Construction and application of mine water inflow prediction model based on multi-factor weighted regression: Wulunshan Coal Mine case

Accurate prediction of the amount of water inflows in mines is of great significance for safe production in mining. To improve the accuracy of the prediction, based on the analysis of the hydrological and geological conditions of the mine, the main factors affecting the water inflows in mines were determined. Using entropy method, the weight values of the factors affecting the water inflows in mines were calculated, and the non-linear regression fitting between the water inflows and various factors was carried out using multiple regression theory and MATLAB function programming. Combining with the factor weights determined by the entropy method, a weighted non-linear regression prediction model for water inflows in mines was established. The model not only takes into account the fact that the water inflows in mines are affected by multiple factors, but also reflects the characteristic that the importance of factors is different. By comparing with the multiple linear regression prediction model and the measured water inflows, it is proved that the weighted non-linear regression prediction model for water inflows in mines can overcome the defects of existing methods, minimize the prediction error caused by low degree of hydrological and geological exploration, and improve the prediction accuracy.