Constitutive Relation Research of Q235 Steel Based on Support Vector Machine

摘要

Aiming at the problem of high cost of mechanics performance testing and uncontrollable of the experimental accuracy, a new modeling method which called "mapping model" based on Support Vector Machine (SVM) is proposed. The SVM regression nonlinear prediction model was established based on the small sample of experimental data by taking the different temperatures of Q235 tensile test as input parameters, the elastic modulus (ultimate strength, yield strength) as output parameters. By MATLAB programming, stress-strain properties of Q235 steel were calculated and analyzed for different temperatures referring to British standard, the constitutive relation which is convenient for engineering use are obtained by fitting the small sample of experimental data and forecast data. The research shows that constitutive relation equations based on experimental data and forecast data could show a complex nonlinear relationship between mechanical properties and temperature. At the same time, according to the comparison of mapping results which is predicted by SVM and neural network, it comes the conclusion that prediction model via SVM have the advantage of less relative error, fewer calculation, high goodness of fit.