基于支持向量机回归算法的疲劳寿命预测研究

Study on fatigue life evaluation based on regression algorithm of support vector machine

根据材料疲劳损伤的特点,提出了基于支持向量机回归算法的材料疲劳寿命预测方法。收集材料疲劳性能数据构建训练样本集,建立基于支持向量机回归算法的疲劳寿命预测模型,对疲劳载荷预处理后就可以计算出疲劳寿命。预测结果表明,该方法可利用较少的材料疲劳性能数据,实现疲劳寿命的预测。

According to the characteristics of the material fatigue damage, the paper puts forward the fatigue life pre- diction method based on regression algorithm of support vector machine. The material fatigue performance data are collected to build the training sample set and the fatigue life prediction model based on regression algorithm of support vector machine in order to figure out the fatigue life after pre-treatment of the fatigue load. The prediction result shows that the method can predict the fatigue life using only some fatigue performance data.