基于SVM的煤与瓦斯突出预测模型及应用

Prediction model and application of coal and gas outburst based on SVM

为有效预测矿井内煤与瓦斯突出的危险程度,对其影响因素做了分析与探讨,分别构建了基于粒子群优化算法以及遗传算法支持向量机的煤与瓦斯突出预测模型,并且通过实例对两种模型预测的准确性进行了验证。分别利用单项以及综合指标、BP神经网络以及PSO-SVM模型、GA-SVM模型,对寺河煤矿二号井的突出区域进行预测比较。结果表明,PSO-SVM的预测模型不仅可以在小样本数据中预测出煤与瓦斯突出程度的大小,而且综合预测结果更加精确,其在解决矿井内煤与瓦斯突出的小样本数据中显示出更加强大、通用的性能。

In order to effectively predict the risk degree of coal and gas outburst in the mine,the influencing factors were analyzed,and the prediction models of coal and gas outburst based on particle swarm optimization and genetic algorithm support vector machine were constructed respectively,and the accuracy of the two models was verified by the example.By using the single and comprehensive indexes,BP neural network,PSO-SVM model and GA-SVM model,the outburst area of No.2 mine in Sihe coal mine was predicted and compared.The results show that PSO-SVM model can not only predict the degree of coal and gas outburst in the small sample data,but also the comprehensive prediction results are more accurate,which has certain advantages in the small sample data.