Research on the Integrated Neural Network Water Inrush Prediction System Based on Takagi-Sugeno Fuzzy Criteria

The accurate prediction of water inrush is the key point of safely mining on artesian water, however, there are still lots of difficulties in the quantitative analysis on the influence degree to water inrush of each factor itself and comprehensive influence mutually, or it is in a fuzzy cognitive state. In this paper, on the basis of artificial neural network, the conventional fuzzy algorithm criteria is improved, the integrated neural network water inrush prediction system based on Takagi-Sugeno fuzzy criteria is established, nonlinear relations of mutual and fuzzy functions of various factors that affecting water inrush is better dealt with and according to the training and testing on the network model by a large number of field data, the feasibility, effectiveness and accuracy of using the integrated neural network water inrush prediction system to forecast the probability and quantity of water inrush is proved and great practical significance to guide and ensure safely mining upon artesian water is provided.