基于Smith预估的神经网络循环流化床锅炉床温控制  

Bed temperature control of circulating fluidized bed boiler based on Smith predictor and BP neural network

作　　者：王佳[1] 刘朝英[1] 宋雪玲[1] 

机构地区：[1]河北科技大学电气信息学院,河北石家庄050018

出　　处：《河北工业科技》2010年第6期483-485,488,共4页Hebei Journal of Industrial Science and Technology

基　　金：国家自然科学基金资助项目(60874003);石家庄市科技局资助项目(08113761A);河北省教育厅应用基础研究项目(2008441)

摘　　要：料床温度是循环流化床锅炉正常运行的一个重要指标,它影响着锅炉的燃烧效率和污染物排放速率。在燃烧过程中,它具有时变性、大惯性和大滞后的特点。针对这一问题,将BP-Smith预估控制算法应用于循环流化床锅炉床温控制,该算法基于BP整定的PID控制,提高对被控对象参数变化的自适应能力和Smith预估控制能够克服对象的大迟延特性,并对BP-Smith预估控制进行了仿真。仿真结果表明,所设计的控制系统性能均优于常规PID控制和Smith预估补偿PID控制系统。Bed temperature, which influences boiler overall efficiency and the rate of pollutants emission, is one of the most significant parameters in the operation of the systems. At the same time, it is a large inertia and lag uncertain control system. In view of this characteristic, a kind of BP-Smith controller is applied in the circulating fluidized bed boiler （CFBB）, in which the PID controller based on the BP algorithm is used to improve the self adaptability, and the Smith predictor is designed to minimize the large time delay. Simulation for bed temperature of CFBB control system is carried out by MATLAB. The results show that the performance of this control system is better than that of the conventional PID controller and Smith-PlD controller.

关 键 词：神经网络 Smith预估 循环流化床锅炉 床温控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]