Research On Multi-peak MPPT And Grid-connected Inverter Control For Three-Phase Photovoltaic Power Generation

In recent years,with the emphasis on the natural environment,renewable energy has been continuously developed.As a clean energy source,solar energy is becoming more and more widely used.In order to achieve efficient and friendly integration of photovoltaic power generation into the power system network,and solve problems from the power supply head,it is necessary to study the grid-connected input.Taking 20KW photovoltaic grid-connected power generation system as the research object,a two-stage photovoltaic grid-connected power generation system is adopted.The pre-stage boost circuit provides a stable DC voltage to the input of the inverter.The post-stage inverter circuit is connected to the LCL filter to output the grid-connected AC power.This thesis mainly explored the maximum power tracking of solar power generation and the stability control of inverter grid connection.Photovoltaic array was built,and the power curve was obtained by simulating local shadow simulation with different irradiation intensities.It is analyzed the principle of maximum power tracking(MPPT)and classic tracking strategy.A Newton method based on particle swarm optimization is proposed to search for the maximum power point under local shadow.The method combines the particle swarm algorithm with the Newton method.The particle swarm algorithm is used to track the multi-peak maximum power curve to find the global maximum value.The output of the particle swarm algorithm is used as the initial point of the Newton method for MPPT.This method not only has the advantages of the global solution of the group algorithm,but also the rapidity of the Newton method to find the maximum value.In this thesis,the weight of the particle swarm optimization algorithm is improved,and the original inertia proportional weight is replaced by the linear differential decrement,which improves the fastness and accuracy of the particle swarm optimization algorithm.The feasibility of the method is verified by simulation.It is selected LCL filters with superior effects,calculated the size of the inductor and capacitor according to the system capacity requirements of the PV system and analyzed the stability of the system according to the actual situation,and using the virtual resistance method of the capacitor parallel resistor.The method measured the capacitance current in the form of proportional feedback and is used in the inverter control to filter,suppress spikes and improve stability.The photovoltaic power generation grid-connected system adopts two-stage inverter control,establishes a mathematical model,and carried out feedforward decoupling control on the coupling quantity after coordinate transformation.The current control is based on the adaptive PI control,and the stability control uses the capacitor current proportional control.Based on the existing traditional current PI,the PI parameters are adaptively tuned by simulated annealing,and feedforward decoupling control and capacitive current feedback control are added.This method can solve the problem of insufficient PI parameter tuning when the system illumination changes.Taking a system with a capacity of 20KW as an example,a simulation circuit is established to output a stable voltage and current in accordance with the power quality of the power network under different illumination conditions.The photovoltaic grid-connected power generation system under local shadow is simulated,and the simulation results are subjected to harmonic analysis.The results show that under different illumination conditions,the current harmonic distortion rate of the grid is far less than the system requirements,and the result reaches the LCL inverter control target.