Accurate Estimation for Stability of Slope and Partition Over Old Underground Coal Workings Using Regression-Based Algorithms

Abstract

Numerical modeling simulation has found to be best solution for predicting slope and partition stability over old underground coal workings. But it has taken huge time to complete a single simulation model. In this regard, machine learning-based framework is used to predict the stability of old galleries. A case study is taken up in opencast mine and simulation is carried out using numerical model and machine learning-based framework. Framework has shown an overall accuracy of 94–95% for different slope and partition stability. Framework shows a speedup of 2366 × against numerical simulator. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.