扩展卡尔曼神经网络在锅炉汽包水位控制中的应用  

Application of extended Kalman neural network in boiler drum water level control

作　　者：刘晓悦 冯立强 LIU Xiaoyue;FENG Liqiang(College of Electrical Engineering,North China University of Science and Technology,Tangshan 063210,China)

机构地区：[1]华北理工大学电气工程学院,河北唐山063210

出　　处：《现代电子技术》2018年第19期96-99,共4页Modern Electronics Technique

基　　金：河北省自然科学基金项目(E2016209357)~~

摘　　要：工业锅炉正常运行的一个重要指标是汽包水位,然而实际运行中出现的虚假水位现象对锅炉的正常运行产生很大影响。当前常规的调节策略是以PID为核心的三冲量策略,但是由于PID的调控效果对被控对象的数学模型依赖度高,而汽包水位难以建立精确的模型,此方案的实际效果并不理想。基于此,引入扩展卡尔曼神经网络方法。此方法具有不依赖模型的特点,并依据此方法设计网络控制器(EK-NNC),目的是替换常规PID控制器。最后给出EK-NNC与常规PID的对比结果。The drum water level acts as an important indicator of the normal operation of industrial boiler,but its false phenomenon in actual operation greatly influences the normal operation of the boiler.The three-impulse strategy takes PID as its core,but the control effect of PID has high dependency on the mathematical model of the controlled object,and it is difficult to establish the accurate model of drum water level,so the practical control effect of this strategy is unsatisfied.On this basis,an extended Kalman neural network method is introduced,which is independent of the mathematical model.The extended Kalman neural network controller(EK-NNC)was designed to replace the conventional PID controller.The comparison results of EK-NNC and conventional PID are given.

关 键 词：扩展卡尔曼算法 人工神经网络 工业锅炉 水位控制 三冲量 实验仿真 

分 类 号：TN876-34[电子电信—信息与通信工程] TP183[自动化与计算机技术—控制理论与控制工程]