Design of a fractional-order PID controller for a pumped storage unit using a gravitational search algorithm based on the Cauchy and Gaussian mutation

作     者：Li, Chaoshun Zhang, Nan Lai, Xinjie Zhou, Jianzhong Xu, Yanhe 

作者机构：Huazhong Univ Sci & Technol Sch Hydropower & Informat Engn Wuhan 430074 Peoples R China 

出 版 物：《INFORMATION SCIENCES》 (信息科学)

年 卷 期：2017年第396卷

页      面：162-181页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National Natural Science Foundation of China [51479076, 51679095] Fundamental Research Funds for the Central Universities, HUST [2015QN214] 

主　　题：Pumped storage unit Pump turbine governing system Fractional order PID controller Cauchy mutation Gravitational search algorithm 

摘      要：A pumped storage unit (PSU) is more difficult to control compared to a conventional hydropower generation unit due to the frequent switching of working conditions and the S-shaped characteristics of pump turbine. The traditional proportional integral derivative (PID) controller typically cannot easily provide high quality control. To overcome these difficulties, a fractional-order PID (FOPID) controller is designed for a PSU in this study. Although the FOPID controller is more effective compared to the traditional PID controller, it is more complex to optimize the parameters of this controller for a pump turbine governing system (PTGS). Thus, a gravitational search algorithm combined with the Cauchy and Gaussian mutation, named as CGGSA, is proposed and used to optimize the FOPID controller parameters. The experimental results indicate that the CGGSA has shown excellent optimization ability compared with some popular meta-heuristics on benchmark functions. Results have also proved that the FOPID-CGGSA controller shows significant advantages over other PID-type controllers with different optimization strategies. Meanwhile the optimally designed controller has shown great potential to improve the control quality of PTGS under multiple water heads. (C) 2017 Elsevier Inc. All rights reserved.