信息熵改进免疫遗传算法在火电厂主蒸汽温度控制系统中应用  

Application of immune genetic algorithm improved by information entropy in main steam temperature control system of thermal power plants

作　　者：孙宇贞 李朵朵 张婷 李康 SUN Yuzhen;LI Duoduo;ZHANG Ting;LI Kang(College of Automation Engineering,Shanghai University of Electric Power,Shanghai 200090,China;Shanghai Engineering Research Center of Intelligent Management and Control for Power Process,Shanghai 200090,China)

机构地区：[1]上海电力学院自动化工程学院,上海200090 [2]上海发电过程智能管控工程技术研究中心,上海200090

出　　处：《热力发电》2018年第8期97-102,共6页Thermal Power Generation

基　　金：上海市科技创新行动计划高新技术领域项目(16111106300);上海市科技创新行动计划国际科技合作项目(15510722100)~~

摘　　要：为提高火电厂主蒸汽温度控制系统控制品质,一般选择非线性PID(NPID)控制器,但NPID控制器参数整定尚无系统方法。本文提出一种基于信息熵改进免疫遗传(GAIE)算法的火电厂主蒸汽温度控制方法,该算法在免疫遗传算法中加入信息熵思想,改善了免疫遗传算法易陷入局部最优、寻优慢的情况。通过采集的电厂实际运行数据,对主蒸汽温度控制系统进行辨识,然后根据所辨识的系统传递函数采用GAIE算法寻优NPID控制器进行控制。实际数据验证及仿真结果表明:GAIE算法辨识模型的输出更好地拟合了现场实际输出,比免疫遗传算法模型平均精度提高40.18%,模型参数寻优速度提高了86.67%;GAIE算法寻优NPID控制器控制效果明显优于传统PID控制器。In order to improve the control quality of main steam temperature system of thermal power plants,nonlinear PID（NPID） controllers are generally selected. However, there is no systematic method for parameter setting of the NPID controllers. The immune genetics algorithm improved by information entropy（GAIE） is proposed in this study. The GAIE algorithm adds information entropy to the immune genetic algorithm to improve the problem of easy to fall into local optimality and slow optimization of immune genetic algorithm. The collected actual operation data of the power plant is used to identify the main steam temperature system. Then the GAIE algorithm is used to optimize the NPID controller for control based on the transfer function of the identified system. The results of validation and simulation of actual data show that, the output of the GAIE algorithm identification model fits the actual output of the site better, the average precision is 40.18% higher than that of the immune genetic algorithm model, and the optimization speed of model parameters is improved by 86.67%. The control effect of the GAIE algorithm is obviously superior to the conventional PID controller.

关 键 词：主蒸汽温度控制 对象辨识 温度控制 非线性PID控制器 信息熵改进免疫遗传算法 

分 类 号：TM621[电气工程—电力系统及自动化]