Fisher Multivariate Statistical Discriminant Method based on Principal Component Analysis for Identifying Water-filled Source in Goaf of Mine: a Case Study from the Majiliang Mine, Northern China

The database of water source types and water quality samples is established based on the samples collected from 8210 working face of Majiliang coal mine. In this paper, the water quality types and the training sample database of each waterfilled source were analyzed by Piper three-line graph analysis method and multi-factor analysis method, and Fisher multivariate statistical theory based on principal component analysis method was applied, fisher discriminant water-filling source model based on principal component analysis is established, and the water-filling source of gushing water in mined-out area at sampling stage is identified according to Euclidean distance discriminant principle, the second is floor limestone water and roof sandstone water. The discriminant precision of Fisher discriminant model based on principal component analysis can reach 99.9%. Therefore, the application of this method has an important guiding significance for the on-site identification of mine water inrush and filling source in coal mining face.