Genetic algorithm- Support vector machine iterative methods have been studied and applied to determine the yield strength, tensile strength and hardness of C-Mn steel are predicted. The Krylov methods such as CGS, Bi-CGSTAB, TFQMR and GMRES are some of the most efficient methods for analyzing steel grades like SAE steel, British standard, ISO standard and European steel. The incomplete factorization-based preconditioned considered here is ILUPBS (l) proposed by Mittal [1]. The numerical implementations of the ILUPBS (l) preconditioned iterative methods are tested and compared on three numerical examples to demonstrate their efficiency. Numerical experiments presented here with show that all methods combined with the ILUPBS (l) (l>0) preconditioner is shown to those extremely effective in achieving optimal convergence rates. But, the convergence rate of all iterative method combined with ILUPBS (2) is found to be the best. Classifiers like KNN, AdaBoost, Decesion Tree, Random Forest, Gradient Boosting are compared with GA-SVM approach.
Volume 11 | 01-Special Issue
Pages: 911-916