瓦斯赋存影响因素的AHPSO-BP预测与权重分析

Prediction and Weight Analysis of Influencing Factors of Gas Occurrence

为深入研究宝雨山矿瓦斯赋存规律,依据矿井构造和瓦斯赋存条件,以鲁沟背斜为界,将矿井分为东西2个瓦斯地质单元。并利用改进的粒子群算法(PSO)优化BP神经网络模型,通过优化后的模型定量化表述不同瓦斯地质单元瓦斯赋存影响因素权重大小。结果表明:改进后的AHPSO-BP神经网络预测模型与标准BP神经网络预测模型对比,具有预测精度高、收敛速度快,运行结果相对稳定等优点:顶板砂地比是影响矿井瓦斯赋存的主控因素,东、西部瓦斯地质单元中煤厚对瓦斯赋存影响差别较大,底板砂地比和褶皱复杂系数对瓦斯赋存影响较小。

In order to further study the gas occurrence law of Baoyushan mine,according to the mine structure and gas occurrence conditions,the mine is divided into east and west gas geological units based on the Lugou anticline.An improved particle swarm optimization(PSO)algorithm is used to optimize the BP neural network model,and the optimized model is used to quantify the weighting factors of gas influencing factors in different gas geological units.The results show that the improved AHPSO-BP neural network prediction model is compared with the standard BP neural network prediction model,which has the advantages of high prediction accuracy,fast convergence speed,and relatively stable operation results;the roof sand content is the main control that affects the gas occurrence in the mine.Factors,the thickness of coal in the east and west gas geological units has a large effect on gas occurrence,and floor sand content and fold complexity coefficient of gas floor have little effect on gas occurrence.