Bee colony intelligence in fatigue life estimation of simulated magnesium alloy welds

Abstract

The fatigue life of magnesium alloy is influenced by many factors such as stress concentration factor, stress ratio and stress amplitude as well as different material states. Since all these factors affect the fatigue life of the alloy, the effects of these parameters need to be investigated. This study aims to obtain fatigue life values with satisfactory results on the samples of magnesium alloy AZ31. In order to solve the problem, firstly an exponential-trigonometric function model is constructed. Secondly, an artificial bee colony algorithm is used to optimize the weights of this function. Sample data of the alloy is used to verify the correctness of the fatigue life estimation model. The estimated results are compared with the results of experimental studies. Simulation results show that the fatigue life prediction model proposed in this paper can fit the experimental Wöhler lines with high estimation accuracy. The model based on artificial bee colony is a good candidate to estimate the fatigue life of magnesium alloy. © 2019 Elsevier Ltd