Source identification of mine water inrush based on principal component analysis and grey situation decision

Abstract

Coal mining is bound to destroy the underground aquifer structure, which will lead to mine water inrush disaster. An accurate and rapid identification of water inrush sources is the crux of preventing the recurrence of water inrush incidents. In this regard, 37 training water samples and 14 verification water samples were extracted from three types of aquifers in the Xieqiao coal mine, China. Na+, K+, Ca2+, Mg2+, Cl−, SO42− and HCO−3 were used as the evaluation variables. The principal component analysis was used to eliminate the redundant ion variables in the training samples. The grey situation decision method combined with the entropy weight was used to establish the recognition model. The ion variables of the verification samples were substituted into the model calculations, and the comprehensive accuracy of the model was found to be 85.71%. The proposed method has the advantages of accuracy and speed compared to other contemporary recognition methods. The grey situation decision-making method overcomes the problem that single-factor evaluation cannot identify water inrush, and the entropy weight method can reflect the degree of difference between the variables. Based upon this recognition model, it provides a new method for recognizing water inrush sources, which would also be beneficial to prevention and control of mine water hazards.