Fisher Discriminant Analysis Predicted Model of Criticality of Karst Water Disasters from Coal Seam Floor and its Application

Abstract

Based on Fisher discriminant theory, the Fisher discriminant analysis (FDA) model was estabilished for forecasting the criticality of karst water disasters from coal seam floor. Five factors were selected such as mining level, digging length, hydraulic pressure, thickness of water-resisting layer and fault throw as the discriminant factors of the FDA model. A series of experimental data of water inrush from coal floor were taken as the training and testing samples, then some other practical data were used to verify this model. The results show that the FDA model is a simple, feasible and high accurate prediction method. It can provide a new way for mine water prevention and control to predict water disaster.