神经网络在锅炉温度控制系统中的应用  

Application of Neural Network in Boiler Temperature Control System

作　　者：李烨[1] 屈宝存[1] 冯苹苹[1] 

机构地区：[1]辽宁石油化工大学信息与控制工程学院,辽宁抚顺113001

出　　处：《自动化仪表》2013年第10期55-59,共5页Process Automation Instrumentation

摘　　要：针对国内电厂锅炉温度控制系统存在大惯性、大滞后等特性,采用RBF神经网络与有源自回归模型(RBF-ARX)相结合的方法对其非线性系统进行优化。利用RBF-ARX模型对锅炉的非线性温度系统进行建模,并且与传统PID控制和基于径向基函数神经网络的PID控制进行比较。Matlab仿真试验表明,该方法能够使系统输出在较短时间内达到平稳状态,使系统的控制性能得到了显著提高。Due to boiler temperature control system in power plant features large inertia and time lag, by combining radial basis function { RBF } neural network and active auto-regressive eXogeneous RBF-ARX model, the optimization of the nonlinear system is researched. The nonlinear temperature system of boiler is modeling by using RBF-ARX model, and the comparison of traditional PID control and the PID control based on RBF neural network is conducted. The experimental result of Matlab simulation verifies that this method makes the system output reaches stable status in a short time, and the control performance of system is greatly improved.

关 键 词：温度控制 非线性系统 RBF神经网络 有源自回归模型 MATLAB仿真 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]