Research on water inrush risk of mine floor based on PCA-Logistic regression model

Abstract

In order to solve the prediction problem of water inrush from coal seam floor, a prediction model of water inrush from coal seam floor based on principal component analysis and Logistic regression method was proposed. Through the analysis of the risk factors of floor water inrush, six variables, including the thickness of the aquifers, the pressure of confined water, the fall of the fault, the distance between the fault and the working face, the mining height of the coal seam, and the dip angle of the coal seam, are selected as the initial impact indicators to study the floor water inrush. First, the principal component analysis method is used to reduce the dimensions of the original index data, and the established Logistic regression model is used to analyze and predict the data. Then, five groups of sample data to be tested are used to verify the model. The results show that the comprehensive prediction accuracy of the model for water inrush samples reaches 90% at the highest, and the prediction accuracy reaches 80% when the data to be measured is used for back judgment, which shows that the prediction model has certain reliability and can be used as a new method for predicting water inrush from coal mine floor.

Keywords

• mining water hazards
• floor water inrush
• water inrush hazard
• principal component analysis method
• logistic regression model
• risk assessment