多机励磁电力系统自适应输出反馈量化控制  

Adaptive Output Feedback Quantization Control of Multi-machine Excitation Power System

作　　者：王顺江 欧秀容 李志伟 辛红伟[2] 祝国强 WANG Shun-jiang;OU Xiu-rong;LI Zhi-wei;XIN Hong-wei;ZHU Guo-qiang(State Grid Liaoning Electric Power Supply Co.,Ltd.,Shenyang 116011,China;School of Automation Engineering,Northeast Electric Power University,Jilin 132012,China)

机构地区：[1]国网辽宁省电力有限公司,辽宁沈阳116011 [2]东北电力大学自动化工程学院,吉林吉林132012

出　　处：《控制工程》2022年第5期781-788,共8页Control Engineering of China

基　　金：国家电网辽宁省电力有限公司科技项目(SGLNDK00DWJS2000249)。

摘　　要：针对一类具有传输时延的多机励磁系统,提出了一种基于K滤波观测器的输出反馈自适应量化控制方案。通过构造K滤波观测器来估计系统中的未测量状态,并对神经网络逼近误差以及引入量化器所带来的非线性和外部干扰进行补偿。此外,采用时滞函数逼近器来消除励磁系统间未知的传输时延的影响。并且通过使用初始化技术,实现了任意小的跟踪误差。最后,通过一组半实物实验结果验证了所提控制方案的有效性。For a class of multi-machine excitation system with transmission delays, an output feedback adaptive quantized control protocol based on K-filter observer is proposed. Firstly, a class of K-filter observers is constructed to estimate the unmeasured state in the system, and compensate the estimated error of the neural network approximator, the nonlinearity introduced by the quantizer and external disturbances. Secondly,time-delay function approximator is used to deal with the influence of transmission delay between excitation systems. Thirdly, the arbitrary small tracking error is achieved by using the initialization technique. Finally, a group of hardware-in-loop experimental results show the effectiveness of the proposed control protocol.

关 键 词：多机励磁系统 动态面控制 K滤波观测器 回滞量化器 神经网络 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]