基于广义回归神经网络的煤与瓦斯突出区域预测  

Study on regional forecasting of coal and gas outburst based on Generalized Regression Neural Network

作　　者：念其锋[1,2,3] 施式亮[2,3] 李润求[2,3] 

机构地区：[1]湖南科技大学计算机科学与工程学院,湖南湘潭411201 [2]中南大学资源与安全工程学院,湖南长沙410083 [3]湖南科技大学煤矿安全开采技术湖南省重点实验室,湖南湘潭411201

出　　处：《矿业工程研究》2014年第1期25-29,共5页Mineral Engineering Research

基　　金：国家自然科学基金资助项目(51274100);湖南省教育厅科研资助项目(10C0690);煤矿安全开采技术湖南省重点实验室资助项目(201002)

摘　　要：针对煤与瓦斯突出存在诸多不确定影响因素的特点,在研究煤与瓦斯突出机理和广义回归神经网络的基础上,建立了煤与瓦斯突出的广义回归神经网络预测模型,提出了广义回归神经网络中的光滑因子优化选择算法,以提高模型预测精度.通过应用实例验证,预测结果精度高,与实际情况相吻合.研究结果证明了该方法的合理性和可行性,对煤矿提高煤与瓦斯突出区域的预测能力具有较大的参考意义.According to the fact that the characteristics of coal and gas outburst exist many uncertain influencing factors, and studying the theory of the coal and gas outburst and set pair analysis, this paper built the regional forecasting model of generalized regression neural network in coal and gas outburst, and put forward the optimization selection algorithm of smoothing factor in the GRNN to raise the precision of forecasting. By proving the application examples, the results had higher precision and were fit to the actual situation. The studying results proved the rationality and feasibility of this method, and it would be more guiding significance for raising the ability of regional forecasting of coal and gas outburst.

关 键 词：煤矿 煤与瓦斯突出 区域预测 广义回归神经网络 

分 类 号：TD712.7[矿业工程—矿井通风与安全]