Optimization of Geotechnical Material Proportion and Forecast of Mechanical Properties Based on Neural Network

Abstract

With the rapid development of economic construction and the rapid development of rock engineering, it is very important to study the mechanical properties of rock mass. This paper prepares the geotechnical materials and finds out the factors that affect the geotechnical properties. In order to reasonably match the rock materials and find the optimal solution, this paper transforms the multi-factor ratio optimization problem into a multi-factor linear optimization problem. By constructing a multiple regression equation, and using the genetic algorithm to solve the multivariate optimization of the objective function, a suitable matching scheme is found. Then, this paper predicts the mechanical properties of the rock, and uses neural network to predict the shear strength of the rock, and establishes an optimization structure to verify the model. The results show that the error of the ratio optimization scheme adopted in this paper is less than 2%, and the forecast scheme of mechanical properties is less than 18%, which has certain reference significance.