Fault Detection and Diagnosis for Photovoltaic Array Under Grid Connected Using Support Vector Machine

Abnormal conditions in a solar photovoltaic (PV) array whether permanent or temporary faults lead to lessening the overall PV system efficiency and might lead to a fire hazard. As a result, it is compulsory to detect and diagnose faults in the PV array in order to ameliorate system reliability, safety, and efficiency. This paper proposes an automatic Fault Detection and Diagnosis (FDD) for the PV array under a grid-connected PV system operation based on the Support Vector Machine (SVM). The PV system utilizes two power processing stages consist of a DC-DC boost converter, followed by the three-phase, two-level, Voltage Source Inverter (VSI). The DC-DC boost converter is used to draw out the maximum PV array power and boosting the PV array output voltage. The Fuzzy Logic Control (FLC) based on the Maximum Power Point Tracking (MPPT) is applied to fine-tune the duty cycle for the DC-DC boost converter to realize maximum power. The VSI is used to inject a sinusoidal current into the grid through the LCL filter. The control strategy of the VSI consists of two control loops, the DC-link voltage and the AC-current control loop. The proposed system is simulated using MATLAB/Simulink (R) to investigate system performance.