回采工作面瓦斯涌出量预测的神经网络方法

Prediction of the Amount of Gas Emitted from the Face Based on Neural Network

回采工作面瓦斯涌出量受煤层瓦斯含量、工作面产量和采煤方法等各种因素的影响 ,笔者通过研究得出 :回采工作面瓦斯涌出量与煤层的赋存条件和开采条件之间是一种非线性关系 ,但目前还难以用精确的数学建模来求解。因此 ,提出了一种应用BP人工神经网络模型和算法 ,建立工作面瓦斯涌出量预测模型 ,从而预测不同开采条件下回采工作面瓦斯涌出量。实际应用表明 ,模型精度能满足要求。笔者还对隐含层神经元数目对步长影响作了讨论。

Amount of gas emitted from the face depends on the gas content of coal layer, coal yield and exploiting method. Research indicates that the amount of gas emitted from the face is non-linearly correlated with the condition of deposit and exploitation of coal seam. However, up to now it is still unable to establish a mathematical model to solve this correlation. The back propagation algorithm of the artificial neural network is applied to establish a prediction model for predicting the amount of gas emitted under different exploiting conditions. Practical application demonstrates that the precision of the model is well enough to meet the requirement. Effects of the number of units in hidden layers on training epoch are also discussed.