Optimization of SVM classifier using Grid Search Method for Line-Line Fault Detection of Photovoltaic Systems

Effective monitoring is very crucial for the long-term performance of Photovoltaic (PV) systems. Line-Line (LL) fault should be detected precisely because it is not scrutable easily by traditional protection devices under low mismatch level and high fault impedance. This paper proposes a novel method for Line-Line fault detection by extracting the main features from Current-Voltage (I-V) curves of PV arrays. In this paper grid search (GS) method used to obtain a reliable method for Line-Line (LL) fault detection by optimizing the parameters of the kernel for Support Vector Machines (SVM) classifier. The simulation results verify that this novel method precisely detects Line-Line (LL) faults in PV arrays with an accuracy of 100%.