基于极值搜索算法的过热汽温控制系统优化方法  

Optimization of superheated steam temperature control system using extremum seeking algorithm

作　　者：费文哲[1] 李益国[1] 沈炯[1] 相晓鹏 

机构地区：[1]东南大学能源与环境学院,南京210096 [2]山西省电力公司运城供电分公司,运城044000

出　　处：《东南大学学报（自然科学版）》2010年第5期952-956,共5页Journal of Southeast University：Natural Science Edition

基　　金：教育部科学技术研究重点资助项目(108060);教育部博士点基金资助项目(20090092110051);国家自然科学基金资助项目(51076027)

摘　　要：提出一种基于极值搜索算法的过热汽温控制系统PID参数优化策略.首先介绍了极值搜索算法的基本原理,然后基于该算法设计了过热汽温优化控制系统,最后通过大量仿真实验详细研究了极值搜索算法主要参数变化,以及采用不同目标函数,对优化结果的影响.该方法不依赖于对象的数学模型,而采用阶跃响应实验来获得系统性能指标相对于参数拢动量的梯度信号,在线调整PID控制器参数.仿真结果表明,基于该方法优化的过热汽温控制系统具有良好的动态调节品质和较强的适应性.Based on extremum seeking（ES） algorithm an optimization strategy for tuning parameters of the PID（proportional-integral-differential） controller of a superheated steam temperature control system is presented.Firstly,the basic principles of ES algorithm are introduced.Then,a superheated steam temperature optimizing control system is designed based on it by adding a new optimizing loop into the ordinary control system.Finally,the impacts of the variation of the main parameters and different objective functions on the optimization results are discussed in detail by a large number of simulation experiments.The proposed method can adjust the PID controller parameters online by use of the gradient information of the performance index to parameters perturbations,which does not rely on the mathematical model,but resorts to step response experiments.Simulation results show that the optimized superheated steam temperature control system based on this method has good dynamic properties and strong adaptability.

关 键 词：热能动力工程 过热汽温控制系统 极值搜索算法 PID控制 

分 类 号：TK323[动力工程及工程热物理—热能工程]