煤体注水湿润半径预测的神经网络模型

合理的注水半径一直是煤体注水防尘技术中难以确定的参数。笔者基于对影响煤体注水半径因素的分析和神经网络理论的原理之上 ,设计网络模型为 3层 ,输入层为 7个节点 ,应用BP网络算法 ,建立了煤体注水湿润半径的预测模型 ,并对其参数进行了讨论。然后 ,用平顶山矿务局和水城矿务局 13个矿 19个回采工作面的统计资料对BP网络进行自适应学习 ,并取η =0 .9,α =0 .82 ,控制网络总误差E≤ 10 6。经过 2 12 34次迭代后 ,网络趋于稳定。用训练好的网络对平顶山矿务局的某矿的 3层煤的注水湿润半径进行预测 ,预测结果与实测值很接近。其误差分别为 0 .5 %,0 .6 %和 0 .7%。

Neural Network Model for Predicting the Radius of Moistened Coal Layer Instilled with Water

The rational radius of moistened coal layer has been a difficulty parameter to determine in applying this instillation dust control method. Based on the analysis of factors affecting the moistened radius and neural network theory, a network model is designed as 3 layers and 7 nodes in input layer. The model for predicting the moistened radius of coal layer instilled with water and its parameters are established and discussed. Then the BP network is learned itself by parameters of 19 faces in Pingdingshan mine administrative bureau and ShuiCheng mine administrative, in which η=0.9, α=0.82,and E≤10 -6. The network keeps stable by 21?234 calculations. Then the moistened radii of three coal layers are predicted with this model in Pingdingshan. The predicted results are almost the same as the tested values with the error of 0.5 %, 0.6 % and 0.7 % respectively.