Research On MPPT And Island Detection Technology In Photovoltaic Grid-connected Power Generation System

In order to improve the power conversion efficiency of photovoltaic power generation systems and ensure the reliable operation of the power grid,this paper had made analyses and simulation verifications of two key technologies in the photovoltaic power generation system:the technologies of MPPT and islanding detection.On the basis of the physical parameters and working principles of the photovoltaic cells,its engineering mathematical model had been made.And then,the effect of local shadow on photovoltaic cells was analyzed.The simulation models about the output characteristics on the two cases above-mentioned lay the foundation of the following researches.Based on a detailed analysis of the principles of MPPT control technology,and a comparison between several conmmon uesd methods was made to find their own disadvantages.In view of the shortcomings of traditional methods,an improved conductance increment method was proposed.It could meet the requirements of the tracking speed and high precision.Then,the former circuit,DC/DC converter,was analyzed to made simulation models on the control technologies.By comparing the waveforms,the advantage of the method proposed was vertified.Later,a analysis of the common control algorithms of several local shadows was made to compare with each other to find open questions..This peper propoesd a method of intermittent global scanning based on variable step length perturbation method.It can accurately track the maximum power point under local shadow,and effectively overcome the shortcomings of traditional algorithm converging to local maximum power point through simulation.Blind assessment was important to the application of islanding detection technology.In this paper,the mechanism and harm of the island effect was analyzed in detail,and the testing standards was also introduced.Some common methods of island detection were studied,By analyzing the no-detection area of the island detection,the coordinate system,Q_{f 0}×C_norm,was applied to evaluate the performance.The working mechanism and blind area distribution of the traditional AFD algorithm and AFDPF algorithm was analyzed.There were still some shortcomings existing in the two algorithms mentioned above.This paper proposed an improved algorithm based on AFDPF.The new algorithm shortened the time of islanding detection,and greatly improved the reliability of island.Finally,the analysis of the grid-connected control technology used in this paper was made and its simulation model was set up.On the basis of grid-connected model,the S-function was uesd to set up its own island detection model.The simulation proves the correctness of the theoretical analysis of the improved algorithm.