中文摘要：

为提高煤层瓦斯含量预测的精度和效率,提出用灰色关联分析从影响因素中筛选主要因素,结合运用GA-BP神经网络预测煤层瓦斯含量的方法。通过遗传算法（GA）优化BP神经网络的权值和阈值,解决BP神经网络易过早收敛极小值以及收敛速度慢的问题。用Matlab构建灰色关联分析-GA-BP神经网络、GA-BP神经网络和BP神经网络模型。选取成庄矿3^#煤层的含量与影响因素作为实验数据对该模型进行实验分析,比较三个的预测模型的预测结果。实验结果表明：顶板泥岩厚度、煤层厚度、基岩厚度、煤层深度是影响成庄矿3^#煤层瓦斯含量的主要因素。灰色关联分析-GA-BP神经网络预测模型平均相对误差为2.77%,比后两种预测模型的预测结果好,能准确预测煤层瓦斯含量。

英文摘要：

In order to enhance the efficiency and accuracy of prediction on the gas content in the coal seam,a method was raised to predict the gas content,which adopted the gray correlation analysis to select the main factors first, then combined BP neural network with genetic algorithm （GA）. Considering the problem of easily trapping into the partial minimum and slow convergence, the algorithm adopted GA to improve the weights and thresholds of BP neural network. Taking Matlab for writing programs, the prediction models of gas content based on gray correlation analysis-GA-BP neural network. GA-BP neural network and BP neural network were established. The gas content and influence factors in the No. 3 coal seam of Chengzhuang mine were taken as experimental data to conduct practical analysis on this model,and the prediction results of BP neural network and GA-BP neural network were compared with the result of gray correlation analysis-GA-BP neural network. The results showed that the thickness of mudstone roof, the seam thickness, the basic rock thickness and the thickness of coal seam should all be taken as the primary influential factors of gas content in the No. 3 coal seam of Chengzhuang mine,and the average relative error of gray correlation analysis-GA-BP neural network prediction model was 2. 77%, which was better than those of BP neural network and GA-BP neural network prediction model, and it can accurately predict gas content in the coal seam.