Review on hybrid mathematical models to estimate Unconfined Compressive Strength of Jet Grouting Composite- Machine Learning approach

Abstract

Coal-grout composites were created in this examination utilizing the jet grouting (JG) procedure to improve coal mass in underground conditions. To assess the mechanical properties of the made coal-grout composite, its unconfined compressive strength (UCS) should have been tried. A numerical model is needed to clarify the unknown nonlinear connection between the UCS and the affecting factors. In this examination, six computational insight methods utilizing machine learning (ML) algorithms were utilized to build up the numerical models, which incorporates back-propagation neural network (BPNN), random forest (RF), decision tree (DT), support vector machine (SVM), k-nearest neighbors (KNN), and logistic regression (LR). Furthermore, the hyper-boundaries in these run-of-the-mill algorithms (e.g., the hidden layers in BPNN, the gamma in SVM, and the number of neighbor tests in KNN) were tuned by the as of late created bug radio wires search calculation (BAS). To set up the ML models dataset, three kinds of cementitious grout and three sorts of synthetic grout were blended in with coal powders removed from the Guobei coalmine, Anhui Province, China, to make coal-grout composites.