Study on the geological meaning and application of mine water inflow series after phase space reconstruction
LI Jianlin
HE Qi
WANG Shuwei
WANG Xinyi
WANG Chong
XUE Yang
Abstract：Objectives To determine the geological meaning of mine water inflow series after phase space re-construction and improve the accuracy of water inflow prediction,Methods taking Wangxingzhuang Mine as an example,the correlation between column vectors in the reconstructed phase space of the water inflow se-quence and the main control factors of mine water inflow was analyzed.On this basis,a water inflow predic-tion model coupled with chaos theory and artificial neural network(Chaos-ENN)was established.Results The results showed that the embedding dimension of the phase space was equal to the number of main con-trol factors of mine water inflow.The reconstructed phase space columns 1,2,4,5 and 6 were highly corre-lated with the depth of the C2tL7-8 aquifer burial,the depth of the O2m+Є3ch aquifer burial,the goaf area,the depth of the C2tL1-4 aquifer burial,and the development length,respectively.The column 3 was related to other comprehensive factors that were difficult to quantify.The prediction accuracy of the Chaos-ENN model at Wangxingzhuang Mine reacheded 97.91%.Conclusions The column vectors of the phase space after the reconstruction of the water inflow sequence had clear geological meaning.Chaos theory quantified the num-ber and values of the ENN input layers in the water inflow prediction model,allowing the establishment of a Chaos-ENN model for mine water inflow prediction using only the sequence values of water inflow.This model addressed the challenges of determining and quantifying the main controlling factors in water inflow prediction.It achieved high prediction accuracy and demonstrated significant potential for broader applica-tion.
Keywords：mine hydrological systemphase space reconstructionmain control factor of water inflowchaotic characteristicChaos-ENN prediction model
Publication Date：2024-10-28
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：10( 43-52 )