Estimation and validation for fatigue properties of steels by symbolic regression

Created by W.Langdon from gp-bibliography.bib Revision:1.8414

@Article{Cao:2024:ijfatigue,

• author = "Weiwen Cao and Xingyue Sun and Xu Chen",
• title = "Estimation and validation for fatigue properties of steels by symbolic regression",
• journal = "International Journal of Fatigue",
• year = "2024",
• volume = "186",
• pages = "108416",
• keywords = "genetic algorithms, genetic programming, Fatigue properties, Symbolic regression, Fatigue life prediction, Machine learning",
• ISSN = "0142-1123",
• URL = "https://www.sciencedirect.com/science/article/pii/S0142112324002743",
• DOI = "doi:10.1016/j.ijfatigue.2024.108416",
• abstract = "Estimating fatigue properties through basic mechanical properties is an expectation for achieving fatigue life prediction in engineering applications. This work aims to employ symbolic regression (SR) methods to explore the relationship among 82 types of steel. With the help of genetic programming and multi-population evolution algorithms, two SR models are established and compared with seven semi-empirical methods and four machine learning models. By comparison, the SR method offers the best balance in terms of prediction performance and interpretability for fatigue properties. Furthermore, based on fatigue properties obtained by SR methods, the predicted fatigue life of up to 91.1percent of the samples can be located within the 2-factor band",

}

Genetic Programming entries for Weiwen Cao Xingyue Sun Xu Chen

Citations