Research On MPPT Of Photovoltaic Generation System Based On Improved Particle Swarm Optimization Algorithm

Maximum Power Point Tracking(MPPT)technology has been a hot research issue in the development of solar Power Generation System(PGS)technology.When the photovoltaic(PV)array is under Partially Shaded Condition(PSC),the Power-Voltage(P-U)curve of PV exhibits multiple peaks.The conventional MPPT algorithm can only work under unimodal conditions with uniform illumination.Under PSC,the conventional algorithm will fail,and can not find the Maximum Power Point(MPP)accurately.Therefore,the research of the MPPT algorithm under PSC is of great significance for improving efficiency of PGS operating.First,this paper establishes a mathematical model of PV cells and gives the changing rules of output characteristics of PV cells under different environmental parameters by experiments.This paper focuses on the equivalent simplification and modeling of PV array under PSC.For different light blocking situations,the experimental design is carried out.Analyzing the variation of the output characteristics of the PV array through simulation,support for the design of the improved MPPT algorithm for shading PV array is provided.Second,the basic principle of MPPT for PGS is analyzed.Three common MPPT algorithms under uniform illumination have been compared and analyzed:Constant Voltage method(CTV),Perturbation and Observation method(P&O)and Increment Conductance method(INC).This paper focuses on the misjudgment phenomenon caused by the improper setting of steps and the sudden change of light intensity,and an improved variable step strategy is proposed,which effectively avoids the misjudgment phenomenon when the lighting conditions change abruptly,as well as increases the variable-step INC tracking speed and tracking accuracy.Thirdly,in view of the facts that basic Particle Swarm Optimization(PSO)exists in the MPPT application,the oscillation and the particles are easy to fall into the local optimum value,the algorithm is improved and an improved Adaptive Particle Swarm Optimization(APSO)algorithm is proposed.The APSO algorithm avoids the particle falling into the local optimal value,adds the improved algorithm restarting mechanism for the global maximum power point shift when the shading situation changes,and selects the basic test function to test the improved APSO algorithm.The results show that the APSO algorithm has better convergence and tracking accuracy than the basic PSO.Finally,the flow of APSO-based MPPT algorithm is designed.The simulation model of MPPT control system of PGS is built on Matlab/Simulink simulation platform.The improved APSO algorithm is applied to MPPT control circuit of PGS,and the simulation experiments are carried out.Experimental results show that the proposed APSO algorithm has high tracking accuracy and fast convergence in MPPT applications.The output power of photovoltaic power generation system is significantly improved under the static and dynamic PSC.And the hardware experiment platform is built,verifing the effectiveness of the proposed APSO.