Research On Fault Detection Method For PV Array With Topology Reconstruction

The energy crisis and environmental pollution are two major worldwide problem, which promote the development of new energy technologies. In this context, photovoltaic technology become the focus of new energy development and utilization. With the promotion of photovoltaic power plants, the demand for PV array fault diagnosis is also increasing. Meanwhile, in order to improve the power generation efficiency of the photovoltaic power generation system, the research direction of the PV array changes from the fixed topology to the variable topology. Because of its electrical connection may vary, PV array with topology reconstructed increased the incidence of failure. PV array is working properly that has a vital role for photovoltaic power generation system. Therefore, the study of fault diagnosis method for topology reconfiguration PV array can be very practical significance.PV array fault diagnosis need to solve two problems that are PV array fault point location and the type of fault diagnosis. In this paper, the method of multi sensor detection is used to locate the fault point of PV array. The optimal placement of the sensor is determined by the weight matrix equivalent analysis, and further point out the fault location strategy. In this paper, the extreme learning machine (ELM) algorithm is introduced to the fault diagnosis of PV array. The network model of fault diagnosis of PV array is established by training. The input of the model, the hidden layer activation function and the number of hidden layer neurons are set by in-depth analysis. A small PV power generation system as an experimental platform was built to collect the sample data. The fault diagnosis model of ELM algorithm is simulated and verified, and compared with the BP neural network algorithm.Finally, this paper designs a remote monitoring system for PV array based on the Web technology. The system monitors the operation condition of PV arrays, diagnoses and locates the fault of PV array. The fault remote monitoring system is tested by the experimental platform.