论文部分内容阅读
矿井瓦斯涌出量受众多因素的影响。经研究表明,煤层埋藏深度、煤层厚度、煤层瓦斯含量、煤层间距、日进度及日产量是影响瓦斯涌出的主要因素。利用多元线性回归和BP神经网络理论,分别对矿井瓦斯涌出量进行了预测,最后建立了多元线性回归与BP神经网络的组合预测模型。该模型兼顾了多元回归分析的非线性特性和神经网络的时序特性,通过具体的实例研究,对比了各种方法的预测结果。结果显示,组合预测的结果与实际有较高的拟合度,可靠性高。
Gas emission from mines is affected by many factors. The research shows that the main factors affecting gas emission are the depth of coal seam burial, coal seam thickness, coal seam gas content, coal seam spacing, daily progress and daily output. Using multivariate linear regression and BP neural network theory, the mine gas emission was predicted respectively. Finally, a combined prediction model of multiple linear regression and BP neural network was established. The model takes into account the nonlinear characteristics of multiple regression analysis and the timing characteristics of neural networks. Through the specific case study, the prediction results of various methods are compared. The results show that the result of combination forecasting has a higher fitting degree with reality and has higher reliability.