电锅炉温度控制系统PID参数整定算法的改进  

The improvement of PID parameters-tuning algorithm applicable to an electric boiler temperature control system

作　　者：赵为光[1] 杨莹[1] 

机构地区：[1]黑龙江科技学院电气与信息工程学院,哈尔滨150027

出　　处：《工业仪表与自动化装置》2009年第1期85-87,100,共4页Industrial Instrumentation & Automation

基　　金：黑龙江科技学院科研启动基金资助项目(07-15)

摘　　要：针对电锅炉温度控制系统常规PID控制性能欠佳的问题,提出一种PID参数整定的改进算法。先用RBF网络在线辨识温度采样信息,得到控制器所需的Jacobian信息,然后用LM算法代替传统的梯度法整定PID参数,得到改进的PID算法。通过电锅炉温度控制系统仿真分析,验证了算法的快速性、抗扰性、鲁棒性,具有较好的应用前景。In view of the fact that the performance of any conventional PID control can＇t meet the requiremnt an electric boiler temperature control system, this paper puts forward a kind of improved algorithm for tuning the PID parameters. The RBF network is used to perform the on - line identification of the mathmatie model of a controlled plant, which can provide the Jacobian information for the System controller. The Levenberg - Marquardt algorithm is used to tune the parameters of the PID controller instead of the conventional gradient descent tuning method. The system simulation has proved that the improved algorithm is fast, anti -jamming and robust.

关 键 词：温度 RBF神经网络 PID控制 Levenberg—Marquardt算法 

分 类 号：TP272[自动化与计算机技术—检测技术与自动化装置]