Enhanced Fault Detection in Photovoltaic Systems Through Hybrid SVM Evolutionary Optimization Techniques

Abstract

Fault detection in photovoltaic (PV) systems is vital for maintaining optimal performance. Early detection of faults can prevent downtime and minimize energy loss. In this study, An approach for fault detection in PV systems is proposed. The method integrates a hybrid support vector machine (SVM) with optimization algorithms such as particle swarm optimization (PSO), genetic algorithm (GA), Bayesian optimization (BO), and Randomized Search CV (RS). Experimental results demonstrate the effectiveness of the approach, notably the SVM-PSO variant achieving a significant precision in fault detection accuracy. Specifically, employing the RBF kernel, the SVM-PSO model exhibits an accuracy of 98.24%, precision of 98.29%, recall of 98.25%, and an F1 score of 98.08%. In contrast, utilizing the linear kernel yields slightly lower performance, with an accuracy of 89.47%, precision of 89.82%, recall of 89.47%, and an F1 score of 89.51%. The proposed system enhances performance and reliability, ultimately leading to increased energy generation and reduced maintenance costs.