A novel fault diagnostic technique for photovoltaic systems based on cascaded forest

A variety of faults often occur during the operation of PV arrays, which may seriously affect the normal operation of the system, the machine diagnosis of the types of fault has become a hotspot in the field of photovoltaic power generation. This paper proposes a novel fault diagnostic technique for photovoltaic systems based on Cascaded Forest. Through the in-depth analysis of the output of PV arrays from a data platform of Shijiazhuang Kelin Electric Co, the input variables of the diagnosis model are obtained. Compared with other fault diagnosis methods for the PV array, the proposed method can work under a small number of tagged data and the system can be run online and real-time. Finally, the experimental results show that the fault diagnosis method for the PV array based on the cascading forests can effectively detect four types of fault for PV array such as short-circuit, open-circuit, abnormal degradation and partial shading. This method has a good value for the intelligent fault diagnosis of PV.