基于主成分分析的运行工况车轮轮心载荷识别方法研究

Research on Operational Six-axis Wheel Force Identification Method based on Principal Component Analysis

针对直接获取车轮轮心六分力难度大的问题,对运行工况下车轮轮心六分力的识别试验方法进行研究。采用主成分分析技术将具有偏相关特性的路噪激励源由耦合的多参考问题分解为独立的单参考问题。结合基于奇异值分解的传递函数逆矩阵方法与自由度转换技术,对车速为匀速60km/h工况下轮胎轮心六分力进行识别。所采用方法解决了在路噪CAE分析中获取载荷边界条件困难的问题;同时,该方法也可用于开发车型路噪激励源的对标分析和目标值设定,并可在解决路噪问题时作为激励源分析优化的依据。

The operational six-axis wheel force identification method is studied. Considering that the road noises are partially correlated, the operational interior noise signal need to be transformed from a coupled multi-reference problem to several independent single-reference problems based on principal components analysis (PCA). The spindle load is identified based on MIM (matrix inverse method) and SVD (Singular value decomposition). Finally, through the DOF conversion technology, the identified spindle force is converted into a force and a torque at each wheel center. This method solves the difficulty for obtaining the load boundary condition in the CAE road noise analysis. In addition, this method can also be used for benchmark analysis and target setting of road noise excitation sources to achieve the desired road noise level as well as for diagnosing root causes in troubleshooting processes.