Research On Power Control Of Wave Energy Generation System Based On Neural Network

As technology,economy as well as demand from human society growing dramatically,problems brought by energy shortages become increasingly serious.Therefore,people have turned their attention to the development of new energy sources.Wave energy as a new energy has gained the attention of many scientific researchers.The existing wave energy power generation technology has low energy conversion efficiency,complicated structure,and high cost,especially the design of the intermediate conversion device is particularly complicated,which not only causes great energy loss,but also increases the cost of the cost.The direct-drive wave power generation system reduces complex intermediate components and directly converts wave energy into electrical energy through linear motion,which improves the conversion efficiency and avoids many disadvantages of the current wave power generation technologies.The controlled object of the power generation system studied in this paper is the asymmetric bilinear linear switched reluctance generation.Its simple structure,low cost and high stability make it able to better adapt to the complex power generation environment in the ocean.This article starts with the systematic controlled object ABLSRG,introduces its mechanical structure and characteristics,analyzes its electromagnetic characteristics with ANSYS Maxwell finite element simulation results,and analyzes it through mathematical modeling(dynamic equations,electromagnetic equations,mathematical models,etc.)power generation process,power generation principle and energy conversion mechanism are explained.Based on the dSPACE platform and the MATLAB/Simulink simulation environment,the ABLSRG generation current control system was developed.Furthermore the output current from the power generation system in response to changes system in parameters and generation movement was tested.Next,as the disadvantage for PID control algorithm is not flexible enough to adjust its own parameters,the PID control algorithm is optimized by selfadaptation,self-learning and self-organization by neural network control,and a BP neural network PID(BPNN-PID)control algorithm is designed.The ABLSRG power generation current control system based on BPNN-PID control was designed and multiple sets of tracking control experiments were conducted.Through the analysis of experimental data,it is proved that after the neural network is utilized,the power generation system has stronger adaptability when dealing with parameters and environmental changes.Finally,based on the ABLSRG generation current control system,voltage control method is introduced,then the stability and anti-interference of the voltage control system are tested.The ABLSRG power generation control system was developed in combination with the BPNN-PID principle and the double closed-loop control theory.Its power generation,and carried out a number of tracking control experiments to prove that the power generation system has better stability and steady-state performance in dealing with changes in environmental factors.This project aims to lay a solid foundation for the subsequent large-scale wave energy cluster network power generation for the wave energy power system developed by a single ABLSRG.The introduction of control concept of neural network will also be used to develop the big data service platform and network intelligence for wave power generation.