Keywords: Automation and Process Control, Industry Applications
Abstract: This paper explores the use of Reinforcement Learning (RL) to autonomously tune controllers that depend on a single adjustable parameter. The proposed approach utilizes system performance and robustness indicators to optimize controller settings, ensuring high efficiency and adaptability. The method was initially validated using a Tacho Generator Motor (TGM) setup to refine various control strategies, such as the Pseudo Proportional Integral Derivative (PPID) Controller and the Generalized Minimum Variance Control (GMVC) approach. Subsequently, the approach was extended to a realistic industrial setup involving a level control system using a frequency inverter-powered pump, with a PLC (Schneider M221) and an ultrasonic sensor. This demonstrated the method's applicability in controlling active industrial systems and optimizing online control via MATLAB. Both tests underscored the potential of RL-tuned controllers to achieve a favorable trade-off between performance and robustness, minimizing tracking errors while ensuring stability in diverse applications.

Keywords: Automation and Process Control, Industry 4.0
Abstract: A orientação a serviços pode fornecer as funcionalidades necessárias para arquiteturas de controle e automação em aplicações na Indústria 4.0. Controle de processos usando serviços e Computação de Borda fornecem maior flexibilidade de desenvolvimento e implantação, melhor eficiência da comunicação em rede e escalabilidade. Este artigo apresenta o desenvolvimento de um controlador adaptativo de processos industriais a partir da orquestração de microsserviços. A estratégia de controle usada foi o adaptativo autoajustável, o qual realiza a sintonia dos ganhos do microsserviço de controle PID4.0 com base na identificação dos parâmetros do processo. A identificação do sistema é realizada via algoritmo de mínimos quadrados recursivos e um modelo ARX, utilizando dados de entrada e saída do processo fornecidos pelo microsserviço DAQ. Experimentos foram realizados numa malha de controle de uma planta piloto, alterando-se parâmetros de projeto do controlador adaptativo e analisando-se o desempenho de controle da abordagem proposta em diferentes condições. Resultados demonstram que o controle adaptativo usando microsserviços fornece rastreamento de setpoint e supera distúrbios no processo, além de facilitar sua implantação em sistemas industriais de controle e automação

Keywords: Robotics and Mechatronics, Automation and Process Control, Electrical Machines and Drives
Abstract: Control methods of dissolved oxygen (DO) in water may involve nonlinearities, non-Gaussian noise, external disturbances, and unknown dead times, which makes its study particularly compelling. Nevertheless, requirements such as extensive infrastructure and expensive instrumentation represent significant limitations for experimentation within a typical control laboratory. Therefore, this paper presents an easy-to-replicate aeration process for DO control experimentation. The proposed approach employs a cascade strategy that combines Fuzzy Logic Control and Sliding Mode Control to regulate DO levels and paddle wheel velocity, respectively. The system features a reconfigurable paddle wheel aerator engineered for experimental evaluation. Moreover, the setup incorporates a DC motor with an encoder, a DO sensor, and a standard computer. For validation purposes, a small water tray was used, in which sodium sulfite was added to reduce the DO level, followed by activation of the paddle wheel to increase it. Finally, simulated and experimental results are presented to validate the proposed system.

Keywords: Automation and Process Control, Artificial Intelligence
Abstract: This study presents the Matlab simulation of PID, type-1 and interval type-2 fuzzy logic controllers (FLC) applied to an Electromagnetic Force Compensation (EMFC) load cell, implemented in the Matlab/Simulink environment. A mass-spring-damper mechanical model is employed to represent system dynamics. The performance of the fuzzy controllers is compared to that of a conventional PID controller. Stability characteristics and frequency response are also investigated. FLCs yield superior performance in EMFC load cell applications when compared to classical PID control.

Keywords: Automation and Process Control, Industry Applications, Robotics and Mechatronics
Abstract: This paper proposes a static output feedback (SOF) design for D-stabilization, disturbance attenuation via H∞ norm, and controller norm constraint for nonlinear systems represented by Takagi-Sugeno (T-S) fuzzy models. The objective is to ensure that the closed-loop system eigenvalues remain within a predefined circular region, guaranteeing a minimum decay rate. Simultaneously, the H∞ approach mitigates the effects of uncertainties and external disturbances. Additionally, a constraint on the controller gain norm is imposed to prevent excessively high gains that could hinder practical implementations. The proposed methodology relies on a matrix decomposition, enabling the static output feedback solution to be derived from a corresponding state feedback solution. The proposed approach is applied to an active suspension system, highlighting its feasibility and performance.

Keywords: Diagnosis, Prognosis and System Identification, Industry Applications, Automation and Process Control
Abstract: The increasing complexity of electronic modules, driven by technological advancements and stringent operational requirements, poses significant challenges for ensuring reliability, especially in harsh environments, such as subsea production systems and other mission-critical applications. This study reports the application of Environmental Stress Screening (ESS) as a quality control technique to identify and filter out electronic modules susceptible to premature failure during manufacturing. By subjecting a statistically relevant sample of newly assembled modules to controlled temperature cycling between -25 to 80 degrees Celsius, latent defects were effectively precipitated without causing life-limiting damage. The ESS process identified early life failures in 12 percent of the tested samples, with failures concentrated in a specific subsystem, highlighting opportunities for targeted design improvements. The results demonstrate that implementing an ESS enhances the overall reliability and longevity of electronic devices, reduces warranty claims and customer dissatisfaction, and provides valuable feedback for continuous product improvement.

Keywords: Diagnosis, Prognosis and System Identification, Electrical Machines and Drives, Industry Data Science Applications
Abstract: Three-phase induction motors (TIMs) play a crucial role in the industrial sector, powering 90% of machines and applications. They are favored for their simplicity, versatility, reliability, and durability. However, at the same time, these motors present failures of various types, such as broken rotor bars, which represent on average 10% of the defects. These failures can lead to significant financial losses, forcing companies to perform preventive maintenance, often unnecessarily disassembling motors to prevent unexpected breakdowns. In this sense, researchers explore non-invasive diagnostic methods such as acoustic sensors and accelerometers, facilitating early detection of faults without disassembly. However, the detection of low-magnitude damage can be masked by noise due to the different operating conditions of the machine. Some of recent research also uses expensive sensors and overly difficult methodologies, which motivated the following proposal: a low-cost accelerometer and simplified signal processing to detect minor damages. Therefore, this study focuses on the application of accelerometers to detect a minor rotor failure in TIMs by analyzing vibration patterns. The paper proposes a method that involves accelerometer sensors, root mean square (RMS), and Skewness signal processing. Experiments were carried out to test motor performance under different voltage and loading conditions that could mask the fault signal and hinder fault classification. The results demonstrate the effectiveness of using accelerometers for fault detection, providing a viable, noise-resistant, and cost-effective option for industrial applications.

Keywords: Diagnosis, Prognosis and System Identification, Electrical Machines and Drives, Artificial Intelligence
Abstract: Motores de indução desempenham um papel crucial em diversos processos industriais. O diagnóstico de falhas nesses motores é essencial para minimizar custos de manutenção, reduzir o tempo de inatividade e evitar acidentes. Este trabalho apresenta o desenvolvimento de uma base de dados para treinamento de Redes Neurais Artificiais com o intuito de diagnosticar falhas em motores de indução. A base de dados foi criada por meio de simulações de motores cujos parâmetros foram obtidos na literatura, emulando dados coletados de diversos motores, abrangendo condições de operação normal e perante falhas. As correntes trifásicas foram utilizadas para criar um vetor de características mediante a extração de informação da transformada de Fourier e de indicadores de qualidade. A rede neural foi desenvolvida utilizando o Matlab online, apresentando altos valores de performance e baixos valores de erro. Para trabalhos futuros, destaca-se a utilização de dados de motores reais para a verificação da eficiência em âmbito industrial.

Keywords: Diagnosis, Prognosis and System Identification, Artificial Intelligence, Machine Learning
Abstract: The identification and online monitoring of wear in single-point dressers is important to achieve the desired surface condition on the grinding wheel and to ensure a satisfactory outcome in the grinding process. However, tool wear is a complex phenomenon that occurs in various forms during the cutting operation and lacks an analytical model capable of representing its wear state. This study aims to develop a method for predicting the effective width of a single-point dresser based on acoustic emission and computer vision data using artificial neural networks. This approach shows satisfactory results in wear prediction, showing error percentages as low as 0.46%,0.19%, and 0.07%.

Keywords: Diagnosis, Prognosis and System Identification
Abstract: This paper explores the application of nonlinear system identification techniques to the actuator system of a Homogeneous Charge Compression Ignition Engine. The study focuses on three essential actuators: the fuel injector, the intake valve, and the exhaust valve. The objective is to establish a dynamic model that accurately represents the behavior of each actuator, with the ultimate goal of integrating them into a modular digital twin that characterizes the dynamics of the key HCCI engine subsystems. Additionally, the study evaluates the performance of different nonlinear model structures to determine their effectiveness in capturing the inherent dynamics of these systems. Ultimately, the systems were described using Hammerstein-Wiener models, resulting in a Normalized Root Mean Square Error (NRMSE) of 86.04% for fuel injection, 88.00% for the intake valve, and 83.02% for the exhaust valve.

Keywords: Power and Eneergy Systems, Diagnosis, Prognosis and System Identification
Abstract: This article analyzes the performance of different filtering techniques applied to the phenomenon known as partial discharges (PD), which occur in electrical insulation and, when undetected, can evolve into serious failures, such as short circuits and power outages. Therefore, PD analysis becomes an essential tool in the operation, maintenance, and asset management of power grids, since identifying these PD pulses is not trivial due to the presence of noise from both internal and external sources within the power systems, such as adjacent equipment and pollution levels. The study compared filtering and analysis methods on non-invasive measurements made with a high-frequency current transformer, evaluating traditional digital filters, adaptive techniques, and the wavelet transform (WT). A set of seven metrics was used to verify their effectiveness in minimizing noise and preserving signal characteristics, such as signal-to-noise ratio (SNR), cross-correlation (CC), and root mean square error (RMSE), among others. The results demonstrated that advanced approaches, such as the use of adaptive filters and wavelet combinations, outperform classical methods, reducing noise in measurement signals and favoring clearer detection of PD pulses in real data. This contribution improves monitoring and increases reliability in high-voltage electrical systems.

Keywords: Power and Eneergy Systems, Industry Data Science Applications, Industry Applications
Abstract: Faults in distribution feeders cause significant losses to society and affect the reliability and energy quality of power distribution systems. Developing methods to estimate regions whose feeders are most vulnerable to faults becomes essential to mitigate these impacts and improve the power grid operation. In this context, this study performs an exploratory spatial data analysis to assess the association between the incidence of atmospheric electrical discharges (AEDs) and the occurrence of faults. The k-means algorithm is applied to produce thematic maps to identify regions with similar characteristics with respect to the incidence of faults. Pearson and Spearman correlation coefficients indicate a positive correlation between the frequency of AEDs and faults in power distribution transformers. In addition, the Poisson probability distribution function is applied to estimate the probability of faults by regions based on historical data. Thematic maps highlight the areas most vulnerable to faults, which power utilities should prioritize with corrective and preventive maintenance of the electrical grid infrastructure.

Keywords: Power and Eneergy Systems, Renewable Energy, Smart Grids
Abstract: This paper presents an analysis of differential protection applied to electric power distribution networks, aiming to enhance reliability and improve fault detection. The investigation evaluates the performance of the protection scheme under scenarios with and without the presence of distributed generation, using a simulated distribution network topology modeled in MATLAB/Simulink. The proposed methodology employs measurements at only two points in the network, eliminating the need for monitoring all branches within the protection zone. The results demonstrate that the proposed differential protection philosophy is robust against varying levels of DG penetration, inherently coordinated, and can be implemented without control algorithms or adaptive adjustments, as well as without the need for network state estimation simulations.

Keywords: Power and Eneergy Systems, Optimization Heuristics and Methods, Renewable Energy
Abstract: Evaluating the operation of electric power distribution systems often requires analyzing a large number of hourly load and distributed generation profiles, which can undermine the computational efficiency of more complex studies. In this context, this paper proposes a methodology for synthesizing representative operational scenarios, based on the K-means clustering technique, with the aim of reducing data volume without significant loss of information. The synthesized scenarios preserve the main operational characteristics of the system, enabling analyses to be performed at lower computational cost. The representativeness of the generated scenarios is assessed through quantitative metrics that evaluate how well the clusters reproduce the original behavior of the operational variables. The results show that the proposed approach is effective in generating reduced sets of scenarios with high representational quality, contributing to planning, operation, and decision-making studies in distribution systems.

Keywords: Power and Eneergy Systems
Abstract: Este artigo analisa a aplicação do elemento de sobrecorrente de sequência negativa na proteção de sistemas elétricos comparando seu desempenho com os tradicionais elementos de fase e de sequência zero. Embora preterido na prática, o elemento de sequência negativa apresenta vantagens quando analisado em relés microprocessados modernos. O estudo utilizou um sistema-teste com 16 barras, modelado no software PowerFactory 2024 com dados reais do sistema Furnas, considerando tensões de 138 kV, 230 kV e 345 kV. Foram simuladas mais de 300 faltas do tipo Fase-Fase, Fase-Fase-Terra e Fase-Terra, avaliando o tempo de atuação e o alcance dos diferentes elementos de proteção. Os resultados indicam que o elemento de sequência negativa teve maior alcance em 100% dos casos de faltas Fase-Fase, além de atuar mais rapidamente em até 90% destas ocorrências. Também se mostrou eficaz em cenários onde os elementos tradicionais não atuariam, devido a correntes de falta inferiores às de carga. A inclusão sistemática do elemento de sequência negativa pode melhorar significativamente a seletividade, reduzir o tempo de interrupção e minimizar danos aos equipamentos sob falta.

Keywords: Electrical Vehicle and Energy Storage, Industry Applications, Renewable Energy
Abstract: The safe and efficient implementation of hydrogen as an ener-gy vector is related to international standards that standard-ize its production, storage, transportation and application. These regulations are essential to guarantee the safety, effi-ciency and interoperability of technologies associated with hydrogen, enabling their insertion into the global energy sce-nario. This article analyzes the regulation of the hydrogen value chain as a sustainable energy carrier. The standards were classified by Organization, Area, Application, and Ob-jective, as well as thirteen other subcategories. The results indicated a predominance of standards from the China Na-tional Standards (GB) organization, with a focus on Fuel Cells (FC), Equipment (EH), and Constructive Aspects (CA). The International Electrotechnical Commission (IEC) focuses on Fuel Cells (FC), Performance (PE), Stationary Plants (SP), and Equipment (EH), while the Canadian Standards Associa-tion (CSA) stands out in Vehicle Refueling Stations (VR), Equipment (EH), and Constructive Aspects (CA). Additional-ly, the International Organization for Standardization (ISO) shows a uniform distribution, with a moderate emphasis on VR, Materials, Accessories or Others (OR), and Equipment (EH).

Keywords: Electrical Vehicle and Energy Storage, Renewable Energy, Smart Grids
Abstract: This paper presents a systematic review and classification of customized dispatch strategies for hybrid renewable energy systems (HRES) using the MATLAB Link Module in HOMER Pro. The review identified 14 studies integrating user-defined control logic, predominantly relying on rule-based approaches, with limited adoption of predictive or optimization techniques. To illustrate the practical application of this methodology, a conceptual case study—Peak-Aware Grid-Support Dispatch—was developed for a 500 kW PV and 200 kW/400 kWh BESS system operating under dynamic tariffs. The MATLAB-based dispatch strategy effectively reduced peak grid demand, enhanced PV self-consumption, maintained battery operation within safe state-of-charge limits, and minimized high-tariff grid imports. The findings confirm the feasibility and potential of HOMER-MATLAB integration for implementing advanced, tariff-aware control strategies in urban microgrids, addressing current gaps in research and practice.

Keywords: Electrical Vehicle and Energy Storage, Power and Eneergy Systems, Renewable Energy
Abstract: This paper presents an innovative approach that combines fuzzy logic with a multi-objective mixed-integer linear programming (MILP) model to address the challenges of optimal planning for electric vehicle charging stations (EVCS). The proposed method integrates fuzzy logic to evaluate potential locations based on various geospatial, economic, and grid-dependence factors. This allows for a more flexible and accurate ranking of candidate sites, ensuring robust decisions for the initial deployment of EVCSs. The MILP model then optimizes multiple objectives, including minimizing installation and maintenance costs, reducing the distance between EVCSs, and aligning with electric vehicle user preferences. Additionally, the model incorporates constraints on user waiting times, ensuring an efficient and functional charging experience. By simultaneously considering these aspects, the approach offers a comprehensive solution that enhances decision-making processes for urban electric mobility planning. The method is scalable and adaptable, making it suitable for dynamic and evolving urban environments. This approach not only improves operational efficiency but also contributes to reducing costs and supporting the widespread adoption of electric vehicles. The proposed solution is expected to provide valuable insights for policymakers, urban planners, and other stakeholders involved in developing sustainable and efficient EVCS network.

Keywords: Electrical Vehicle and Energy Storage, Power and Eneergy Systems, Smart Grids
Abstract: The insertion of battery energy storage systems (BESS) has grown exponentially and may become one of the main energy storage resources in a few years. However, this recent growth brings some challenges since there is an incomplete knowledge of how these storage devices work, including their operational requirements, especially how they degrade throughout their operation. Battery designers and suppliers usually make acceptable estimates of these aspects; however, the actual operating conditions of a BESS may differ significantly from the optimal conditions under which the original estimates were made. Irregular charge/discharge regimes and an unbalanced state of charge (SOC) are common issues in real applications. As a result, the useful life of the batteries can be drastically reduced, leading to significant errors, particularly in the economic evaluation of the power system's operation. Predicting the degradation of these batteries is a complex task. Therefore, evaluating the State of Health (SOH) of the batteries becomes an important metric from the choice of the type of battery to the operation. In this paper, we make a critical review of different methods for estimating the state of charge of lithium-ion batteries; finally, new perspectives and challenges are discussed as well.

Keywords: Electrical Vehicle and Energy Storage, Smart Grids, Power and Eneergy Systems
Abstract: This paper presents an enhanced semi-empirical model for lithium-ion battery degradation that accounts for both calendar and cycle aging mechanisms under various operating conditions. Building upon a previously established double-exponential formulation, the proposed model introduces three key improvements. First, it employs an asymmetric Gaussian function to represent the State of Charge (SoC) stress factor, capturing the non-monotonic degradation pattern observed in calendar tests. Second, a novel two-exponential formulation is developed for the Depth of Discharge (DoD) stress factor, enabling better representation across shallow and deep cycling regimes. Third, the model explicitly includes the C-rate as an independent degradation stressor, using a linear function to isolate its impact—an aspect neglected in the prior work. Model parameters were calibrated using digitized experimental data from controlled tests, and the results show strong agreement with empirical observations across multiple stress profiles. The resulting formulation is compact, interpretable, and suitable for integration into Battery Energy Storage System (BESS) optimization frameworks, enabling degradation-aware decision-making in industrial applications.

Keywords: Renewable Energy, Resource Efficienty & Circular Economy Tracking, Bioprocessess Applied to Industry
Abstract: Aspectos técnicos são apresentados acerca da aplicação de um sistema de cogeração em um tipo específico de Estação de Tratamento de Esgoto (ETE) que integra reatores anaeróbios e pós-tratamento por processo aeróbio, utilizando o biogás como combustível em motor de combustão interna (MCI), gerando eletricidade e calor útil. A energia elétrica irá suprir um eletrolisador para produção de hidrogênio (H2), constituindo um ativo energético para o operador. A energia térmica residual da cogeração é aproveitada na secagem do lodo - subproduto úmido da ETE que exige desidratação e transporte para a destinação final adequada - enquanto o oxigênio (O2) proveniente da eletrólise é direcionado para a fase secundária de aeração mecanizada, mitigando a demanda de energia elétrica nesta etapa. A proposta busca sinergia entre eficiência energética e a promissora aplicação do método lodos ativados com aeração prolongada como pós-tratamento de efluentes de reatores UASB, este oferendo biogás como subproduto de seu processo anaeróbio e aquele com alta demanda elétrica em seu processo aeróbio.

Keywords: Renewable Energy, Power and Eneergy Systems, Optimization Heuristics and Methods
Abstract: Climate change has intensified extreme weather events, such as heatwaves and cold fronts, disrupting economic activities and accelerating the transition to low-carbon energy systems. Renewable hydrogen is a key enabler of this transition, offering a clean fuel alternative for the industrial and transportation sectors. However, its large-scale deployment faces technical and economic challenges, especially in production. This paper presents an optimization model to minimize the operational costs of a grid-connected renewable hydrogen production and storage hub, utilizing wind power as the primary energy source and the electrical grid as a secondary energy source. The results provide valuable insights for operational planning, supporting the cost-effective deployment of renewable hydrogen aligned with sustainable energy transition goals.

Keywords: Renewable Energy, Power and Eneergy Systems
Abstract: An aquifer thermal energy storage (ATES) can mitigate CO2 emissions of heating, ventilation, and air conditioning (HVAC) systems for buildings. In application, an ATES keeps large quantities of thermal energy in groundwater-saturated aquifers. Normally, an ATES system comprises two (one for heat and one for cold) storages and supports the heating and cooling efforts of simultaneously present HVAC system components. This way, the operation and emissions of installed and, usually, fossil fuel-based components are reduced.

The control of ATES systems is challenging, and various control schemes, including model predictive control (MPC), have been proposed. In this context, we present a lightweight input-output-data-based autoregressive with exogenous input (ARX) model of the hybrid ATES system dynamics. The ARX model allows the design of an output-based MPC scheme, resulting in an easy-to-solve quadratic program and avoiding challenging state estimations of ground temperatures. A numerical study discusses the accuracy of the ARX predictor and controller performance.

Keywords: Renewable Energy, Power and Eneergy Systems, Power Quality
Abstract: As part of an initiative by the Brazilian Air Force (FAB) aimed at promoting sustainability and reducing operating costs, several photovoltaic (PV) plants have been installed in its Military Organizations. Given the criticality of the systems they supply, it is essential to ensure the continuity of operation and energy reliability. With this focus, this study simulates a 33~kWp PV plant using MATLAB/Simulink, with modules representative of those used by the FAB. Four recurring anomalies on the direct current (DC) side are analyzed: dirt on the modules, cable breakage, short circuit between the negative pole and ground, and imbalance between modules. The impacts of these anomalies on electrical parameters such as voltage, current, and active and reactive power, in steady state, are evaluated. The results obtained contribute to the identification and classification of electrical signatures associated with different types of anomalies, using machine learning algorithms. This provides subsidies for effective monitoring, as well as for preventive and corrective maintenance strategies in critical PV systems.

Keywords: Renewable Energy
Abstract: This study investigates the deposition of Nb₂O₅ thin films on glass substrates using DC-pulsed Magnetron Sputtering, focusing on the effects of pulse length and oxygen pressure. The sputtering system employed a stainless-steel chamber with a Nb-metallic target under controlled conditions. Two experimental series were performed: the first varied pulse lengths (170, 250, 330, 410, and 600 µs) at fixed O₂ (1 mTorr) and Ar (2 mTorr) pressures; the second fixed pulse length (250 µs) and Ar pressure (2 mTorr), varying O₂ pressure (0.25, 0.50, and 1.00 mTorr). Optical Emission Spectroscopy (OES) analyzed the plasma, revealing distinct behaviors compared to RF excitation. Film thickness was measured with a profilometer, optical properties via UV-Vis-NIR spectroscopy, wettability through contact angle, and structure by X-ray diffraction. Results show that increasing pulse length enhances deposition rates, while O₂ pressure has a more complex influence. OES identified Ar I and O₂⁺ emission lines, indicating ionization and dissociation processes. Optical characterization revealed absorption edges related to band gap transitions. Band gaps, refractive indices, and extinction coefficients were determined using Tauc plots and the Cisneros method. While pulse length had minimal effect on optical properties, lower O₂ pressure increased refractive index, suggesting changes in film density or phase. This study provides insights into tailoring Nb₂O₅ films by adjusting DC-pulsed sputtering parameters to optimize structural and optical properties for various applications.

Keywords: Automation and Process Control, Industry Applications
Abstract: This paper presents a comprehensive approach to the modernization of a Variable Refrigerant Flow (VRF) refrigeration bench for technical education. The project goes beyond simple component replacement, focusing on rigorous system modeling, the design and implementation of a discrete-time PID RST controller, and the integration of a SCADA system for supervision and data acquisition. A detailed mathematical model of the system’s thermal dynamics was developed, based on the literature studied. This model informed the design of a PID RST controller, which was implemented on a Schneider Modicon M221 PLC. The controller achieves precise temperature regulation, demonstrating fast settling times, minimal overshoot, and zero steady-state error in response to both setpoint changes and thermal load disturbances. The integration of an Elipse E3 SCADA system provides real-time monitoring, control, and data logging capabilities, enhancing the bench’s usability and educational value. Experimental results, including performance and robustness indices derived from Bode diagram analysis, confirm the effectiveness of the implemented control strategy. The modernized SVRF bench serves as a valuable platform for hands-on learning in advanced control techniques, automation, and refrigeration systems.

Keywords: Automation and Process Control, Industry Applications, Industry 4.0
Abstract: This paper presents the improvement of the cooking-pressing system in a fishmeal plant to produce high-quality meal with low moisture and fat content. Due to limitations in accessing the plant, a computational model based on a Nonlinear AutorRegressive model with eXogenous inputs (NARX) neural network was developed using experimental process data. A decoupled neuro-fuzzy control strategy was proposed, employing Adaptive Neuro-Fuzzy Inference System (ANFIS)-tuned membership functions to optimize the fuzzy controllers. Different fuzzy controller structures—Proportional-Derivative (PD)-type, Proportional-Integral (PI)-type, and hybrid PI-PD-type—were evaluated under typical disturbances and noise to assess robustness in a simulation environment. Implementation considerations for deployment on a real plant were also discussed. Results demonstrate that the proposed neuro-fuzzy controllers meet design specifications and improve disturbance rejection, enhancing the overall efficiency of the cooking-pressing system.

Keywords: Automation and Process Control, Robotics and Mechatronics, Industry 4.0
Abstract: Drug dispensing errors are a significant concern in hospital environments, impacting patient safety and operational efficiency. Although automated drug dispensers (ADDs) have proven to reduce human errors and improve inventory management, their high acquisition and maintenance costs often prevent adoption within Brazil’s public health system (Sistema Único de Saúde - SUS). This study presents the development and evaluation of a Cartesian robotic arm designed for drug selection in hospital pharmacies, emphasizing the application of control strategies to ensure precise and reliable operation. Three control approaches—classical Proportional-Integral-Derivative (PID), Incremental PID, and Sliding Mode Control (SMC)—were implemented and tested in a mechatronic system prototype. The controllers were evaluated based on final positioning error and stabilization time across 350 experimental movements. The results indicate that the SMC achieved superior performance, offering critically damped responses, high precision, and low stabilization times. These findings suggest that the proposed solution has strong potential for improving drug dispensing processes in Brazilian public hospitals, enhancing safety, efficiency, and resource management.

Keywords: Automation and Process Control, Internet of Things, Bioprocessess Applied to Industry
Abstract: The automation of irrigation systems is essential for optimizing water resource usage and increasing agricultural efficiency. This paper presents an approach based on Production Flow Schema (PFS) and Object-Oriented Petri Nets (OOPNs) to model and control irrigation devices intelligently. PFS is used to structure the activity flow and define the operational logic of the devices, while OOPNs enable the modeling of the system’s dynamic behavior, ensuring synchronization and efficient control, as well as providing modularization capabilities for intermediate networks. The proposed methodology was implemented in a computational environment, demonstrating the control of information flow within an irrigation system. The results indicate that the combination of PFS and OOPNs provides a robust and scalable system model for irrigation automation, contributing to sustainability and operational efficiency in the field.

Keywords: Automation and Process Control, Machine Learning, Industry 4.0
Abstract: O presente estudo apresenta uma proposta de integração entre a linguagem R e o sistema supervisory control and data acquisition (SCADA), visando o monitoramento de recursos hídricos. Neste sentido, a proposta consiste em executar scripts em linguagem R por meio do sistema SCADA, facilitando o processamento estatístico e a predição do índice de qualidade da água (IQA), uma métrica utilizada na gestão ambiental. O estudo foi desenvolvido no software AVEVA Edge e testado em ambiente virtual, por meio da simulação do monitoramento da qualidade da água em um município da Região Metropolitana do Vale do Paraíba e Litoral Norte (RMVP-LN), utilizando-se de scripts em Visual Basic para acionar modelos de regressão implementados em R, para a exibição no sistema SCADA. Os resultados obtidos demonstraram o potencial relevante da proposta de integração da linguagem R ao sistema SCADA, a qual poderá ser utilizada como uma alternativa acessível no contexto industrial e ambiental acerca do monitoramento de recursos hídricos.

Keywords: Diagnosis, Prognosis and System Identification, Electrical Machines and Drives, Machine Learning
Abstract: The main objective of this work is to develop an automatic fault detection system in electrical machines, based on a microcontroller embedded in an FPGA - Field Programmable Gate Array device. The microcontroller monitors in real time the vibration signal coming from an acceleration sensor, installed in the electrical machine, aiming to detect new vibration patterns that differ from the typical patterns recorded for the proper functioning of a healthy machine. The detection method depends on the determination of characteristics of the vibration signal, selected in the frequency domain, and the application of an Artificial Neural Network - ANN for fault detection. It was demonstrated that, by monitoring the frequency spectrum of the vibration signal, specific characteristics of the signal can be extracted and, through the application of a neural network, it is possible to detect faults in the operation of the electrical machine. The microcontroller embedded in the FPGA can perform all necessary calculations. The microcontroller allows high-speed calculations with low power consumption, high reliability and low cost compared to solutions commonly found in the market.

Keywords: Diagnosis, Prognosis and System Identification, Industry Applications, Smart Grids
Abstract: With the rapid growth in the development of sensors, actuators, and microcontrollers, an entire field of research has emerged around low-cost instrumentation for various types of sensors used in the study of physical quantities. In the field of Structural Health Monitoring (SHM), the early detection and quantification of damage in civil structures is crucial. This allows for reduced maintenance and repair costs, while also enhancing the overall safety of the structures through continuous data acquisition and analysis. In this context, the present work focuses on the development of low-cost electronics for the excitation and signal conditioning of a displacement sensor based on inductive principles: the Linear Variable Differential Transformer (LVDT). The proposed design includes wireless connectivity for the sensor to be integrated into a monitoring network. In this approach, an Arduino MKRWAN 1310 is used.

Keywords: Diagnosis, Prognosis and System Identification, Industry Data Science Applications, Deep Learning and Machine Learning
Abstract: Predictive maintenance, through regular monitoring of the actual electromechanical condition, efficiency, and other operational indicators of machines and process systems, ensures maximum intervals between repairs and reduces the frequency and cost of unplanned shutdowns caused by failures. Therefore, it is essential to develop efficient methodologies that are feasible for implementation in industrial plants, enabling control and predictability of maintenance downtime. In this context, the Envelope Technique—a reference method for fault analysis, diagnosis, and monitoring in bearings—is applied through frequency-domain spectral analysis of vibration signals, based on the propagation of amplitude near the natural frequency of the component under analysis. This work presents a simple, efficient, and innovative methodology for experimental simulations, monitoring, diagnosis, fault prediction, and estimation of the Remaining Useful Life (RUL) of bearings in rotating machines using vibration signals acquired by a sensor installed on a test bench specifically designed and built to replicate the dynamics of real-world rotating machinery. The implemented experimental simulations generated the progressive increase of vibration peaks—characteristic of fault evolution—through a novel method involving the controlled application of dynamic radial loads. Vibration spectral datasets were collected from an MPU6050 sensor via an Arduino Mega 2560 board, integrated with a model developed in Simulink, which enabled real-time signal acquisition and data file generation for critical (or admissible) fault prediction through processing in MATLAB. The prediction of the critical fault occurrence and RUL estimation was performed using a simple numerical method.

Keywords: Diagnosis, Prognosis and System Identification, Industry Data Science Applications, Industrial Ultrasound Theory and Application
Abstract: The present work proposes the use of the macro fiber composite (MFC) transducer for damage detection through ultrasonic wave propagation. The methodology is based on the transmission and reception of ultrasonic waves accomplished through an innovative sensor monitoring approach called chirp-through transmission (CTT), in which a low-cost piezoelectric diaphragm is used as a transmitter and, as presented in this paper, the mentioned MFC transducer is used as a receiver. The acoustic waves were generated in the chirp form with a 500 ms time window and frequencies from 1 Hz to 250 kHz. The waves received by the MFC transducer were sampled at 2MS/s. The damage tests were induced by mass addition, which included attaching three metallic nuts of increasing sizes to the aluminum beam chosen as the structure under inspection. To accurately assess damage using the received acoustic waves, preliminary tests were conducted on the structure in its healthy condition. This initial evaluation established a baseline for comparison. Subsequently, tests were performed under different conditions of damage, allowing for a comprehensive analysis and clear differentiation between conditions. A study was also conducted on the frequency content of the received waves to select the most representative frequency bands. The results demonstrated a satisfactory characterization of the actual structure condition obtained by the representative metrics root-mean-square deviation (RMSD), correlation coefficient deviation metric (CCDM), and root mean square (RMS) computed at the frequency bands of 100- 125 kHz and 150-175kHz. Based on the results, the MFC transducer used in cooperation with CTT can be a simple and effective sensor monitoring tool for non-destructive applications.

Keywords: Diagnosis, Prognosis and System Identification, Electrical Machines and Drives, Automation and Process Control
Abstract: This work proposes the experimental modeling of a system composed of a CFW-10 inverter and a WEG induction motor (1.5 hp, 1715 rpm), integrated into a test bench developed at the Power System Control Laboratory (LACSPOT), UFPA. Treating the plant as an unknown block, a step input of 10 Hz (1 p.u.) was applied in open loop via the inverter to analyze the transient response of the output (angular velocity). Parametric identification and least squares (LS) methods were employed, both based on transient response analysis, for systems characterized as first and second order. However, all methods yielded suitable models for system representation, according to validation metrics such as the sum of squared errors (SSE) and the coefficient of determination (R²). Moreover, the second-order parametric identification method proved to be the most suitable for representing the system.

Keywords: Power and Eneergy Systems, Artificial Intelligence, Industrial Lighting
Abstract: This paper presents an intelligent adaptive search method for assessing the backflashover rate (BFR) of overhead transmission lines, optimizing the search process by progressively narrowing the search space based on an initial critical current. Unlike the conventional fixed-increment search method, in which the peak lightning current is gradually increased until flashover occurs, the proposed approach significantly reduces the number of simulations required by learning to identify the region where the critical current is likely to be found. The method was applied to evaluate the BFR of a 220 kV overhead transmission line, implemented in PSCAD. The evaluation considered 25 distinct power frequency angles ranging from 0° to 360°, with tower footing impedances varying from 20 Ω to 50 Ω. Both the classical and proposed methods were implemented in Python, and critical current results were obtained through an interface between Python and PSCAD. The results indicate that the proposed method achieves accuracy comparable to conventional approaches while substantially reducing the associated computational burden. These findings highlight the potential of the method as an efficient alternative for simulation-based studies, particularly in applications involving extensive BFR assessments

Keywords: Power and Eneergy Systems, Renewable Energy, Power Quality
Abstract: This paper presents and evaluates two distributed consensus-based methodologies for active power control of distributed generation (DG) units in low-voltage distribution networks. The objective is maintaining voltage levels within operational limits under varying load conditions and heterogeneous DG capacities. The first one is based on the classical Olfati-Saber consensus algorithm, which assumes uniform agent capabilities and seeks average convergence. The second one, called Supportive Consensus, introduces an equity-based mechanism that allows agents with greater available capacities to compensate for those operating under constraints. Simulations are conducted on an IEEE 34 bus network. The results demonstrate that both methods perform well in homogeneous systems. However, the Supportive Consensus algorithm offers superior adaptability and constraint handling in heterogeneous configurations. The proposed approaches have the potential to reduce reliance on auxiliary voltage control equipment, enhance system resilience, and provide a scalable solution for real-time DG coordination in smart distribution grids.

Keywords: Power and Eneergy Systems, Renewable Energy, Diagnosis, Prognosis and System Identification
Abstract: This paper investigates the coordinated control of reactive power injection from inverter-based wind farms as a strategy to improve voltage security in power systems. The objective is to propose an alternative voltage regulation method that avoids both the curtailment of renewable generation and the investment in additional regulation devices. A case study involving the contingency of the 500 kV Santa Luzia II – Campina Grande III transmission line, in the Northeast subsystem of the Brazilian Interconnected Power System (BIPS), is analyzed. Dynamic simulations using Brazilian Electrical Energy Research Center (CEPEL) software showed that wind farms connected to Santa Luzia II were disconnected due to under-voltage protection. Subsequent steady-state and dynamic simulations revealed that coordinated reactive power control prevented the voltage at Santa Luzia II from falling below the under-voltage emergency threshold. The results demonstrate that such coordination is effective in avoiding renewable generation curtailment and highlight its potential as a viable solution to improve voltage stability in power systems, as well as further reinforcing the need to create structures of incentives for intermittent generation to provide ancillary services to the bulk power system.

Keywords: Power and Eneergy Systems, Renewable Energy
Abstract: This paper introduces a benchmark test system for frequency stability studies, as part of the efforts by the IEEE PES Task Force on “Benchmark Networks for Low-Inertia Systems” under the Power System Dynamic Performance Committee. The proposed test system is based on the Northeast (NE) electrical power subsystem (or simply, NE subsystem) of the Brazilian Interconnected Power System (BIPS), considering the high penetration of converter-interfaced generation (CIG) units from renewable energy sources (RESs), particularly wind energy. Given the large size of the NE subsystem, a simplified model was created using network reduction techniques and simplifications applied to the complete BIPS and NE subsystem. This simplified model consists of 224 buses, 422 AC transmission lines and transformers, 60 aggregated wind farms (WF), 19 hydroelectric power plants (HPPs), and 18 thermal power plants (TPPs). The total capacity is approximately 22,700 MW, and half of this capacity comes from wind farms. Standard models for the HPPs, TPPs and WFs controllers were adopted, which facilitates the reproducibility of the test system. Data for this model are available in a public repository. To address frequency stability issues arising from the high penetration of CIG units from RES, the test system is considered a control area responsible for balancing its own power imbalances, while maintaining constant power exchanges with other subsystems. In this operating scenario, dynamic RMS simulations show that increasing CIG units from RESs may compromise frequency stability. Scenarios where some conventional generators are inoperative can lead to cascading disconnections of the TPPs, ultimately resulting in blackouts.

Keywords: Power and Eneergy Systems, Electrical Vehicle and Energy Storage, Renewable Energy
Abstract: Battery energy storage systems in containerized environments often face two significant challenges: the high energy demand of air-conditioning systems for thermal management and the accelerated battery degradation due to frequent discharge cycles. This work presents an optimization model that addresses both issues by integrating a three-parameter thermal model with a battery degradation function sensitive to temperature and cycling. A comparative analysis is conducted between a simplified two-parameter model - considering only the thermal inertia of the container interior and the batteries - and a complete three-parameter model that also accounts for the thermal inertia of the container walls. Simulate results using temperature and irradiance data from Santo Andre-SP, Brazil, show that including the container wall inertia leads to more efficient air-conditioning operation and smoother battery cycling, yielding an average instantaneous degradation rate of less than 1%. In contrast, the simplified model results in higher energy consumption for cooling and more aggressive battery operation, with sharper power peaks and increased degradation.

Keywords: Electrical Machines and Drives, Diagnosis, Prognosis and System Identification, Deep Learning and Machine Learning
Abstract: Three-phase induction motors (TIMs) are the primary source of mechanical energy in industrial applications. With advances in power electronics, TIMs can be easily controlled for various applications, including pumps, treadmills, exhausters, compressors, and fans. However, their widespread use requires efficient and rapid maintenance strategies. Performing a timely diagnosis of faults is crucial, as unexpected downtime can cause significant economic losses for industries. One of the most common faults in TIMs is short circuits in the stator winding, which can cause overload, excessive heat generation, increased vibrations, and insulation degradation. Modern signal processing techniques, combined with advances in neural networks, offer new approaches to fault detection and classification with high speed and accuracy. The integration of electro-acoustic (EA) sensors further enhances fault detection capabilities, making them a powerful tool for improving motor maintenance. Therefore, this paper presents a method to identify and classify the severity of short circuit faults in TIMs, applying discrete wavelet transform (DWT) as a preprocessing of signal, and then the extracted features is compacted as a tensor, which is used to train and validate a deep neural network (DNN).

Keywords: Electrical Machines and Drives, Electrical Vehicle and Energy Storage, Industry Applications
Abstract: This work presents a control methodology for reducing torque ripple in synchronous reluctance motors (SynRM). The strategy proposes the use of a torque function reconstructed from finite element analysis (FEA) and harmonic interpolation for different current levels. The method was analyzed through co-simulations between Matlab/Simulink and Ansys Electronics, considering field-oriented vector control. The results demonstrated a significant reduction in torque ripple, from 5.28% to 1.53%, while maintaining system efficiency. The analysis also highlighted the influence of magnetic saturation on the reference currents. The proposed approach shows strong potential for industrial and automotive applications requiring both robustness and high efficiency.

Keywords: Electrical Machines and Drives, Diagnosis, Prognosis and System Identification, Industry 4.0
Abstract: This work presents a hybrid modeling methodology based on the dynamic response, filtering and identification techniques, considering time and frequency domains, to determine the representative model of a fixed-coupled ball mill. That is, models were proposed for the electric drive, gearmotor and load, without the need for physical decoupling. The electrical parameters are determined using state variable filtering, linear regression and recursive least squares techniques, and the mechanical parameters are identified by considering only the acceleration time of the system. A final adjustment of the parameter set is performed using the nonlinear least squares technique. Numerical simulations of the given model, under different mill operating conditions, show a good approximation with experimental results. Therefore, the proposed hybrid methodology, based on both dynamic modeling and signal analysis, has the potential to assist in the design of monitoring and control processes for the fixed-coupled ball mill.

Keywords: Electrical Machines and Drives, Power Electronics, Personalized Products
Abstract: Este trabalho apresenta o desenvolvimento do projeto eletromecânico, simulação eletromagnética e construção de uma Máquina a Relutância Chaveada (MRC) na topologia 12x10. A escolha dessa configuração foi motivada por sua robustez e confiabilidade, uma vez que a máquina mantém seu funcionamento mesmo em caso de perda de fase, além de contribuir para a diminuição da amplitude de ruído e vibração. O projeto foi fundamentado na literatura clássica de máquinas elétricas, sendo adaptado para se adequar a uma carcaça comercial, com foco na viabilidade técnica e redução de custos de fabricação. O método dos elementos finitos foi utilizado para a simulação e validação do desempenho eletromagnético da máquina, permitindo identificar regiões de saturação, caminho e densidade do fluxo magnético. Essa etapa foi essencial para minimizar erros de projeto antes da sua construção física, economizando assim tempo, recurso material e financeiro. Por fim, houve a construção da máquina, contemplando o estudo sobre o corte das lâminas e a preparação dessas peças. Além da realização de serviços de usinagem, envernizamento, fixação do conjunto de lâminas na carcaça e bobinagem. Os passos apresentados neste trabalho podem ser replicados para o desenvolvimento de qualquer MRC, com os devidos ajustes conforme a topologia selecionada.

Keywords: Electrical Machines and Drives
Abstract: The aim of this work is to comparatively analyze the iron losses in a three-phase induction motor (1 HP, 60 Hz, 127 V) when using the ferromagnetic materials 10JNEX900 (6.5% Si) and M235-35A (3.5% Si) . To this end, an electro magnetic simulation based on the finite element method was developed to estimate the levels of magnetic induction in the machine under investigation. Once these induction profiles were available, the losses for each motor configuration were estimated using the Bertotti method, whose hysteresis and eddy current coefficients were estimated by fitting the loss density x magnetic induction curves, in such a way that the information was extracted from the Ansys Maxwell database, for the M235-35A the data is for a frequency of 50 Hz and for the 10JNEX900 the information is for 100 Hz. In addition, the accuracy of the approximations via the Bertotti method was checked by constructing the relative error curves, after which the Bertotti coefficients were adjusted for the 60 Hz frequency. Therefore, the iron losses using M235-35A were approximately 80% higher compared to the arrangement with 10JNEX900.

Keywords: Industry Applications, Industry 4.0, Artificial Intelligence
Abstract: This article describes a study on the application of fuzzy logic for industrial alarm generation, focusing on a storage system used in a food manufacturing facility. The system monitors data from flour silos equipped with weight indicators as inputs and operational status as output. Using simulated data, a fuzzy logic model is developed to evaluate operational conditions based on the levels of the silos and trigger alarms based on the silo data. The model efficiency is validated through a simulated storage system dataset. The system is being developed for use in a factory that is digitalizing its operations to achieve greater process efficiency.

Keywords: Industry Applications
Abstract: Despite the high level of sophistication of machine tools, they are still subject to geometric errors arising from various sources, which directly affect the dimensional accuracy of machined parts. Given the limitations of traditional calibration methods, such as high cost, time consumption, and complexity, this study proposes and applies an alternative method based on error separation for geometric evaluation in a three-axis CNC machining center, demonstrating its practical feasibility. Using a slender 1045 steel artifact and a pair of LVDT sensors with a resolution of 0.0001 mm, measurements were performed in six different setups, with the instruments and artifact fixed to the machine structure. The mathematical model developed has a sparse matrix structure and was solved numerically using the LSQR algorithm, reaching satisfactory convergence in 130 iterations. The straightness profiles obtained through the proposed method were compared to measurements taken with a CMM, showing agreement. The analysis of angular and linear errors allowed the identification of a critical region at 320 mm along the y-axis, where the largest deviations occur, indicating points for applying numerical compensation or mechanical adjustments. The method stood out for enabling the mathematical removal of errors and reconstruction of the artifact’s profile, as well as for its ability to locate and quantify geometric errors in regions crucial for defining the cutting height of the tool. The modeling and algorithm used allow for precise results with low computational cost. Thus, the study presents significant potential for calibration and predictive maintenance of machine tools, providing technical information for decisions on error compensation and quality improvement in the machining process.

Keywords: Industry Applications, Renewable Energy, Power and Eneergy Systems
Abstract: The photovoltaic (PV) module’s position in relation to the Sun significantly influences its performance due to its dependence on the incidence of direct solar irradiation. Most small and medium-sized PV installations are located on rooftops, taking advantage of idle areas. However, they introduce challenges, such as partial shading, greater maintenance requirements, and the impact on operating temperature due to the proximity of the rooftop. This paper investigates the influence of the proximity of modules installed on rooftops 0 and 15 cm apart, and in structures installed on the ground, with and without active cooling. The results show that a distance of 15 cm between the modules and the rooftops provides effective natural ventilation, similar to that of structures on the ground, where for every 1°C reduction, generation increased by an average of 0.45%, demonstrating the effect of temperature on efficiency.

Keywords: Industry Applications, Automation and Process Control, Renewable Energy
Abstract: The intermittence in the performance of photovoltaic modules is directly associated with atmospheric conditions, such as cloudiness, rain and dust. However, under these conditions there is also production of electrical energy, even if smaller. Therefore, this article investigates, over a period of one-year, the effect of solar tracking on the performance of photovoltaic modules under predominantly cloudy and rainy sky conditions, as well as the best static orientations under these conditions. The results show that modules with solar tracking on two axes generate 20.83% less than the static configuration to the horizontal plane under sky conditions. Therefore, the best performance on predominantly cloudy days occurs with the modules facing the horizontal plane, where the distribution of radiation is predominantly isotropic and diffuse throughout the sky.

Keywords: Industry Applications, Internet of Things, Smart Grids
Abstract: This paper presents a performance evaluation of Broadband over Power Line (BPL) networks with cascaded Time Division Repeaters (TDRs), focusing on Quality-of-Service parameters under real-world electrical conditions. To extend coverage of smart grids in Industrial Internet-of-Things (IIoT) applications, BPL with cascaded TDRs can be a promising solution. Thus, in this study, TDRs were used to assess their impact on network scalability and, in particular, the impact of time delay in end-t-end connectivity. We achieved transmission response times between ≈ 2 and ≈ 3 ms in links with 3 TDRs. This experimental analysis paves the way for our future deployments of BPL in smart grids, within the IIoT context.

Keywords: Industry 4.0, Internet of Things, Blockchain Technology
Abstract: The exchange of data between Industrial Internet of Things (IIoT) devices from different organizations is critical to fully exploiting the technological potential of Industry 4.0. However, interoperability and data privacy remain challenges due to the lack of standardization among devices and services. This paper proposes using the MQTT protocol for distributed messaging, integrated with blockchain technology, to enable secure and seamless data exchange between IIoT devices from different organizations. The solution includes a specific topic in MQTT broker for sensitive data transactions, which will be processed and forwarded to the blockchain. Smart contracts will be implemented using Ethereum, leveraging Attribute-Based Encryption (ABE) to manage device permissions and ensure fine-grained access control. This approach aims to enhance IIoT interoperability, secure sensitive data, and facilitate new business opportunities, such as real-time supply chain tracking and automated supply processes, fostering digital transformation and advancing Industry 4.0.

Keywords: Industry 4.0, Diagnosis, Prognosis and System Identification, Cyber Physical Systems, Digital Twins and Knowledge Systems
Abstract: This paper presents a methodology for analyzing industrial PROFINET networks. It focuses on integrating automated topology reconstruction based on collected data and data preprocessing to estimate cumulative distribution functions and probability density functions for failures. Results establish a foundational framework for predictive maintenance through probabilistic analysis. The approach enables temporal error pattern visualization and supports statistical reliability threshold estimation. The work contributes to predictive maintenance in Industry 4.0 environments.

Keywords: Internet of Things, Industry Data Science Applications, Industry 4.0
Abstract: A conservação adequada de vacinas depende de um rigoroso controle de temperatura, pois pequenos desvios podem comprometer a eficácia dos imunobiológicos, resultando em perdas financeiras e riscos à saúde pública. Este artigo propõe uma arquitetura baseada em Internet das Coisas (IoT) com aplicação de técnicas de inteligência artificial, voltada ao desenvolvimento de um sistema inteligente de monitoramento e classificação dos estados operacionais de câmaras de conservação de vacinas. A solução permite o monitoramento de variáveis como temperaturas interna e externa, umidade interna e externa, tensão da rede elétrica, corrente elétrica do compressor e fator de potência do compressor, as quais também compõem a base de dados utilizada para o treinamento supervisionado de uma rede neural densa. O sistema utiliza um microcontrolador ESP32, conectado a sensores de temperatura DHT22 e de energia PZEM-004T. O artigo também apresenta o esquema elétrico de conexão do hardware e o modelo do circuito impresso, além da implementação de uma rede neural para classificação automática dos estados operacionais da câmara, com base nas variáveis monitoradas.

Keywords: Industry 4.0, Internet of Things, Industry Applications
Abstract: The integration of emerging technologies into educational environments is essential for preparing engineers to meet the demands of Industry 4.0. This paper presents the partial results of a project aimed at modernizing a didactic industrial process control plant through the incorporation of Internet of Things (IoT) technologies and the Message Queuing Telemetry Transport (MQTT) communication protocol. The plant, originally operated by programmable logic controllers (PLCs) and composed of classical control loops for flow, temperature, and level, was upgraded to include digital connectivity and real-time monitoring. The proposed architecture includes the deployment of an MQTT broker on a repurposed TV Box running a Debian-based system, ultrasonic sensors for level measurement integrated with ESP32 microcontrollers, and communication with a supervisory system. Integration was tested using the Elipse E3 platform, and system performance was evaluated through tests that measured message transmission response time, confirming stable communication and suitable performance for real-time monitoring applications. This implementation promotes practical learning of Industry 4.0 concepts and demonstrates a sustainable, low-cost approach to modernizing legacy educational systems. The results indicate that the proposed architecture is scalable, robust, and well-suited for future expansions, including the incorporation of additional sensors and supervisory interfaces.

Keywords: Industry 4.0, Industry Applications
Abstract: This paper investigates the repeatability of a structured light 3D scanner applied to scanning a SWIP (Swiss Federal Office of Metrology and Accreditation) certified standard sphere. Based on the substitution method, 12 consecutive scans were performed, whose data were processed in GOM Inspect. The statistical analysis of deviations from the nominal value revealed adequate metrological performance for application in tire molds, evidencing the technology's potential for industrial applications.

Keywords: Autonomous Vehicles and Drones, Diagnosis, Prognosis and System Identification
Abstract: Unmanned aerial vehicles are increasingly offering several benefits to society. The series of functional characteristics for these actions justify this importance. However, such systems present challenges, as they are multi-input and multi-output systems, time-varying and non-linear. Given this, this work proposes the application of an extended Recursive Least Squares algorithm with Multiple Forgetting Factors to identify, in real-time, the external and internal dynamic parameters of the Parrot AR. Drone 2.0 drone, considering its non-linear nature. Using the Python 3 PS-Drone library for data collection and modeling based on discretized differential equations, the study incorporates dynamic couplings between the axes and parametric variations caused by operational conditions. The proposed algorithm allows a differentiated adaptation for each parameter, improving the estimation accuracy and the robustness of the model against noise and sudden changes. Controlled experiments demonstrated that the technique obtains excellent adherence to the real behavior of the Quadcopter, obtaining a Normalized Root Mean Squared Error index greater than 0.8 in all system states. The parametric convergence of the proposed algorithm was compared to other algorithms based on Recursive Least Squares, making it evident that it is more sensitive to the detection of operational failures, reinforcing its potential for applications in adaptive control and real-time diagnostics.

Keywords: Autonomous Vehicles and Drones, Optimization Heuristics and Methods, Robotics and Mechatronics
Abstract: The autonomous navigation of robotic systems is one of the main problems of this industry today. Due to the simplicity of its implementation and the ability to follow trajectories, the Pure Pursuit controller has become a handy and popular tool among researchers. However, the controller’s performance depends directly on the adjustment of its parameters, which must be selected following the track’s characteristics. This study investigates the Particle Swarm Optimization (PSO) algorithm application for automatically tuning multiple parameters within the Pure Pursuit controller, to improve trajectory tracking accuracy on a complex closed-loop circuit. For this purpose, the track is segmented according to the smoothness of the curves present, and the controller parameters are automatically adjusted according to the characteristics of each segment. The results of this work show that segment-wise parameter optimization of a Pure Pursuit controller using the PSO algorithm enables the vehicle to successfully complete complex tracks, whereas both empirical and globally optimized fixed-parameter configurations fail early in the trajectory. The proposed approach demonstrates significant improvements in trajectory tracking accuracy, suggesting that PSO-based parameter optimization can be a valuable tool for enhancing the performance of Pure Pursuit controllers in real-world autonomous navigation applications.

Keywords: Autonomous Vehicles and Drones, Robotics and Mechatronics
Abstract: This paper addresses the research gap in applying geometric-based vehicle motion controllers, such as the Stanley algorithm, to heavy-duty trucks, with a specific focus on lateral path tracking control. We experimentally validate a modified version of the Stanley controller on a 45-ton 8×4 agricultural truck in a test field environment. Two different routes were defined to simulate real-world operation, using high-precision GPS data collected at multiple speeds (5, 10, and 15 km/h) and across five repetitions. The controller achieved a maximum lateral error of 0.7 meters, remaining within the defined tolerances (0.1m for straight segments and 1 m for curves), which are appropriate for heavy-duty agricultural operations. Results confirm the algorithm’s reliability without requiring a full dynamic model. Future improvements with dynamic modeling and noise filtering are also discussed.

Keywords: Personalized Products, Robotics and Mechatronics, Human Machine Symbiosis
Abstract: Driver drowsiness impairs vigilance and increases crash risk, making real-time monitoring vital for Advanced Driver Assistance Systems (ADAS). While Detection Response Task (DRT) reaction times correlate with Karolinska Sleepiness Scale (KSS) scores, existing mapping techniques often suffer from overfitting and poor generalization due to unregularized high-degree polynomials and ad hoc filtering. This study presents a regression framework for estimating KSS from DRT data, integrated into a multimodal fusion and embedded-systems architecture. The pipeline applies quantile-based scaling, RANSAC-based outlier removal, low-degree polynomial expansion, and Ridge regularization, with hyperparameter tuning via GridSearchCV. Evaluation on data from 20 drivers across 80 simulated sessions showed a 17% reduction in mean absolute error and a 20% drop in mean squared error, enhancing prediction accuracy. Correlation coefficients between predicted and actual KSS more than doubled, confirming stronger alignment and increased reliability for real-time monitoring. Multimodal fusion further improved detection sensitivity and reduced false alarms by 25%. Distinct model coefficients across users indicate potential for personalized calibration. Overall, the results validate the feasibility of lightweight, interpretable, and scalable sleepiness estimation models for embedded ADAS, enabling continuous, real-time driver monitoring in safety-critical environments.

Keywords: Autonomous Vehicles and Drones, Diagnosis, Prognosis and System Identification, Cyber Physical Systems, Digital Twins and Knowledge Systems
Abstract: This article proposes a fault detection and isolation (FDI) strategy for quadcopter actuators, addressing lock-in-place (LIP), hard-over failure (HOF), Float, and loss-of-effectiveness (LOE) faults. The attitude kinematics are formulated using a unit-quaternion rotation matrix to avoid axis singularities. External disturbances are explicitly incorporated into the model, while actuator faults are parameterized in discrete time. Robust residuals are generated through eigenstructure assignment, where carefully selected left eigenvectors decouple the disturbance subspace while retaining sensitivity to actuator faults. Numerical experiments evaluate residual matrices over discrete time steps, analyzing both residual magnitudes and directions. The effectiveness of the proposed approach is evaluated in terms of isolation accuracy (IA), false alarm rate (FAR), and detection delay (Delta k).

Keywords: Power and Eneergy Systems
Abstract: In unbalanced distribution networks with radial topology, the use of sweeping methods is advantageous compared to traditional methods for solving power flow. Therefore, in this work, the well-known bi-quadratic equation for the nodal voltage magnitude through branch power flow is expanded to the three-phase distribution networks. As the sweep methods deal well with distribution network ill-conditions, due to their simplicity and fast convergence, further studies, variations, and enhancements of the algorithm have been constantly done. In order to increase the algorithm efficiency in terms of computational performance, the compensation technique for voltage-controlled nodes is also improved with the Newton-Raphson approach. To verify performance improvement, the proposed algorithm is evaluated in a 135 bus test network with radial topology and several electrical devices installed, such as capacitors, voltage regulators, and distributed generators. The results show the accuracy of the three-phase formulation and a better computational performance of the proposed compensation approach with a gain around 57%.

Keywords: Power and Eneergy Systems, Power Quality, Power Electronics
Abstract: Wind parks can interact with weak AC grids, potentially causing instability. This paper presents the use of Impedance-Based Stability Assessment for a type IV wind park connected to a weak grid. The generic EMT-model is used for representing the wind park. The results show that type IV wind turbines can interact with weak grid in supersynchronous frequencies, given the behaviour of its impedance in that range. Simulations in time-domain are used to validate the results. Bode Plot, Nyquist Criterion and the Vector Fitting algorithm were used to identify the instability and the frequency where it occurs.

Keywords: Power and Eneergy Systems, Optimization Heuristics and Methods, Industry Applications
Abstract: Um sistema de proteção em redes de distribuição de eletricidade é essencial para garantir a segurança e a eficiência do fornecimento. A coordenação e a otimização dos dispositivos de proteção são um problema complexo, cujo objetivo é garantir que as faltas sejam detectadas e isoladas adequadamente, para minimizar as interrupções e garantir a estabilidade da rede. Várias técnicas são empregadas para obter a coordenação ótima, incluindo métodos heurísticos, metaheurísticos (MHs) e modelos linearizados. Esses métodos têm suas limitações, pois não garantem soluções de boa qualidade e simplificam o problema ao reduzir o número de variáveis. O comportamento do algoritmo de gradiente reduzido generalizado (GRG) é apresentado neste artigo, devido sua capacidade de encontrar soluções de boa qualidade para coordenação e seletividade entre relés de sobrecorrente (OCRs), bem como coordenação entre OCRs e elos fusíveis. Fatores como qualidade da solução, eficiência computacional e robustez são analisados por meio de um estudo comparativo com uma MH apropriada, como o algoritmo genético. O algoritmo GRG tem várias vantagens, como melhor coordenação e seletividade entre OCRs e elos fusíveis, rápida capacidade de processamento computacional e resposta mais rápida do sistema de proteção, conforme demonstrado pelos resultados obtidos.

Keywords: Power and Eneergy Systems
Abstract: The expansion planning of electric distribution system (EPEDS) is modeled as a mathematical programming problem aimed at minimizing costs while satisfying a set of technical constraints. However, traditional models often fail to consider the practical experience of system operators when adjusting network topology or final costs according to their preferences. To address this limitation, this work proposes a multicriteria decision support methodology that incorporates the perceptions of operators and specialists regarding the relative importance of different planning aspects. The optimization is performed using the Tabu Search metaheuristic, while operator preferences are simulated through the Preference Ranking Organization Method for Enriched Evaluation (PROMETHEE), ensuring that the resulting solutions are not only technically sound but also aligned with decision-makers’ expectations. The decision criteria considered include the costs of constructing and/or repowering substations, the costs of constructing and/or reconductoring circuits, the operational costs of substations, active power losses, and system reliability indices. The proposed methodology was tested on a 27-node distribution system, demonstrating its ability to generate flexible, cost-effective solutions that respect operator-defined priorities.

Keywords: Power and Eneergy Systems, Renewable Energy
Abstract: A expansão da geração eólica no Brasil tem sido limitada por restrições de operação ligadas à infraestrutura de transmissão, especialmente em regiões com elevado potencial renovável. Nesse cenário, torna-se essencial dispor de metodologias que permitam estimar a capacidade de aumento de geração e identificar potenciais gargalos da rede elétrica. Este trabalho apresenta uma abordagem em duas etapas baseada em injeções incrementais de potência ativa nas barras do sistema, acompanhadas de ajustes proporcionais de carga e verificação dos limites operacionais de tensão e carregamento em regime permanente e avaliação sob contingências N-1, por meio de um Índice de Performance (PI) que quantifica a segurança operativa do sistema. O processo foi automatizado em linguagem Python, com simulações realizadas por meio do software ANAREDE. A aplicação foi realizada ao sistema teste simplificado de 107 barras, representativo da rede de transmissão brasileira. Os resultados demonstram que a metodologia proposta pode ajudar a diagnosticar restrições estruturais da rede para o planejamento estratégico da integração de novos parques eólicos.

Keywords: Artificial Intelligence, Deep Learning and Machine Learning, Industry Data Science Applications
Abstract: This paper presents an AI-driven methodology for segmenting and annotating digitized historical newspapers, with applications in museum and archival contexts. The proposed pipeline combines Tesseract OCR, image preprocessing, named entity recognition using spaCy, and automatic DOCX report generation to extract structured content from scanned newspaper pages. Applied to a dataset of 3,360 pages published between 2010 and 2024, the system demonstrated high performance in text block recovery, headline and date detection, and entity extraction. The approach is lightweight, language-adaptable, and suitable for integration into curatorial workflows in Latin America. Evaluation metrics and a case study are presented to illustrate the effectiveness of the method and its potential for enhancing access to historical press collections.

Keywords: Artificial Intelligence, Machine Learning, Industry Applications
Abstract: This study presents an end-to-end pipeline for the digitization, segmentation, and semantic processing of historical newspapers using a hybrid architecture that combines image- based and language-based models. The corpus, drawn from editions of the Diário dos Campos sampled from 1932 to 1984, was digitized using a planetary scanner and processed through Google Keep’s Optical Character Recognition (OCR) capabilities. Visual segmentation leveraged deep learning models based on convolutional neural networks (CNNs) and long short- term memory (LSTM) layers to extract and transcribe content from heterogeneous and degraded print layouts. The resulting textual data was structured using the Gema3 large language model, which enabled disaggregation of OCR outputs into distinct articles, title inference, removal of duplication, and thematic classification. The final dataset consists of semantically enriched, searchable newspaper articles, offering significant utility for cultural heritage preservation, historical research, and digital humanities initiatives. The pipeline demonstrates high adaptability, low cost, and scalability for similar archives in Latin America and beyond.

Keywords: Artificial Intelligence, Deep Learning and Machine Learning, Industry Applications
Abstract: This paper presents a case study on the implementation of artificial intelligence (AI), semantic web technologies, and the Omeka digital repository platform in the digital preservation efforts of university museums in southern Brazil. Centered on the transition from Omeka Classic to Omeka S at the Multiuser Laboratory for Digital Humanities and Innovation (LAMUHDI), based at the Campos Gerais Museum (MCG), the study explores how metadata interoperability, AI-assisted workflows, and linked open data contribute to more accessible, inclusive, and sustainable memory preservation. The project is part of the broader NAPI (New Research and Innovation Arrangements) initiative, funded by Fundação Araucária, which seeks to establish a federated network of interoperable repositories across higher education institutions in Parana. By integrating tools such as Optical Character Recognition (OCR), Resource Description Framework (RDF), and Dublin Core metadata standards within the Omeka S environment, the initiative positions Digital Humanities as a hub for innovation and public engagement in cultural heritage dissemination.

Keywords: Artificial Intelligence, Machine Learning, Industry Applications
Abstract: The recent development of Agentic AI systems, empowered by autonomous large language models (LLMs) agents with planning and tool-usage capabilities, enables new possibilities for the evolution of industrial automation and reduces the complexity introduced by Industry 4.0. This work proposes a conceptual framework that integrates Agentic AI with the intent-based paradigm, originally developed in network research, to simplify human–machine interaction (HMI) and better align automation systems with the human-centric, sustainable, and resilient principles of Industry 5.0. Based on the intent-based processing, the framework allows human operators to express high-level business or operational goals in natural language, which are decomposed into actionable components. These intents are broken into expectations, conditions, targets, context, and information that guide sub-agents equipped with specialized tools to execute domain-specific tasks. A proof of concept was implemented using the CMAPSS dataset and Google Agent Developer Kit (ADK), demonstrating the feasibility of intent decomposition, agent orchestration, and autonomous decision-making in predictive maintenance scenarios. The results confirm the potential of this approach to reduce technical barriers and enable scalable, intent-driven automation, despite data quality and explainability concerns.

Keywords: Deep Learning and Machine Learning, Artificial Intelligence, Machine Learning
Abstract: This paper explores how Explainable AI (XAI) techniques, applied to machine learning models like Transformers and LSTMs, can help analyze behavioral patterns in gaming. Drawing on insights from the field of educational games for neurodivergent children — where explainability is crucial for success — We propose adapting these approaches to serious games for professional training. By capturing worker engagement and explaining performance outcomes, XAI can support not only the workers themselves, but also instructors and managers in comprehending why failures occur. Our experiments with Portuguese gaming datasets show that LSTM models, combined with explainability tools like LIME, outperform Transformers with a relative gain of 7% for LSTM, when identifying nuanced emotional patterns. The obtained metrics from the validation phase are an accuracy of 0.78 for the LSTM model and 0.73 for the Transformer. We argue that such explainable insights can enhance professional training by providing more targeted feedback and improving the overall training process.

Keywords: Industry Applications, Renewable Energy, Personalized Products
Abstract: Passive ejectors, which require no additional energy input, offer an efficient alternative to mechanical pumps for recirculating H₂ on the anode side of PEMFCs. In this article, a 3-D CFD simulation is performed for an ejector serving a stationary 3 kW PEMFC. The results are compared with data from the literature. The main points assessed were: (i) the effect of primary and secondary pressures on the entrainment ratio; (ii) the influence of the anode-return and feed temperatures; and (iii) the impact of N₂ and H₂O molar fractions in the secondary flow. Simulation results indicate that the ejector’s optimal operating pressure lies between 2.0 and 2.5 bar, a range in which the entrainment ratio reaches its maximum. For temperatures above 40 °C, the entrainment ratio decreases by roughly 0.5 % per degree Celsius due to the drop in gas density. Moreover, higher N₂ or H₂O content increases the entrainment ratio but lowers the anode stoichiometric ratio. As design guidelines, we recommend operating within the optimal pressure range, keeping the anode return below 40 °C, and limiting N₂ and H₂O according to the desired stoichiometry.

Keywords: Industry Applications, Robotics and Mechatronics, Virtualization, Simulation Techniques and Augmented Reality
Abstract: The use of Hierarchical Systems (HS) technology in conceptual design of industrial robot and its work cell (RW) is proposed in the paper. In comparison with known design, AI, and mathematical models, HS conceptual formal models contain connected models of the robot and its work cell subsystems, their processes, RW structure, its dynamic representation in its environment and RW coordinator. The design and control system of RW is given in the form of HS coordinator. Visual Components system was applied in the paper for RW detailed design. Conceptual model of the RW is considered in the paper first. Implementation of the Visual Components software for RW detailed design of the RW is described after that. Third, the laboratory test with KUKA KR 6 R700-2 robot is presented. The results and conclusive remarks are finally given in the paper.

Keywords: Industry Applications, Resource Efficienty & Circular Economy Tracking, Personalized Products
Abstract: O cimento Portland (CP) é a pozolana mais utilizada como aglomerante para concreto, mas seu processo de fabricação requer altas temperaturas e queima de combustíveis fósseis, gerando alta emissão de gás carbônico, principal gás do efeito estufa. Para reduzir esse impacto, são estudados ligantes inorgânicos compostos por uma fonte de aluminossilicato, solução alcalina e resíduos industriais como substitutos do CP, chamados de geopolímeros. Sabe-se também que o consumo de água tem aumentado a cada ano e, consequentemente, o lodo proveniente da etapa de decantação do tratamento de água também cresce, dificultando seu descarte. A adição de lodo de estação de tratamento de água (LETA) em geopolímeros vem sendo gradualmente estudada. Nesse contexto, investigou-se o comportamento mecânico e físico do geopolímero à base de metacaulim e LETA. Assim, foram produzidos geopolímeros de metacaulim com diferentes quantidades de LETA e feitos testes de resistência à compressão e absorção de água, variando o tratamento térmico do lodo e do geopolímero. O maior valor de resistência à compressão alcançado para os geopolímeros com LETA ocorreu para amostras curadas em temperatura ambiente com 5% de substituição do metacaulim por lodo sem calcinação. Os valores de absorção de água foram próximos para todas as amostras de geopolímeros e superaram os valores obtidos para o cimento Portland.

Keywords: Industry Applications, Optimization Heuristics and Methods, Diagnosis, Prognosis and System Identification
Abstract: Este trabalho propõe a otimização de um processo de decapagem química industrial por meio da aplicação da técnica de Projeto e Análise de Experimentos, utilizando o método de Taguchi como ferramenta estatística de apoio à tomada de decisão. A decapagem química é essencial para a remoção de oxidações superficiais em metais, especialmente em aços de baixo carbono, sendo o HCl o ácido mais comumente empregado devido à sua eficiência. A pesquisa destaca a importância da experimentação planejada para compreender a influência de variáveis críticas no processo e promover melhorias em qualidade, produtividade e eficiência operacional. O método adotado possibilita reduzir significativamente o número de experimentos necessários, mantendo a precisão dos resultados. Além disso, o estudo contribui para a expansão do conhecimento sobre a decapagem química sob uma perspectiva gerencial e estatística, pouco explorada na literatura técnica e acadêmica, tornando-se uma referência prática para a indústria na busca pela excelência operacional.

Keywords: Industry Applications, Virtualization, Simulation Techniques and Augmented Reality, Cyber Physical Systems, Digital Twins and Knowledge Systems
Abstract: In an era marked by uncertainty and disruption, supply chain (SC) resilience and robustness have become strategic imperatives. This study introduces the Supply Chain Resistance Index (SCRESIS Index), which was developed employing simulation methods. Digital Twin and artificial neural network technologies are conceptually discussed as potential enablers for future applications; however, they are not implemented in this study. The proposed framework enhances adaptability and stability by integrating resilience and robustness. The results demonstrate that combining rapid adaptation with structural resistance provides a competitive advantage in volatile environments.

Keywords: Power Quality
Abstract: This work presents the development of a low-cost datalogger for monitoring the charging of electric vehicles and analyzing power quality. An ESP32 microcontroller and an MCP3208 A/D converter were used. The system collects 30,000 voltage samples per second across three different channels (one for each phase), and 3,000 current samples per second across three different channels (1,000 samples per channel). The data is stored on a microSD memory card for subsequent analysis of 16 seconds of one phase of the collected signal. The device enables the evaluation of voltage and current variations during the charging process, which is essential to ensure both efficiency and safety. The results demonstrate the datalogger's capability to capture high-frequency data, making it a viable and cost-effective solution for monitoring the electrical grid during electric vehicle charging.

Keywords: Power Quality, Smart Grids, Industry Applications
Abstract: The optical voltage transducer offers several advantages over traditional inductive and capacitive transformers, such as wide bandwidth, complete galvanic isolation, and immunity to electromagnetic interference. These voltage transformers can be designed using electro-optical amplitude modulators, which are based on the Pockels effect in crystals such as lithium niobate. The quadrature switching method is employed to obtain two signals from the photodetected one. Inflection points from these two signals are then retrieved using the Discontinuity Point Judgment (DPJ) method. Based on the preliminary converted signal and the identified inflection points, an orthogonal signal is constructed. This approach achieves the quadrature condition without the need for a feedback control loop. The two resulting quadrature signals are subsequently analyzed using the high-gain and sliding mode control approach, which involves employing the control system to fully compensate for the phase shifts induced by the voltage applied to the sensor crystal. Data were collected using a myRIO card and stored in .lvm files; later, they were compiled offline using Matlab. Sinusoidal signals ranging from 7 kV to 14 kV (peak-to-peak) were tested, with amplitude relative errors remaining below 1% when compared to a high-voltage (HV) probe reference.

Keywords: Power Quality, Power and Eneergy Systems, Diagnosis, Prognosis and System Identification
Abstract: Resilience is a relatively new concept in microgrids and is distinctly different from traditional concepts such as reliability and security in electrical systems. When analyzing critical mission microgrids, it is essential to consider the specific characteristics and particularities of the system. In this article, we evaluate and discuss resilience quantifiers applied to the critical mission microgrid implemented at the Alcântara Launch Center, which has been operational since early 2024. Scenario studies are conducted under islanded operation conditions. The results showed maximum resilience, with no load shedding under textit{N-1} contingencies, due to generation oversizing to maintain supply to critical loads.

Keywords: Power and Eneergy Systems, Power Quality
Abstract: The objective of this work is to employ an empirical approach to simplify the state matrices that represent a transmission line, describing the equivalent π circuit directly in the time domain through state-space equations. The formation rules of these matrices are addressed in an overview manner, with emphasis on the matrices related to the sensitivity study, enabling the empirical removal of terms that do not significantly impact the system response. This results in a simplified matrix with straightforward computational implementation. For validation purposes, a simplified model was compared to a model based on formation rules validated in previous studies, both simulated in MATLAB and in the time domain, considering energizations with step inputs. The results demonstrate the feasibility of the proposed simplification, highlighting its applicability in electromagnetic transient studies and its computational cost efficiency without compromising system accuracy. It is important to note that the models analyzed in this work are simple and characterized by a low number of poles and branches.

Keywords: Power Quality, Electrical Machines and Drives
Abstract: O artigo avalia a qualidade da energia em um sistema real de bombeamento de água, destacando a importância da conformidade com os padrões estabelecidos pelo Procedimentos de Distribuição de Energia Elétrica no Sistema Elétrico Nacional - PRODIST e a norma técnica ABNT NBR 17094-1. São analisados distúrbios elétricos, como variações de tensão, desequilíbrio entre fases, flutuações de frequência e distorções harmônicas, utilizando medições reais obtidas com um analisador de classe A no prédio do Centro de Excelência em Eficiência Energética da Amazônia - CEAMAZON (UFPA). Os resultados mostram que, embora a maioria das leituras de tensão e frequência esteja dentro dos limites aceitáveis, eventos críticos prolongados, como quedas de tensão abaixo de 110 V, foram registrados, podendo comprometer o desempenho e a vida útil dos motores. Além disso, o artigo discute os impactos desses distúrbios em motores de indução, incluindo aumento de perdas, superaquecimento e redução de eficiência.

Keywords: Machine Learning, Power and Eneergy Systems, Smart Grids
Abstract: The monitoring of electrical distribution systems plays a fundamental role in the development of modern society and is a challenging task due to its radial nature and the limited number of meters installed. Generally, the method based on weighted least squares has been applied to this problem with the main objective of determining, with a high level of precision, the values of the magnitudes and phase angles of the voltage at all the nodes of the network, allowing other quantities of interest such as power flows to be calculated. In this context, this article proposes a graph neural network for estimating the state of electrical distribution networks. The proposal integrates layers of a graph network that models the topological relationships of neighboring nodes, together with a simple linear network, where the combination of both components extracts the complex characteristics and non-linear relationships of the data. Generation and demand profiles drawn up from historical data were used to create the database. The results obtained from the 13 and 123 node IEEE test networks showed significant accuracy, demonstrating the viability of the approach for small and medium-sized systems.

Keywords: Machine Learning, Artificial Intelligence, Internet of Things
Abstract: A detecção e a localização de vazamentos em Redes de Distribuição de Água (RDA) constituem atividades fundamentais para garantir a eficiência operacional e a sustentabilidade dos sistemas de abastecimento urbano. Neste contexto, o presente estudo investiga o impacto da substituição da matriz de adjacência binária por uma matriz de ponderamento baseada nas distâncias físicas entre os nós no desempenho de modelos de Redes Neurais de Grafos (GNNs) aplicados à tarefa de localização de vazamentos. A abordagem proposta explora a correlação entre a topologia da rede hidráulica e a estrutura do grafo utilizado na modelagem, considerando medições de pressão hidráulica obtidas em pontos estratégicos da RDA. Foram avaliadas as arquiteturas Gated Graph Neural Network (GGNN) e GraphSAGE, com os agregadores Mean e LSTM, aplicadas a uma rede simulada no software EPANET. Os resultados obtidos indicam que a utilização de uma matriz de ponderamento com base na distância física entre os nós contribui para a melhoria da acurácia na localização de vazamentos, especialmente em arquiteturas que dependem fortemente das informações de vizinhança, como a GGNN e o GraphSAGE-Mean. Essa conclusão reforça a relevância da incorporação de características físicas da RDA na estrutura dos grafos, de modo a potencializar a capacidade expressiva dos modelos de GNN em aplicações de diagnostico em sistemas hidráulicos urbanos.

Keywords: Deep Learning and Machine Learning, Machine Learning, Human Machine Symbiosis
Abstract: Training neural networks with EEG data is often challenging due to small sample sizes. The main objective of this work is to implement a data augmentation strategy for EEG datasets, based on the fact that EEG recordings are composed of extensive strings of signals. The method splits subsections of the original data while maintaining key information regarding overall synaptic dynamics, which is then provided for training and evaluating predictive models. We observe that it is possible to implement deep learning algorithms with improved accuracy using the presented method. The best mean accuracy achieved using only the proposed data augmentation method was 0.89, while the same model trained on the unaltered data had an accuracy of 0.69, and the augmented model also showed to be more reliable than the original dataset with a p-value of 0.0016 on the McNemar's test. We also investigate the influence of some random factors associated with the dataset and obtained results, as well as how such variables may be addressed in the future in order to further enhance training procedures using the augmentation method.

Keywords: Machine Learning, Big Data in Industry Applications, Artificial Intelligence
Abstract: This study explores the application of Machine Learning techniques to predict delays in a public bus system, using the city of Curitiba, Brazil, as a case study. Leveraging historical data on vehicle operations, the research frames the problem as a binary classification task, determining whether a bus is “DELAYED” or “ON TIME” at a specific moment. Key features such as route code, direction of travel, and geolocation (latitude and longitude) were extracted and preprocessed. The methodology included data cleaning, class filtering, dataset sampling, one-hot encoding of categorical variables, and division into training and testing subsets. A Logistic Regression model was trained on the resulting dataset. Performance was assessed using standard classification metrics—accuracy, precision, recall, and F1-score—as well as a confusion matrix for interpretability. The model achieved an accuracy of 0.9677, precision of 0.9135, recall of 0.8501, and F1-score of 0.8806, indicating robust predictive performance with a balanced trade-off between false positives and false negatives. These results highlight the potential of Machine Learning models to support urban mobility planning by anticipating delays and enhancing real-time passenger information systems.

Keywords: Electrical Vehicle and Energy Storage, Machine Learning, Artificial Intelligence
Abstract: This paper presents a hybrid modeling framework that integrates physics-informed empirical equations with supervised machine learning to estimate the State of Health (SoH) of lithium-ion batteries using real operational data. A sixmonth dataset from a LiFePO4 battery system (24V, 160Ah) was used to combine simplified aging equations with Random Forest regression. Additionally, a synthetic SoH proxy was constructed to emulate operational variability in the absence of measured capacity data. The model achieved low error rates under weekly cross-validation (MAE ∼ 0.0002%, R2 up to 0.99), though these results were strongly influenced by the specific SoH behavior in the dataset. The empirical model produced nearconstant degradation, limiting predictive complexity, whereas the proxy introduced greater variability but lacked physical grounding. The proposed approach is suitable for early-stage model development in data-limited environments. However, broader validation across multiple systems and with real SoH measurements is necessary to confirm its robustness and practical applicability in battery lifecycle management.

Keywords: Power and Eneergy Systems, Industry Applications
Abstract: This work evaluates radiality constraint formulations for optimizing the topology of radial distribution systems, with a focus on computational efficiency. The goal is to identify network configurations that minimize active power losses while maintaining bus voltages and line currents within operational limits. The analyzed frameworks employ a mixed-integer second-order cone programming model, compatible with commercial optimization solvers. Four radiality formulations are tested using the 14-, 33-, 69-, 84-, 118-, 119-, and 136-bus systems. Results demonstrate that a directed graph-based model for radiality delivers the highest efficiency for large-scale networks.

Keywords: Power and Eneergy Systems, Industry Applications
Abstract: The continuous growth of electricity demand and the global transition towards dependence on clean energy sources intensify the need for efficient and reliable long-term expansion planning of transmission networks. In this framework, transmission network expansion planning (TNEP) is a challenging problem involving nonlinear constraints and discrete decision variables, with inherent complexity often resulting in high computational costs, especially with more accurate models such as the disjunctive formulation for the DC model. This paper proposes an enhanced linear formulation for the TNEP problem that incorporates the power transfer distribution factor (PTDF) concept to reduce problem dimensionality and computational effort. The PTDF is a strategic tool to reduce the number of decision variables and constraints, enabling a more scalable optimization of the problem. Comparative results across several test systems confirm the effectiveness of the model and its scalability.

Keywords: Power and Eneergy Systems, Optimization Heuristics and Methods, Industry Applications
Abstract: This paper presents a hybrid optimization approach for the optimal power flow problem using an enhanced tabu search integrated with the KNITRO solver. The method handles binary and integer variables, including discrete tap settings and shunt statuses, while continuous variables are optimized via nonlinear programming. A memory-efficient elite-based restart strategy and path-relinking intensification are proposed to enhance global exploration and convergence speed. The performance of the proposed approach is validated on the IEEE 14-, 30-, 57-, and 118-bus systems, demonstrating its effectiveness in finding high-quality solutions when compared to traditional mixed-integer nonlinear programming solvers.

Keywords: Power and Eneergy Systems, Smart Grids, Renewable Energy
Abstract: The continuous energy supply to critical loads is vital for the resilience of essential infrastructures. The increasing frequency of extreme weather events increases concerns about electrical system reliability, making uninterrupted operation crucial for public safety, health, and economic stability. In this context, hybrid renewable energy systems (HRESs) emerge as a reliable and environmentally solution for backup systems, enabling extended service during main grid contingencies. This paper proposes a multi-criteria Analytic Hierarchy Process (AHP)-based decision support framework for managing non-essential loads in substations during power outages. The method facilitates intelligent decision-making by considering the dynamic availability of HRES resources, which integrate wind and solar power, battery energy storage, and electric vehicles, alongside the critical load. This approach defines three distinct operational modes for non-essential loads, enabling autonomous consumption adjustment and ensuring uninterrupted power supply to essential loads. The tests and simulations confirm the framework’s effectiveness in capturing variations in renewable generation and storage system state of charge, optimizing available energy utilization. Results demonstrate the proposed framework’s significant potential to support the substation’s operational resilience and ensure critical load supply continuity, promoting more effective energy management and increasing system availability during power outages.

Keywords: Power and Eneergy Systems, Renewable Energy, Optimization Heuristics and Methods
Abstract: The high penetration of distributed energy resources (DERs) may cause problems in the operation of the power network in the future. An alternative to mitigate these problems is the implementation of operating envelopes (OEs), which restrict the export and import of power. Commonly, a fixed OE is used to limit the operation of DERs; however, in recent years, this strategy has become excessively restrictive. In this context, new approaches have emerged, one of them being dynamic operating envelopes (DOEs), which allow the use of DERs throughout the day in a more efficient and less restrictive way. This work proposes a nonlinear programming mathematical model aimed at the calculation of DOEs for each active customer. The tests were carried out on the IEEE 34-bus network, considering both photovoltaic systems and electric vehicle charging stations, obtaining the DOEs that allow extracting the maximum benefit from DERs while ensuring safe and reliable network operation.

Keywords: Electrical Machines and Drives, Industry Applications, Power Electronics
Abstract: Este trabalho avalia a substituição de um motor trifásico com rotor bobinado e partida por reostato líquido por uma máquina de nove fases alimentada por inversor, destacando os ganhos em confiabilidade, segurança elétrica, redução na complexidade de manutenção e impacto econômico .

Keywords: Automation and Process Control, Industry Applications
Abstract: Purpose: the construction sector in Brazil, despite its social and economic importance, faces bottlenecks in its development, such as the slowness of building permit processes. The aim of this study is to fill a gap identified in the Brazilian scientific literature, going beyond diagnosing problems in licensing, but rather seeking solutions to overcome the barriers identified, describing the results of a partnership between the construction industry, government and academia to improve licensing processes, carried out in a sustainable manner.

Design/methodology/approach: applied research, with exploratory and descriptive objectives, qualitative approach, method of Soft Systems Methodology for investigation of problems within a system to improve licensing processes, using Statistical Process Control combined with Value Stream Mapping. Data was collected from the city hall's and fire department’s protocol system and from the experience of its employees.

Findings: the results had a positive impact on the sustainable licensing processes for buildings in the municipal governments of Três Rios, Nova Friburgo, Magé, Maricá and Miguel Pereira, located in the state of Rio de Janeiro, as well as in the Military Fire Brigade of the state of Rio de Janeiro.

Originality: the subject of innovation and modernization in urban licensing is relevant and has been little explored in the academic field. There is a scarcity of papers dealing with initiatives to support municipalities in improving the performance of their work processes, especially in sustainable construction licensing, justifying this research.

Keywords: Industry Applications, Automation and Process Control, Electrical Machines and Drives
Abstract: A água é um recurso natural essencial para a sobrevivência humana, mas para que possa ser consumida por nossa população, necessita ser tratada. Para realizar esse tratamento utiliza-se as Estações de Tratamento de Água (ETAs), sendo o modelo de ciclo completo o mais utilizado, pela sua capacidade de processar variações na qualidade da água. Com o objetivo de proporcionar sugestões de inovações tecnológicas ao modelo de ETA de ciclo completo, como o controle a malha fechada da velocidade das pás agitadoras e medição instrumentada da vazão de entrada da água, este trabalho acadêmico registra a descrição de um projeto, o processo construtivo, a operação e validação de uma ETA didática em escala reduzida de acordo com as normas NBR 12216/1992, Portaria 888/2021 e literaturas de referência. A ETA didática foi construída em vidro transparente, a instrumentação e a automação do controle de velocidade dos agitadores foram realizadas com sensores eletrônicos e microprocessamento, o que viabilizou a realização de testes parametrizados e monitorados em tempo de execução, com a obtenção de resultados compatíveis com a prática do ensino e pesquisa tecnológica.

Keywords: Industrial Ultrasound Theory and Application, Industry Applications
Abstract: Este trabalho apresenta um método não invasivo para a medição de temperatura em células ultrassônicas de baixa frequência (10 MHz), aplicada à caracterização de líquidos. A precisão dessas medições depende da temperatura do líquido, uma vez que variações térmicas afetam propriedades acústicas. Além disso, métodos convencionais de medição, que utilizam sensores em contato direto com o líquido, podem introduzir contaminações. Para contornar essas limitações, propõe-se uma abordagem baseada em estimação de estados como o filtro de Kalman Estendido (EKF), eliminando a necessidade de sensores internos e evitando interferências. Foram realizados dois ensaios experimentais, modulando-se a corrente aplicada por meio de sinais chirp, nos seguintes cenários: i) somente aquecimento e ii) aquecimento e resfriamento. O EKF alcançou índices RMSE de 1,13% e 2,51% para os cenários 1 e 2, respectivamente. Os resultados obtidos sugerem que o EKF é um método promissor para a estimação de sinais térmicos, tornando-se uma ferramenta útil para aplicação de métodos ultrassônicos.

Keywords: Virtualization, Simulation Techniques and Augmented Reality, Machine Learning, Artificial Intelligence
Abstract: Abstract — The inspection of fire protection engineering projects is essential to ensure safety in buildings and public spaces. In Brazil, the Military Fire Departments are responsible for this analysis, which requires a high level of attention and technical detail. Due to the complexity and workload involved, technological solutions, such as the use of intelligent systems, are necessary to automate and optimize this process, reducing errors and approval time. This study presents the development of a customized dataset for training a symbol detection model in fire protection projects using the YOLOv8 architecture. To achieve this, a dataset containing 400 real images of fire protection projects was created, with 3,120 annotated instances distributed across 12 classes of standardized fire safety symbols, following technical regulations. The results demonstrated the high efficiency of the model, achieving a mAP@50 of 92.1% during validation and 96.2% in the final test set. These findings highlight the potential of deep learning to automate the analysis of technical drawings, improving both the speed and accuracy of fire safety project evaluations.

Keywords: Renewable Energy, Power and Eneergy Systems
Abstract: There is a growing demand for renewable energy sources to replace fossil fuels, including solar energy. Its intermittency and the restricted energy generation interval are obstacles to this source becoming the main source of energy for society. To mitigate these issues, a potential solution is its implementation combined with storage systems, which allow the management of the energy generated. However, this solution must be economically attractive to make the implementation viable. One way to make it viable is to store the energy generated during periods when the tariff is cheaper and use it during periods when the tariff is more expensive, generating savings on the energy bill. This paper presents a methodology for the economic evaluation of the implementation of energy storage systems in residential consumer units that opt for the white hourly tariff modality with the implementation of a photovoltaic generation system, verifying the sensitivity to the specific cost (R/kWh) of the storage systems, and applying it to a case study. It is concluded that the implementation of storage systems for the studied scenario is not viable.

Keywords: Virtualization, Simulation Techniques and Augmented Reality, Renewable Energy, Power and Eneergy Systems
Abstract: Diante da crescente pressão por sistemas de energia mais limpos e eficientes, este estudo apresenta o modelo de Fluxo de Potência Ótimo por Sensibilidade de Carbono (FPO-SC), uma abordagem inovadora que alia viabilidade econômica à mitigação das emissões de CO2 no planejamento e operação de sistemas elétricos. A proposta incorpora as Emissões Marginais Locacionais (LMEs), um indicador estratégico que revela como variações na demanda impactam diretamente os níveis de emissão, permitindo decisões mais assertivas e ambientalmente responsáveis. Utilizando ferramentas consolidadas como CVX e MATPOWER, o FPO-SC otimiza simultaneamente os custos operacionais e o fluxo de carbono, assegurando a confiabilidade do sistema e atendendo metas ambientais de forma eficaz. Aplicado ao sistema-padrão IEEE 118 barras, o modelo demonstrou significativa redução nas emissões sem comprometer a estabilidade da rede. Os resultados evidenciam o potencial do FPO-SC como instrumento decisivo para o avanço de políticas públicas e estratégias voltadas à transição energética. Ao mapear com precisão os pontos críticos de emissão e sensibilidade do sistema, o estudo oferece subsídios técnicos valiosos para operadores, planejadores e reguladores comprometidos com um futuro energético sustentável.

Keywords: Smart Grids, Power and Eneergy Systems, Renewable Energy
Abstract: A crescente inserção de geração distribuída intermitente nas redes elétricas tem gerado desafios relevantes para o planejamento dos sistemas de distribuição. Para enfrentar esses desafios, torna-se necessário estimar com precisão o potencial de geração fotovoltaica em telhados residenciais urbanos. Este artigo apresenta um método inovador composto por três etapas integradas: (i) estimativa do potencial de mercado para sistemas fotovoltaicos (SFV) residenciais; (ii) classificação dos consumidores com base no tipo e na quantidade de eletrodomésticos presentes nos domicílios; e (iii) dimensionamento dos SFV necessários para suprir a demanda energética característica de cada perfil de consumidor. O método proposto é aplicado com base em dados socioeconômicos e informações geoespaciais de uma cidade brasileira de médio porte. Os resultados indicam variações significativas entre os setores censitários, tanto em relação à capacidade econômica para aquisição de SFV quanto aos padrões de consumo energético, evidenciando as limitações das abordagens que utilizam estimativas homogêneas. Esses achados demonstram que o método proposto representa com maior precisão a heterogeneidade espacial da adoção urbana de SFV; assim, oferece um suporte técnico mais robusto para a avaliação dos impactos dessa tecnologia nas redes elétricas e para a formulação de estratégias de planejamento energético sustentável.

Keywords: Renewable Energy, Optimization Heuristics and Methods, Power and Eneergy Systems
Abstract: Energy communities facilitate the more efficient use of distributed energy resources, such as photovoltaic systems, to meet high-demand loads, including public charging stations for electric vehicles. However, to ensure that this integration contributes to improving the quality indicators of the electric energy delivered by the power distribution network, it is essential to define both the connection point and the operational area of the energy community. This paper presents a methodology for forming energy communities in urban distribution networks by clustering photovoltaic systems and public charging stations for electric vehicles to minimize voltage drops within their operational area. The approach integrates geospatial analysis, reinforcement learning techniques, and a Density-Based Spatial Clustering of Applications with Noise Algorithm to identify optimal configurations. We applied the proposal to a power distribution feeder in the city of Santo André, in the state of São Paulo, Brazil. The results indicate a potential reduction of up to 15% in voltage drops for the demand of charging stations, thereby improving voltage profiles across the network and enhancing the local use of solar energy to meet electric vehicle demand.