基于自适应遗传算法的过热汽温PID参数优化控制仿真研究  

SIMULATING STUDY OF ADAPTIVE GA-BASED PID PARAMETER OPTIMIZATION FOR THE CONTROL OF SUPERHEATED STEAM TEMPERATURE

作　　者：李萌[1] 沈炯[1] 

机构地区：[1]东南大学动力系,江苏南京210096

出　　处：《中国电机工程学报》2002年第8期145-149,共5页Proceedings of the CSEE

基　　金：教育部高等学校骨干教师资助项目;江苏省"333人才工程"基金项目

摘　　要：该文从控制的角度出发对遗传算法进行分析研究,指出加强算法和优化对象的联系是提高遗传算法效率的根本途径。并从算法和对象的外部联系以及内部联系这两个不同的角度分别提出了遗传算法的适配值函数的一般构造原则和遗传算法的自适应机制,并给出了算法过程中各个环节新的实现策略。与以往经典遗传算法相比,采用文中提出的策略的遗传算法具有更高的求解精度、更强的全局搜索能力。并在此基础上将其运用于电厂锅炉过热气温的PID参数优化控制,仿真结果表明了其优越性,具有一定的价值。并且文中所提出的算法及策略具有一般性,很容易运用到其他优化问题中去。From the point of view of control,this paper presents that enhancing the relationship between algorithm and object is the basic approach to improve the efficiency of GA. And the general rule of the selection of GAs fitness function and the adaptive mechanism of GA are put forward from the external and internal relationship of algorithm and objectrespectively. Then new methods to achieve the steps in GA are given. Compared with the classic GA, this new GA has higher operational precision and global search ability. This new GA is applied to PID controller parameter optimization on powerstation superheated steam temperature. The result of simulatingindicates the high ascendance and value. Moreover thealgorithm and strategy discussed in this paper has general sense and can be applied to many other problems.

关 键 词：自适应遗传算法 PID参数 优化控制 仿真 锅炉 过热器 汽温调节 

分 类 号：TK223.73[动力工程及工程热物理—动力机械及工程] TP391.9[自动化与计算机技术—计算机应用技术]