基于灰色关联一遗传神经网络的煤与瓦斯突出预测模型

摘要

在综合分析影响煤与瓦斯突出的各种评价指标的基础上,基于人工神经网络极强的非线性逼真能力,建立了煤与瓦斯突出强度预测的遗传神经网络模型.模型采用灰色关联理论完成了评价指标的优化,并利用遗传算法对BP网络初始权值和阈值的确定进行了优化.以重庆南桐矿区砚石台矿为例,对煤与瓦斯突出强度进行了预测,结果表明,采用本模型的预测结果与矿井实际突出状况一致,模型可靠,具有一定的理论与实际意义.%Considered the impact of the various coal and gas outburst indexes, based on the powerful non-linear capacity of artificial neural network, a genetic neural network prediction model that could predict coal and gas outburst events is made. Indexes are optimized by using gray theory and BP neural network's initial weights and thresholds; and by using genetic algorithm as well. This model was used to predict coal and gas outburst in the Nantong mine area. The result shows that the prediction result of the model is reliable. So it has certain significance of the theory and practice.