ALRIFAI, Yehya (Univ. Bordeaux, ESTIA INSTITUTE OF TECHNOLOGY), AGUILERA GONZALEZ, ADRIANA (ESTIA - INSTITUTE OF TECHNOLOGY), Vechiu, Ionel (ESTIA Recherche)

Fault Detection and Diagnosis of PV Systems Using Kalman-Filter Algorithm Based on Multi-Zone Polynomial Regression

Scheduled for presentation during the Regular Session "Observers and Estimation" (WeA1), Wednesday, November 16, 2022, 17:00−17:20, MAIN ROOM - E406

16th European Workshop on Advanced Control and Diagnosis, November 16-18, 2022, Nancy, France

This information is tentative and subject to change. Compiled on April 25, 2024

Abstract

Faults must be timely diagnosed, detected, and identified to enhance photovoltaic PV system’s dependability. In this context, this paper presents a novel Fault Detection and Diagnosis (FDD) methodology based on a hybrid combination of model-based, through Kalman Filter (KF), and a statistical data-driven regression approach for online monitoring of a PV system’s DC side. This statistical approach is formulated on Multi-Zone non-linear Polynomial regression (MZP) techniques of PV characteristics under Global Maximum Power Points (GMPP) at the array level. In particular, the proposed method effectively detects intermittent soft Short-Circuit (SC) even at very low irradiation. The performance of the proposed FDD methods is evaluated via MATLAB/Simulink® considering varying * weather conditions.