The Application of Support Vector Machine in the Calculation of Pavement Structural Modulus

Abstract

In this paper, support vector machine (SVM) is introduced into back analysis, and the inversion method of pavement modulus based on SVM is discussed. The main work can be summarized. In the process of modulus optimization inversion, the forward analysis process needs to be called repeatedly, which results in a large amount of inversion calculation. In this paper, support vector machine is introduced into the back analysis, and a new idea of pavement modulus inversion based on support vector machine is proposed, that is, firstly, the support vector machine model is established through the learning of samples, Then, the trained support vector machine model is used to calculate the pavement deflection corresponding to the modulus parameters instead of the numerical model, so as to reduce the amount of back analysis calculation. Taking advantage of the strong global search ability and high search efficiency of particle swarm optimization algorithm, this paper combines it with support vector machine, and successively applies it to the optimization selection of support vector machine model parameters and intelligent search of pavement modulus parameters. The analysis results show that with the help of particle swarm optimization algorithm, the support vector machine model with the best prediction performance and the optimal parameter inversion results can be effectively obtained.