橡胶隔振器疲劳寿命智能预测

Intelligent Prediction of Fatigue Life for Rubber Vibration Isolators

在不同应变比恒幅疲劳载荷环境下,橡胶隔振器常用一种填充天然橡胶材料制成的哑铃型试件进行单轴疲劳试验,进而构建一种用来预测橡胶疲劳寿命的支持向量回归(Support vector regression,SVR)模型.为改善预测模型的收敛速度、准确度和稳定性,引入改进的麻雀搜索算法(Improved sparrow search algorithm,ISSA)来优化SVR模型超参数.与遗传算法(Genetic algorithm,GA),粒子群算法(Particle swarm optimization algorithm,PSO),标准的麻雀搜索算法(SSA),差分进化-灰狼算法(Hybrid grey wolf optimization,HGWO)优选参数的SVR模型进行对比结果表明,基于ISSA-SVR模型的预测精度、速度和稳定性都表现最佳.为进一步说明ISSA-SVR模型的预测能力,建立考虑应变比影响的疲劳寿命解析模型.对ISSASVR模型、解析模型和两个已发表模型的比较分析表明,ISSA-SVR具有最准确的预测寿命,预测结果聚集在1.5倍分散线以内.

Under constant amplitude fatigue loads with different strain ratios,uniaxial fatigue tests are conducted on dumbbell-shaped specimens made from a commonly used filled natural rubber material for rubber vibration isolators.Subsequently,a support vector regression(SVR)model is constructed to predict the fatigue life of the rubber.To improve the convergence speed,accuracy,and stability of the prediction model,an improved sparrow search algorithm(ISSA)is introduced to optimize the hyperparameters of the SVR model.Comparison results with SVR models optimized using the genetic algorithm(GA),particle swarm optimization algorithm(PSO),standard sparrow search algorithm(SSA),and hybrid grey wolf optimization(HGWO)show that the ISSA-SVR model performs best in terms of prediction accuracy,speed,and stability.To further demonstrate the predictive capability of the ISSA-SVR model,a fatigue life analytical model considering the impact of strain ratio was established.Comparative analysis of the ISSA-SVR model,the analytical model,and two published models indicates that the ISSA-SVR model has the most accurate life predictions,with results concentrated within a 1.5 times dispersion line.