Prediction of coal mine gas emission based on hybrid machine learning model

Coal mine gas accident is one of the most serious threats in process safe coal mining, making it important to accurately predict emission. To improve accuracy emission prediction, a hybrid machine learning prediction model combining random forest (RF) algorithm, improved gray wolf optimizer (IGWO) algorithm and support vector regression (SVR) proposed, effect validated by using actual measured data from mine. Firstly, RF used screen 13 influencing factors emission, finally 6 are selected as input variables model; Secondly, GWO nonlinear convergence factor DLH search strategy obtain IGWO algorithm; Finally, optimize parameters SVR RF-IGWO-SVR established. The results show that mean absolute percentage error, error root square RF-GWO-SVR 1.55%, 0.0759, 0.1103, respectively, this result better than other comparative models, which indicates can effectively provide new for prediction.