抽水蓄能机组复杂空载工况增益自适应PID控制  

Heuristic Gain-Scheduling Nonlinear PID Control of Pumped Storage Units Under Complicated No-Load Condition

作　　者：赵志高 周建中[1] 张勇传[1] 李超顺[1] 许颜贺 郑阳[1] ZHAO Zhigao;ZHOU Jianzhong;ZHANG Yongchuan;LI Chaoshun;XU Yanhe;ZHENG Yang(School of Hydropower and Information Engineering,Huazhong University of Science and Technology,Wuhan430074,Hubei Province,China)

机构地区：[1]华中科技大学水电与数字化工程学院

出　　处：《电网技术》2018年第12期3918-3925,共8页Power System Technology

基　　金：国家重点研发计划项目(2016YFC0402205,2016YFC0401910);国家自然科学基金资助项目(51679095)~~

摘　　要：针对抽水蓄能机组低水头空载工况频率易波动、常规PID控制对不同水头空载工况适应性不强及现有用于控制参数优化的抽水蓄能调节系统模型过于简化的问题,提出了非线性抽水蓄能机组精细化模型的增益自适应PID控制方法(Heuristic gain—scheduling nonlinear PID,HGS-NPID),建立了基于特征线法和改进Suter变换的抽水蓄能调节系统精细化模型,利用人工羊群算法(artificial sheep algorithm,ASA)优化该模型的控制参数,讨论了HGS-NPID控制在不同水头空载开机和空载频率扰动时的应用;仿真结果表明:“抽水蓄能机组精细化模型的HGS-NPID控制”既能充分适应调节系统水击、水泵水轮机强非线性以及各模块间耦合特性,又能改善常规PID及分数阶PID(FOPID)在不同水头尤其是低水头条件下空载工况的动态品质。Heuristic gain-scheduling nonlinear PID (HGS-NPID)is proposed for accurate nonlinear model of pumped storage hydro units(PSHU)to solve the problems of PSHU instability under complicated no-load condition with low waterheads,inadaptability of traditional PID under diverse waterheads and simplification of existing PSHU models.The characteristic method and improved Surer transformations are respectively applied,toderiveac curate PSHU model. Moreover,regulatory parameters of HGS-NPID are optimized with artificial sheep algorithm.This paper is mainly concerned with HGS-NPID application for two experiments:no-load start-up operations and no-load frequency perturbation conditions under diverse waterheads.Numerical simulations are performed to verify effectiveness of the designed controller. Results show that HGS-NPID based on accurate PSHU can fully accommodate water hammer,nonlinearity and coupling characteristics between modules and improve dynamic quality of the no-load conditions under diverse waterheads,especially low heads,compared to traditional PID and fractional order PID.

关 键 词：抽水蓄能机组 增益自适应PID 空载工况 人工羊群算法 控制优化 

分 类 号：TM721[电气工程—电力系统及自动化]