基于奇异值分解的马氏距离光伏故障监测

Photovoltaic Fault Monitoring Based on Singular Value Decomposition and Mahalanobis Distance

及时检测出光伏阵列短路、断路、阴影遮挡等故障并作出相应处理,不仅有利于提高光伏阵列收益,也有利于降低光伏阵列安全风险,避免火灾等事故。基于马氏距离方法,计算出光伏阵列实测电流电压和仿真模型输出电流电压的相似程度,以此用于光伏阵列故障监测。正常条件下,光伏阵列实测电流电压值和仿真模型输出电流电压值较为接近,传统马氏距离方法在求解逆矩阵时将产生奇异解。结合奇异值分解方法对传统马氏距离方法进行了改进,试验证明,所提出的方法能够精准检测出相应的短路、断路、阴影遮挡等故障。

Detecting the short circuit,open circuit,shadow shielding and other faults of the photovoltaic array in time and dealing with them will not only help increase the revenue of the photovoltaic array,but also help reduce the safety risk of the photovoltaic array such as fires.Based on the Mahalanobis distance method,the similarity between the measured current and voltage of the photovoltaic array and the output current and voltage of the simulation model is calculated,which can be used for photovoltaic array fault monitoring.Under normal conditions,the measured current and voltage values of the photovoltaic array are relatively close to the output current and voltage values of the simulation model.The traditional Mahalanobis distance method will produce singular solutions when solving the inverse matrix formula.The singular value decomposition method is combined to improve the traditional Mahalanobis distance method.Experiments show that the method proposed can accurately detect the corresponding short circuit,open circuit,shadow occlusion and other faults.