Research On Maximum Power Point Tracking Of Photovoltaic Power Generation System

Energy is used heavily in the world because of social science and technology progress.In the meanwhile,the eco-environment has also been destroyed to a certain extent.In addition,traditional fossil energy on earth is limited and nonrenewable.A thorough solution to the problem is planned to use clean energy,solar energy is the most abundant and common on earth.Photovoltaic power generation technology is the most significant factor in the application of solar energy.To raise the maximum utilization efficiency of photovoltaic power generation,maximum power point tracking technology for photovoltaic modules plays an important role in photovoltaic power generation.This thesis takes photovoltaic cells and photovoltaic arrays as the research object,besides that maximum power output is the research goal.Firstly,this thesis sums up the research background of photovoltaic power generation with development situation.After that,the photovoltaic cell is modeled and analyzed.Under different influence factors,we can know that every single photovoltaic cell has a different output power value.In addition,the photovoltaic circuit can increase the output power by adjusting the duty cycle.Secondly,traditional MPPT algorithms are introduced and analyzed by simulating.After this,an improved MPPT algorithm combining the attention mechanism with the LSTM recurrent neural network is proposed.This method trains and analyses historical data to obtain the predicted maximum power point,and verifies whether it is the maximum power point through a judgment condition,and then performs small-step tracking.This thesis uses Matlab/Simulink to carry out simulation experiments and analysis.It is obvious that the maximum power can be tracked more quickly and accurately from the output curve.Finally,it can be concluded that every single photovoltaic module has a different output power value under different influencing factors.The photovoltaic module can be damaged because of the thermal spot effect.Therefore,it is necessary to track the global maximum power point of the photovoltaic array.In this thesis,an improved particle swarm algorithm combined with a bacterial foraging algorithm is used to optimize the group of photovoltaic arrays.PSO algorithm,BFOA algorithm,and an improved particle swarm algorithm combined with bacterial foraging algorithm are used to optimize the photovoltaic array.From the output curve,the improved algorithm can get a higher maximum power point output and the tracking speed is faster in this paper.