Discrimination Method for Water Inrush Source of Mine Based on Rough Sets Theory and BP Neural Network

Based on rough set theory and BP neural network theory, discriminating method for water inrush source of mine was studied. The chemical indicators of inrush water were chosen to constitute a water sample matrix. First, the rough sets theory was applied to sample information reduction, then the BP neural network was applied to water source discrimination. Discriminating model for water source was established based on rough set theory and BP neural network, and compared with the traditional BP neural network model. Taked Panyi mine in Huainan for example, the results are compared with those of BP neural network model and show that the discriminating method based on rough set theory and BP neural network theory had a higher discrimination accuracy (92.5%)than the BP neural network method(82.5%). The principle of the method was clear and applied easily.