Fault diagnosis of photovoltaic strings by using machine learning‐based stacking classifier

Photovoltaic (PV) modules are prone to short circuits, open cracks, which can bring serious harmful effects. It is difficult establish the corresponding PV fault models diagnose status of strings. The paper proposes a machine learning-based stacking classifier (MLSC) for accurate diagnosis Specifically, operating state modules, parameter sensitivity algorithm used analyze impact characteristic factors on characteristics modules. Then based (irradiance, temperature, current, and power), MLSC proposed realize This structure integrate all kinds classifiers by play each classifier. combines various types learning algorithms improve overall classification effect using advantages Finally, experiments reveal that improves accuracy diagnosis.