基于改进麻雀搜索算法优化RBF神经网络的光伏阵列故障诊断

针对传统BP神经网络在光伏阵列故障诊断时受初始权值阈值的影响，易导致全局搜索过程陷入局部最优这一问题，提出了一种基于改进麻雀搜索算法优化RBF神经网络（ISSA-RBF）的光伏故障诊断方法。首先，利用Matlab建立光伏阵列故障仿真模型，提取出故障诊断模型的特征参数；其次，融入Levy飞行和自适应权重φ对麻雀搜索算法进行改进，用优化后的算法建立ISSA-RBF故障诊断模型；最后，与传统BP和SSA-RBF模型进行对比验证，实验结果表明，ISSA-RBF模型在故障诊断精度上达到94.8%，可以有效诊断光伏阵列的故障类型。

Fault Diagnosis of Photovoltaic Array Based on Improved Sparrow Search Algorithm Optimized RBF Neural Network

Aiming at the problem that the traditional BP neural network is affected by the initial weight threshold during the fault diagnosis of photovoltaic array， which can easily lead to the global search process falling into local optimization， the paper proposes a photovoltaic fault diagnosis method based on an improved sparrow search algorithm（ISSA） optimized radial basis function（RBF） neural network. Firstly， the photovoltaic array fault simulation model is established by Matlab， and the characteristic parameters of the fault diagnosis model are extracted. Secondly， Levy flight and adaptive weight φ are combined to improve the sparrow search algorithm（SSA）， and the ISSA-RBF fault diagnosis model is established using the optimized algorithm. Finally， compared with the traditional BP and SSA-RBF models， the results show that the ISSA-RBF model has a fault diagnosis accuracy of 94.8%， thus effectively diagnosing the fault types of the photovoltaic arrays.