摘要
针对煤体瓦斯渗透率在预测方法上存在计算量大、样本选取量多、智能化程度低等问题,首先分析了瓦斯压力、有效应力、温度变化和抗压强度等4个影响煤体瓦斯渗透率的主要因子,然后根据煤体的力学特性建立了煤体瓦斯渗透率的PSO-SVM预测模型,应用PSO算法优化了支持向量机模型的参数,最后将该模型应用于实际工程中,并将该模型的预测结果与BP模型的预测结果进行比较,结果表明,在样本数据较少的情况下,PSO-SVM模型的预测误差较小,准度更高,能够更好的对煤体瓦斯渗透率进行预测.
To overcome the problem that large amount of calculation,more samples,low degree of intelligence and coal mine safety and production is unable to provide a quick guide of the prediction method of coal gas permeability,four main influential factors affecting coal seam gas permeability were analyzed and summarized in this study,which were gas pressure,temperature,effective stress and the compressive strength.According to the mechanical properties of coal,the support vector machine forecasting mode for coal gas permeability is established.PSO algorithm is used to optimize the parameters of the SVM model.A PSO-SVM model was used to predict coal gas permeability in the actual engineering.The prediction results show that the precision of the PSO-SVM model is superior to that based on BP neural network and is consist with the actual situation.
作者
古勇
GU Yong(Langfang Teachers University, Langfang 065000, Chin)
出处
《数学的实践与认识》
北大核心
2016年第20期149-155,共7页
Mathematics in Practice and Theory
关键词
瓦斯
渗透率
支持向量机
PSO算法
gas
permeability
support vector machine
PSO algorithm
