基于专家系统的回转窑窑头工况识别  

Recognition of Working Condition for Rotary Kiln Hood Based on Expert System

作　　者：王孝红[1] 房喜明[1] 于宏亮[1] 

机构地区：[1]济南大学控制科学与工程学院,山东济南250022

出　　处：《控制工程》2010年第3期309-312,共4页Control Engineering of China

基　　金：山东省科技发展计划资助项目(2007GG10004007);山东省自然科学基金资助项目(Z2006G02)

摘　　要：针对新型干法水泥生产线窑头环节非线性、多变量、强耦合、大时滞等特性导致的工况复杂,控制难度大的问题,在深入分析窑头工况特点的基础上,提出了一种基于专家系统的工况智能识别方法。该方法首先对从集散控制系统中采集的现场参数实时值进行数据预处理,运用ART-2神经网络算法对关键参数的变化趋势进行在线辨识,并输出趋势类别,同时对主要参数的实时值进行模糊化处理,输出模糊档位。基于专家规则的工况识别系统根据参数的辨识结果,通过精确推理输出当前工况。工况识别软件采用C++编写,现场实际运行表明了该方案的正确性和实用性。A scheme for recognition of working condition is proposed by using expert system based on the analysis of the rotary kiln hood characteristics.The real-time data of the parameters collected from the distributed control system is pretreated at first.The ART-2 neural network is used to recognise the change trendency of pivotal craft parameters,and the species of trendency are given out.The real-time data of the main craft parameters are fuzzified.The expert system outputs the species of current working condition by the accurate consequence according to the result of parameter recognition.The software is compiled by C＋＋.The application result shows the correctness and practicability of the proposed scheme.

关 键 词：回转窑 ART-2神经网络 模糊 专家系统 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]