Year: 2012
Type: article
Abstract: the model, predicting fatigue life of ductile iron, based on SVM (Support Vector Machine, SVM) has been established. For it is easy to fall into local optimum during parameter optimization of SVM, DE (Differential Evolution algorithm, DE) algorithm was adopted to optimize to improve prediction precision. Fatigue life of ductile iron is predicted combining with concrete examples, and simulation experiment to optimize SVM is conducted adopting GA (Genetic Algorithm), ACO (Ant Colony Optimization) and POS (Partial Swarm Optimization). Results reveal that DE-SVM algorithm is of a better prediction performance.
Source: Physics Procedia
Authors: Ma Yiqun, Wang Xiaoping, An lun
Language: English
Subfield: Mechanics of Materials
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2013251502