煤矿瓦斯浓度预测的ANFIS方法研究  

Forecasting Coalmine Gas Concentration Based on Adaptive Neuro-Fuzzy Inference System

作　　者：张剑英[1] 程健 侯玉华[1] 白静宜[1] 裴小斐[1] 

机构地区：[1]中国矿业大学信息与电气工程学院,江苏徐州221116

出　　处：《中国矿业大学学报》2007年第4期494-498,共5页Journal of China University of Mining & Technology

基　　金：国家自然科学基金项目(70533050)

摘　　要：将时间序列分析方法与自适应神经模糊推理系统(ANFIS)结合,构建煤矿瓦斯浓度的预测模型.根据Takens理论,重构煤矿瓦斯浓度相空间,分别采用互信息法确定相空间时延和假近邻法确定相空间维数;然后在重构相空间中,运用自适应神经模糊推理系统构建煤矿瓦斯浓度的预测模型,并应用混合学习算法整定模型参数.结果表明,得到的模型训练和检验均方根误差分别为0.021 4和0.021 6,充分体现了ANFIS具有显著的学习能力和良好的泛化能力,同时也表明该预测模型是切实可行的.Forecasting model of coalmine gas concentration was built using time series and adaptive neuro-fuzzy inference system （ANFIS）. The gas concentration phase space was reconstructed according to Takens theory, and time delay and embedding dimension were determined by mutual information method and false nearest neighbor method, respectively. Then, the forecasting model of gas concentration was constructed via ANFIS in the reconstruction phase space, and the parameters of ANFIS were tuned by hybrid learning algorithm. The results show that the training and checking root mean squared error are 0. 021 4 and 0. 021 6, respectively, which indicates that the ANFIS has better learning ability and generalization performance, and the model is feasible.

关 键 词：时间序列 相空间重构 自适应神经模糊推理系统(ANFIS) 瓦斯浓度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TD712[自动化与计算机技术—计算机科学与技术]