自动变速器行星轮系疲劳点蚀故障诊断

Fatigue pitting fault diagnosis of planetary gear train of automatic transmission

随着新能源汽车、航空航天和工业机器人等行星轮系传动系统对疲劳寿命要求的提高,齿轮的疲劳点蚀的接触疲劳失效成为限制服役性能和可靠性的重要瓶颈。以某款国内汽车无极自动变速器的斜齿行星轮系为研究对象,提出基于粒子群优化算法自适应优化最大相关峭度解卷积参数的斜齿行星轮系点蚀故障诊断方法;通过优化周期性冲击信号时间和滤波器长度,利用无量纲的包络谱峰值因子评价函数自适应提取齿轮点蚀特征,对行星齿轮和太阳轮疲劳点蚀故障实验振动加速度信号进行特征提取与诊断。结果表明:该方法有效过滤外部噪声激励,突出故障特征,实现单一和复合故障特征的有效提取与识别,包络谱幅值因子寻优最大值、迭代次数和寻优速度均明显优于其他算法。

As the demand for improved fatigue life in planetary gear transmission systems used in new energy vehicles,aerospace,and industrial robots increases,gear contact fatigue failure due to pitting becomes a critical bottleneck restricting service performance and reliability.Focusing on the helical planetary gear system of a specific automobile automatic transmission,this paper proposed a pitting fault diagnosis method for helical planetary gear systems based on particle swarm optimization algorithm,which adaptively optimized the deconvolution parameters using the maximum correlation kurtosis.By optimizing the periodic impulse signal time and filter length,and adaptively extracting the characteristics of gear pitting using the dimensionless envelope spectrum peak factor evaluation function to extract and diagnose experimental vibration acceleration signals from various categories of fatigue pitting faults in planetary gear and sun gear.The experimental result indicates that this approach filters external noise excitation,highlights fault features and achieves effective extraction and recognition of single and compound fault features with a notable improved performance in the maximum value of envelope spectrum amplitude factor optimization,iteration times,and optimization speed than other algorithms.