欢迎访问《太阳能学报》官方网站,今天是 分享到:
ISSN 0254-0096 CN 11-2082/K

太阳能学报 ›› 2022, Vol. 43 ›› Issue (10): 364-370.DOI: 10.19912/j.0254-0096.tynxb.2021-0436

• • 上一篇    下一篇

基于Rife法的波浪发电RBF神经网络功率优化控制

林汇金, 杨俊华, 邱孟, 谢子森, 黄纬邦   

  1. 广东工业大学自动化学院,广州 510006
  • 收稿日期:2021-04-21 出版日期:2022-10-28 发布日期:2023-04-28
  • 通讯作者: 杨俊华(1965——),男,博士、教授,主要从事新能源发电及其电机控制方面的研究。yly93@gdut.edu.cn
  • 基金资助:
    国家自然科学基金(51370265); 广东省科技计划(2016B090912006); 广东省自然科学基金(2018A030313010)

OPTIMAL POWER CONTROL OF RBF NEURAL NETWORK STRATEGY FOR WAVE POWER GENERATION SYSTEM BASED ON RIFE ALGORITHM

Lin Huijin, Yang Junhua, Qiu Meng, Xie Zisen, Huang Weibang   

  1. School of Automation, Guangdong University of Technology, Guangzhou 510006, China
  • Received:2021-04-21 Online:2022-10-28 Published:2023-04-28

摘要: 为提高不规则海况下直驱式波浪发电系统的功率捕获能力,通过建立水动力模型及直线电机模型,分析等效模拟电路共振状态,构造最优功率捕获条件;针对波浪工作环境强噪声干扰问题,采用Rife法获取波浪主导激励分量,设置等效阻尼系数,计算q轴最优期望电流跟踪值;通过设计径向基函数(RBF)神经网络PI控制,提高等效控制电流信号的跟踪精度,改善系统动态性能。仿真结果表明,所提策略对含强噪声干扰的波浪信号主频段幅值和频率预估精度高,跟踪期望电流效果好,鲁棒性强,系统输出功率优化效果显著。

关键词: 波浪能, 最大功率点跟踪, 永磁直线同步电机, Rife法, RBF神经网络

Abstract: In order to improve the power capture efficiency of the direct-drive wave power generation system under irregular sea conditions, optimal power capture condition was constructed by establishing the hydrodynamic model and linear motor model as well as analyzing the resonance state of the analog equivalent circuit. Aiming at the problem of strong noise interference in the wave environment, the Rife algorithm was used to obtain the dominant wave excitation component, which contributes to set the equivalent damping coefficient and calculates the q-axis optimal desired current tracking value. With designing the RBF neural network PI control algorithm,both the tracking accuracy of the equivalent control current signal and the dynamic performance of the system are improved. The simulation results show that the proposed strategy has high prediction accuracy for the amplitude and frequency of the main frequency band wave signals with strong noise, good tracking effect on the expected current value, strong robustness and significant optimization effect of the output power.

中图分类号: