电工技术学报2024，Vol.39Issue(14)：4391-4404,14.DOI:10.19595/j.cnki.1000-6753.tces.230906
基于模型预测的直驱式波浪发电机机侧最优功率控制技术
Optimal Power Control Technology of Direct-Drive Wave Power Generation System Based on Model Prediction
摘要
Abstract
The direct-drive wave power generation system has the advantages of a simple structure,low cost,and high efficiency.Permanent magnet linear generators are implemented in direct-drive wave power generation systems for their high-power density,and the electromagnetic force of the linear generator is controlled to realize maximum energy capture of the system.However,the power capture strategy of the direct-drive wave power generation system greatly affects the energy obtained from the system.Predictive models with multiple constraints can solve the safety problems of wave power generation devices in extreme sea conditions,which is suitable for power capture control in direct-drive wave power generation systems.Therefore,based on Model Prediction,this paper proposes an optimal power control strategy for direct-drive wave power generation systems to improve the energy capture of generators. Firstly,the hydrodynamic mathematical model of the direct-drive wave power generation device is established,and the frequency response of the wave-front equation to an excitation force is obtained using the boundary element method.In addition,the dynamic mathematical model of the permanent magnet linear generator is analyzed.Secondly,a power control strategy based on model prediction is derived for the direct-drive wave power generation system.The power control strategy is divided into power capture and power tracking control strategies to convert wave energy into electrical energy and import it into the DC bus.This paper analyzes the optimal power capture strategy from three perspectives:control strategy,prediction interval,and cost function.The state-space equations,cost function,and constraints are analyzed.Then,the model predictive current control is implemented for dynamic power tracking.Thirdly,based on the theoretical derivation,the control block diagram of the direct-drive wave power generation system is established.The simulation model of the system is built,and simulation parameters of the wave power generation are listed.Moreover,the energy captured by the PMLG under different control strategies is compared,and the prediction interval effects on power capture under different cost functions are analyzed.The optimized prediction interval parameters are obtained through comparative analysis.Different cost functions are compared,considering the most power in the DC bus and the electromagnetic power of PMLG.All the simulation models are built.Simulation results show that the optimal power capture strategy based on model prediction effectively improves the energy capture of generators under irregular waves from wave power generation systems.In addition,maximizing the power imported into the DC bus can increase the energy imported into the DC bus by the generator,which has significant practical value.Experimental results under different driving speeds and dynamic wave condition adjustments are obtained. The following conclusions can be drawn.(1)The proposed optimal power control strategy improves the captured energy from the waves,making it suitable for direct-drive wave power generation systems.(2)The prediction interval of EMPC greatly affects the energy absorption of the system,and the cost functions significantly affect the energy feeds into the DC bus.Therefore,selecting an appropriate prediction interval and cost function is helpful for performance improvement.(3)Experimental results show that the system effectively achieves tracking control of current under different working conditions using the proposed control strategy.关键词
直驱式波浪发电/最优功率捕获/模型预测控制/功率跟踪控制Key words
Direct-drive wave power generation/optimal power capture/model predictive control/power tracking control分类
信息技术与安全科学引用本文复制引用
黄磊,魏莱,杨建龙,胡敏强..基于模型预测的直驱式波浪发电机机侧最优功率控制技术[J].电工技术学报,2024,39(14):4391-4404,14.基金项目
国家自然科学基金(41876096)和国防科技重点实验室基金(6142217190101)资助项目. （41876096）
