A multi-scale smart fault diagnosis model based on waveform length and autoregressive analysis for PV system maintenance strategies

Nonlinear photovoltaic (PV) output is greatly affected by the nonuniform distribution of daily irradiance, preventing conventional protection devices from reliably detecting faults. Smart fault diagnosis and good maintenance systems are essential for optimizing overall productivity a PV system improving its life cycle. Hence, multiscale smart model improved strategies proposed. This study focuses on diagnosing permanent faults (open-circuit faults, ground line-line faults) temporary (partial shading) in arrays, using Random Forest algorithm to conduct time-series analysis waveform length autoregression (RF-WLAR) as main features, with 10-fold cross-validation MATLAB/Simulink. The actual irradiance data at 5.86 °N 102.03 °E were used inputs produce simulated that closely matched on-site data. Fault database 2 MW power plant Pasir Mas Kelantan, Malaysia, field testing verify developed model. RF-WLAR achieved an average fault-type classification accuracy 98%, 100% classifying partial shading