论文部分内容阅读
为对矿井涌水量进行准确预测,结合矿山地下水特征,分别选取采空区累积面积、降雨量、充水通道3个影响因素作为研究对象,利用20组矿井涌水量数据作为随机森林预测模型的训练数据集,进行模型的学习训练,另用3组边矿井涌水量数据作为预测模型的测试数据,通过训练好的矿井涌水量预测模型进行测试。
In order to accurately predict the mine water inflow and mine groundwater characteristics, three influencing factors of cumulative area, rainfall and waterflooding channel are selected as the research objects respectively. The data of 20 groups of mine water inflows are used to train the random forest prediction model Dataset, model learning and training. In addition, three groups of side gushing water inflow data were used as the test data of the prediction model and tested by well-trained mine gushing water forecasting model.