基于神经网络的电厂磨煤机运行状态智能分析技术  

Intelligent analysis technology of coal mill operating status in power plant based on neural network

作　　者：宿星会 张宝国[1] 吴家伟[1] 韩建刚 闫超 SU Xinghui;ZHANG Baoguo;WU Jiawei;HAN Jiangang;YAN Chao(Shiliquan Power Plant,Huadian Power International Corporation Ltd.,Zaozhuang 277103,China)

机构地区：[1]华电国际十里泉发电厂,山东枣庄277103

出　　处：《电子设计工程》2022年第1期94-98,共5页Electronic Design Engineering

基　　金：中国华电山东公司科技项目(CHDKJ18-02-39)。

摘　　要：针对火力发电厂制粉系统中,磨煤机长期处于恶劣的工作环境且相关状态监测理论发展不完备,导致实际运行中磨煤机的状态不能得到有效预测等问题,文中设计了一种基于LM_BP算法和时间序列预测理论的神经网络预测模型。通过选取合适的特征参数在Matlab中完成对应预测模型的构建,分别对正常状态下的磨出口温度模型和少煤故障状态下的磨煤机电流模型进行实验,从而得到预测结果。分析与测试结果表明,文中所建立的磨煤机运行状态预测模型的准确率可达98%以上,可以有效地为磨煤机故障与非故障状态下的运行状态分析判断提供技术支撑。In view of the coal pulverizing system in thermal power plant,the pulverizer has been in a bad working environment for a long time,and the development of related state monitoring theory is not complete,which leads to the problems that the actual operation of the pulverizer can not be effectively predicted.In this paper,a neural network prediction model based on LM_BP algorithm and time series prediction theory.By selecting appropriate characteristic parameters,the corresponding prediction model is constructed in Matlab,and the mill outlet temperature model under normal state and the current model under the condition of few coal fault are tested respectively,and the prediction results are obtained.The analysis and test results show that the accuracy of the model can reach more than 98%,which can effectively provide technical support for the analysis and judgment of the operating state of the coal mill under fault and non fault conditions.

关 键 词：中速磨煤机 神经网络 运行状态分析 时间序列预测理论 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]