第30卷
第3期
2010年
9月
核科学与工程
Chinese Journal of Nuclear Science and Engineering
Vol.30 Sep.
基于模糊神经网络白
的直流蒸汽发生器
压力协调控制
明，张伟
夏国清，艾甲
（哈尔滨工程大学自动化学院，黑龙江哈尔滨150001）
No.3 2010
摘要：针对直流蒸汽发生器水容积小，蓄热能力小，数学模型具有不确定性和非线性，在负荷变化或受到扰动的动态过程中出口蒸汽的压力难于控制的特点，引人了协调控制思想，将蒸汽发生器出口蒸汽流量与给水流量引人到控制系统，将模翻神经网络应用到直流蒸汽发生器出口压力的控制中，在线调警神经网络的权值，在此基础上设计了基于模糊神经网络的直流蒸汽发生器压力协调控制器。仿真结果表明，
该控制较PID控制具有更小的超调量和调整时间，对实际控制系统设计具有一定的参考价值。关键词：直流蒸汽发生器；模糊控制；神经网络
中图分类号：TL61.3
文献标志码：A
文章编号：0258-0918(2010)03-0266-06
Pressurecoordinatedcontrolbased onfuzzyneuralnetwork
of once-throughsteamgenerator XIA Guo-qing,AI Ming,ZHANG Wei(College of Automation , Harbin Engineering University
Harbin of Heilongjiang Prov, 150001, China)
Abstract: According to the characteristics of the small water volume and the wake capa bility of saving heat, uncertain and non-linear of the mathematics model, and it is hard to control of the exit steam pressure when load changed or disturbed, coordinated con trol ideology was introduced, exit steam flow and feedwater flow of once-through steamgenerator（OTSG）wasworkedinthecontrolsystem，fuzzyneuralnetwork(FNN) based on standards was applied to pressure control system of OTSG, joint nu-meral of FNN was adjusted on line. Pressure coordinated control based on FNN of OTSG was designed. Simulation results show that the controller has smaller overshoot and shorter steady time. This effort could provide some guidance for designing control system of OTSG.
收稿日期：2009-09-21；修图日期：2009-12-03
基金项目：黑龙江省博士后基金资助项目（AUGA41100088)：哈尔滨工程大学基础研究基金资助项目（HEUFT08004）作者简介：夏国清(1962一)，男，黑龙江哈尔滨人，数授，博土生导师，主要从事核动力装置建模与仿真及控制技术研究 266