Research On Islanding Detection For Photovoltaic Grid-Connected Inverter Based On Active Frequency Drift Method

From the industrial revolution to the present,due to the traditional fossil fuels are becoming less and less and the environmental problems are increasingly serious dual problems,human beings have increased the use of wind,solar and other clean energy.The application and development of distributed generation(DG)based on renewable energy sources has attracted wide attention all over the world.Islanding refers to the phenomenon when the mains failure or set off for maintenance in case of power failure,the distributed power generation system without disconnecting the response to the main power grid,power supply continued to form an independent condition of the local load.The occurrence of island hidden to the public distribution network,power generation equipment and personnel and other aspects of the formation of security risks,so the research of reliability,timeliness and practicability of anti islanding detection method is very important.First of all,respectively,respectively,on the island effect and the corresponding means of dealing with the specific background,status and algorithm optimization of the actual value,and the islanding effect mechanism of introduction on the islanding detection effectiveness evaluation standards at home and abroad,comparing the advantages and disadvantages of the passive evaluation is widely applied at present and two kinds of local active islanding detection method.The common islanding detection algorithm based on coordinate system is analyzed in theory,simulation and performance evaluation.The method of finding the boundary curve of Zone(Non-Detection,NDZ)is given.In view of the inherent contradiction between the detection speed,the success rate of detection,the detection blind area and the harmonic distortion rate,the parameter selection is difficult to be unified.This paper puts forward an intelligent model to avoid the blind islanding detection method(IAFDPF),based on the load phase identification on the directional variation of the initial factor and follow the direction of the chopper,the real-time frequency distribution of blind spot detection and adjustment.Finally,a simple photovoltaic grid connected inverter system is built by MATLAB/SIMULINK software,and the fault mode is set by the BREAK switch.The improved intelligent island detection algorithm(IAFDPF)and AFD and AFDPF detection algorithm are compared.The simulation results show that the intelligent algorithm(IAFDPF)can greatly reduce the non detection zone,and can realize the quality factor of blind spot detection conditions,and according to the load characteristics of intelligent change blind distribution,improve the success rate of islanding detection can improve the detection rate,while reducing the harmonic distortion rate.The coordination of the detection speed,the success rate of detection,the harmonic distortion rate and the detection blind area is realized,which provides the reference value for the design and improvement of the islanding detection system.