基于可分裂组织型P系统的光伏阵列故障分类方法

Abstract To fully leverage the output characteristics of photovoltaic arrays for diagnosing photovoltaic array fault types, this paper introduces a photovoltaic array fault classification approach based on divisible tissue-like P systems. Firstly, a simulation model of the photovoltaic array is established, voltage and current at the maximum power voltage, sum of the power values at all peak points in P-V curves and the fill factor are extracted as features. These features are then compared against the constructed feature library, and the features are used as inputs of the photovoltaic array fault classification method after eliminating the influence of environmental factors. Secondly, an adaptive density based spatial clustering of applications with noise algorithm is proposed to cluster the samples. Then, the fault classification models based the divisible tissue-like P system are constructed on the basis of the clustering results, and a reasoning algorithm is executed. Finally, the fault types of photovoltaic arrays are determined based on the probability vectors yielded by the reasoning algorithm. A comparison was conducted between the proposed method and several other machine learning algorithms, the results demonstrated that the proposed method achieved the highest accuracy, which prove that the proposed method is effective.