<![CDATA[This paper presents an approach to detect faults in Photovoltaic (PV) systems based on a state estimation and residual analysis. After constructing the mathematical model of the nonlinear system and the extended state observer, we establish the fault detection method of the Single-Phase Grid connected PV system subjected to external disturbances based on estimation and residual analysis of nominal and extended states. To manage the uncertainties due to the modeling and external disturbances, an active disturbance rejection control (ADRC) is used, thanks to its robustness. We generated residuals by comparing the actual and estimated states with a threshold set at a tolerance of 5% from the nominal residual value. The external disturbances, such as PV generator and grid voltage variations, are defined as the external sources of disturbances. The faults occasioned by these disturbances are detected by the presence of peaks exceeding the thresholds. The results obtained by simulation in MATLAB environment demonstrated that with a threshold set at a tolerance of 5% from the nominal residual value, the proposed residual analysis method achieves 62.5 % of detection of faults from the PV source and 100% detection of faults from the grid side. The state estimation-based approach is verified by a direct visual observation of the nominal (current and voltage) and extended (disturbances) state estimation curves. Given the satisfactory results of this work, this diagnosis approach offers an interesting outlook for ensuring the productivity and lifespan of the PV system.]]>
Copyrights © 2026
