Elman神经网络在动力配煤煤质预测技术中的应用  

Application of the steam coal quality predication technique base on Elman network

作　　者：董平[1] 陈彦杰[1] 王鹏[1] 

机构地区：[1]黑龙江科技学院资源与环境工程学院,黑龙江哈尔滨150027

出　　处：《煤炭学报》2007年第12期1312-1315,共4页Journal of China Coal Society

基　　金：黑龙江省科技厅攻关项目(GC03J102)

摘　　要：为了满足动力配煤数字化信息控制系统对配煤煤质在线预测的精度要求,利用E lman神经网络建立了动力配煤煤质预测模型.该预测模型综合学习了各煤质指标内部以及和配比的关系,能够根据各参配煤种的Mad,Aad,Vad和对应的配比X,直接预测出配煤产品的Mad,Aad,Vad和Qnet,ad.采用置信区间分析法和相关性分析法,对该预测模型的预测效果进行了检测.检测结果表明,该预测模型具有较高的预测精度和可信度.The predication model of the steam coal quality base on Elman network was presented to meet the requirement for the information control system of steam coal blending. The model was adopted to determine the relations between qualities of the blended coal and its components and was established for predicting blended coal＇ s moisture content, ash content, volatile content and heating value. The model was verified by confidence interval analysis and checked degree of reliability of predicated results. Test results show that the model is feasible and effective.

关 键 词：ELMAN神经网络 动力配煤 预测模型 数字化控制 

分 类 号：TQ531[化学工程—煤化学工程] TP389.1[自动化与计算机技术—计算机系统结构]