矿井煤与瓦斯突出强度机器学习预测方法探讨

Discussion on Prediction Method of Coal and Gas Outburst Intensity based on Machine Learning

本文通过对高瓦斯突出矿井利民煤矿“四位一体”防突资料的数据挖掘,研究选用了机器学习方法中MARS算法,构建了该矿井突出强度回归预测模型。只要取得并输入待预测地点对应的矿井最相关的突出预兆信息,就可以随时预测计算出对应的突出强度,达到防突措施程度“精准施策”目的,以实现最佳的安全经济效果。

In this paper,through the data mining of"four in one"outburst prevention data in Limin Coal Mine of high gas outburst mine,MARS algorithm in machine learning method is selected to construct the regression prediction model of outburst intensity in the mine.As long as the most relevant outburst precursor information of the mine corresponding to the predicted location is obtained and input,the corresponding outburst intensity can be predicted and calculated at any time,so as to achieve the goal of"precise strategy"of outburst prevention measures,so as to achieve the best safety and economic effect.