Presented paper

Bivariate Multivariate Nonlinear Prediction Model for the Depth of Damaged Floor in Working Area Based on MATLAB

Qu, Xingyue (1,2); Shi, Longqing (1,2); Xu, Dongjing (1,2); Qin, Daoxia (3)
1: College of Earth Sciences & Engineering, Shandong University of Science and Technology, Qingdao 266590, China;; 2: Shandong Provincial Key Laboratory of Depositional Mineralization & Sedimentary Minerals, Qingdao 266590, China;; 3: Feicheng Mining Grou

In the process of coal seam mining, the water-resisting strata deform, and then produce floor heave and cracks, which makes the underground water stored in aquifers pour into mine, causing water inrush accidents. Therefore, the depth of damaged floor is the key data to evaluate the water resistance performance of rock strata under coal seam floor. Aimed at the influences of various complicated factors, a bivariate multivariate non-linear model based on MATLAB was constructed. Six factors were selected as indices. Based on factor analysis, three exogenous latent variables of structural equation model were determined by dimensionality reduction. On this basis, MLP neural network and Deng's grey correlation were used to calculate weights of each factor, then the combined weights of three exogenous latent variables were solved by conflicting evidence fusion. Structure optimal bivariant multivariate nonlinear regression modified model. Taking the No. 21 coal seam mining in Guhanshan coal mine as an example, predict its depth of damaged floor, then auxiliarily prove the accuracy of the modified model by Flac3D numerical simulation. The results show that the bivariate prediction model has higher accuracy, providing theoretical basis for preventing water-inrush from floor.

Presenter:
Room:

IMWA2019 Conference

Genkel st. 4, Perm, Russia, 614990

MWA2019 website uses cookies.

We inform you that this site uses own, technical and third parties cookies to make sure our web page is user-friendly and to guarantee a high functionality of the webpage. By continuing to browse this website, you declare to accept the use of cookies. More information

How to delete cookies

© 2018-2019 IMWA2019