Research On Photovoltaic System Control Based On Improved MPPT Algorithm

With the development of cities and the improvement of human living standards,the long-term goal of promoting the construction of green and low-carbon cities is put on the agenda.People begin to seek alternative energy sources such as wind energy and solar energy to reduce environmental pollution and the use of fossil energy.Solar energy,as the primary choice,has gradually received attention in the field of power generation.However,in practical applications,the power generation efficiency of photovoltaic systems is often affected by environmental factors.The most typical problem is the "multi peak" phenomenon,that is,the output P-U characteristic curve of photovoltaic cells will show multiple peaks instead of the conventional single peak under the effect of local shadow.At this time,the traditional maximum power point tracking(MPPT)algorithm is no longer applicable.Therefore,it is of great significance to find algorithm suitable for local shadow occlusion and apply it to photovoltaic MPPT control.First of all,this thesis analyzes the working principle and equivalent circuit structure of photovoltaic cells,and studies its output characteristics by building corresponding engineering simulation models.At the same time,aiming at the special situation of local shading,the simulation model of photovoltaic array is established,and the output characteristics of photovoltaic array in this state are analyzed emphatically.Three traditional MPPT algorithms(disturbance observation method,constant voltage method and conductance increment method)are used to simulate and analyze the photovoltaic array by setting up two experimental groups of uniform illumination and local shading.The results show that all three algorithms can track the maximum power point under uniform illumination,but will lose the ability of MPPT under local shading.Therefore,considering that the traditional MPPT algorithm can not solve the "multi peak" problem,this thesis introduces the idea of swarm intelligence search into photovoltaic MPPT control,and selects harmony search(HS)algorithm as the MPPT algorithm studied in this thesis.In order to solve the influence of HS parameter setting on convergence and global search ability,a double reverse learning strategy and an adaptive spiral search strategy are introduced in its parameter memory bank generation and tone updating,and an improved tent self-adaptive opposition based harmony search algorithm(TSOHS)is proposed.Through four sets of benchmark functions to test the algorithm and apply it to photovoltaic MPPT,it is found that TSOHS algorithm has obvious advantages over particle swarm optimization algorithm,whale optimization algorithm and original HS algorithm in optimization ability and convergence performance.Finally,an experimental platform of photovoltaic energy storage system based on STM32F103C6T6 was built,and the improved MPPT algorithm was tested and proved to be effective in improving the output power of the photovoltaic system under natural light.