Rerearch On Maximum Power Point Tracking Algorithm Under Partial Shading

The world gradually steps into the post-industrial society, the excessive use of non-renewable energy makes the human facing the serious problems such as environmental pollution and energy depletion. So it’s a urgent work for us to find alternative energy. Solar energy, as a clean and renewable energy, is starting to be more widely appreciated. The comprehensive promotion of solar energy is difficulty due to the high cost of solar power. The thesis focuses on the research of MPPT(Maximum Power Point Tracking) algorithm under partial shading to make PV system can always output with maximum power under different environmental conditions, improving the efficiency of solar power and reducing the unit cost of solar power.Firstly, single diode model of the photovoltaic cell is studied. NIM(Newton Iterative Method) and SMM(Simplified Model Method) is used to obtain analytical solution from the equation of PV model. The principle analysis of PV cell’s series-parallel model is taken, and the I-V and P-V curves with different shadow mode are drawn by the model simulation.Secondly, the DC-DC circuit and the basic MPPT algorithm is discussed. The DC-DC circuit provides a adjusting mechanism to realize impedance matching of PV system. The basic condition of MPPT algorithm optimization is obtained through the analysis of Buck and Boost type of DC-DC circuit. The basic MPPT algorithm has good optimization results under uniform illumination, but it is unable to get the global maximum power in multi-peak curve under partial shading, so that the efficiency of the photoelectric conversion will decreased.Thirdly, the basic PSO(Particle Swarm Optimization) algorithm is improved. The PSO is a swarm intelligence algorithm which has a good effect in the optimization of complex and nonlinear curves. The PSO algorithm is easy to fall into local extremum since it’s post optimization ability of PSO is weak. To add chaos search into the optimization process of PSO, in order to avoid excessive concentration of population and elimination of inferior particle. Lastly, the superiority of CIPSO(Improved Particle Swarm Optimization based on Chaos search) algorithm is proved through classical test functions of the numerical experiments.Finally, CIPSO-MPPT(the MPPT algorithm based on CIPSO) is applied to centralized and distributed MPPT architecture. 3 conclusions is obtained by numerical experiments and simulation of PV model:(1) CIPSO-MPPT can get the global maximum power in a multi-peak curve under partial shading.(2) In the same MPPT architecture, CIPSO-MPPT is superior to PSO-MPPT(the MPPT algorithm based on PSO) both in the convergence precision and the speed of response.(3) Comparison of central MPPT algorithm, distribute MPPT algorithm can get more power under partial shading, but a substantial increase in the complexity of the algorithm makes it spend more time to converge.