Research On Islanding Detection Methods For Three-phase Photovoltaic Grid-connected Inverter

With the increasing consumption of fossil fuels,photovoltaic power generation technology has developed rapidly.Photovoltaic grid-connected power generation system has also been widely used.More and more photovoltaic power generation system access to the grid.The corresponding islanding effect also followed.The islanding effect will not only cause damage to the power grid equipment but also threaten the life of the grid maintenance personnel.Can effectively detect the islanding effect is a key factor in the safe operation of PV grid-connected power generation system.Therefore,the research of islanding and islanding detection technology is of great significance.In this paper,three-phase photovoltaic grid-connected inverter is used as the research object,and the islanding detection technology of three-phase grid-connected inverter is analyzed and researched in detail.Firstly,the basic structure and working principle of the three-phase photovoltaic grid-connected inverter are introduced.The mathematical model of the DC/AC inverter bridge in dq coordinate system is deduced,and the structure diagram is drawn according to the mathematical model.The principle of PQ power control and current loop decoupling is analyzed in detail,and at the same time,the principle of software phase-locked loop is introduced.Secondly,the commonly used active islanding detection method is analyzed,and the research model of islanding effect is established.The electrical characteristics of PCC in islanding effect are analyzed in detail.The single-phase active frequency offset method is analyzed and the three-phase active frequency offset method is designed according to its characteristics,at the same time,the three-phase frequency feedback positive reactive current disturbance method is introduced,and the two methods are simulated and compared.And then for the active islanding detection method will affect the power quality problems,a passive islanding detection method based on wavelet transform and neural network is proposed.The basic principle of wavelet transform and neural network is introduced,and the extraction of eigenvector for islanding detection is analyzed in detail,and this method is simulated and verified.Finally,the experimental platform of three-phase PV grid-connected system based on TMS320F28335 is designed,and the hardware and software design of the system are described in detail.Experiments on the islanding method introduced in this paper are carried out to verify the effectiveness and feasibility of the islanding method.