Study On The Maximum Power Point Tracking Strategy And Energy Loss Of Photovoltaic Grid-connected System

The photovoltaic arrays will be under partial shading because of the shadow of clouds, buildings and trees. The P-U output characteristics of photovoltaic arrays are multiple peaks when photovoltaic arrays are under partial shading. It is great significant to study the multiple peaks MPPT control strategies when photovoltaic arrays are under partial shading, because the conventional MPPT control strategies will fall into local maximum power point and loss power.The MPPT control process cause energy losses in photovoltaic grid-connected system, mainly including the oscillation near the maximum power point and the misjudgment phenomenon due to the changes of the environment. The study of energy losses will make easy for us to understand the detail of the maximum power point tracking process and helpful to improve the efficiency of MPPT strategies and photovoltaic grid-connected system.This thesis introduced the principle of photovoltaic power generation firstly. The physical model and engineering model of photovoltaic cells were analyzed. The mathematical model of photovoltaic arrays under uniform illumination and partial shading were established. The simulation model of photovoltaic arrays was builded in Matlab/Simulink environment and the effects of light intensity and temperature on the output characteristic of photovoltaic arraya were studied. The results show the P-U output characteristics of photovoltaic arrays are multiple peaks under partial shading. It is theoretical basis for the follow-up study.Some conventional MPPT methods were analyzed when the photovoltaic arrays were under uniform illumination. On the basis of that, the MPPT method based on fuzzy control was focused on. The fuzzy controller was designed and simulated to validate in Matlab/Simulink environment. The global maximum power point tracking methods were studied under partial shading based on the conventional MPPT methods. Because the particle swarm algorithm is suitable for multimodal optimization, it was applied for the global maximum power point tracking in this thesis. In order to solve the question that the basic particle swarm algorithm slowing convergence in the later stage, this thesis improved the particle swarm optimization algorithm for the global maximum power point tracking. The results of simulation show that the improved algorithm has the advantages of rapidity and accuracy.The energy loss of the maximum power point tracking process was analyzed in this paper, focusing on the misclassification and oscillation near the maximum power point on the perturbation and observation MPPT method and incremental conductance MPPT method.Finally, the experiment of perturbation and observation MPPT method under uniform illumination and the output characteristics of photovoltaic arrays under partial shading were studied on the 10 k W single-stage photovoltaic grid-connected system experimental hardware platform. The experiments of global maximum power point tracking strategy based on global scanning were studied and the experimental results show the validity of the tracking strategy.