为汽车试验与动力学仿真提供更准确的路面激励,提出了一种对实测路面谱进行二维等效重构的方法.结合某三级公路实测路面数据特征,系统研究了经验模态分解法去除趋势项、经验模态分解小波阈值法滤波去噪、实测路面二维自回归模型(auto regressive,AR)重构等方法,并利用相关系数方法进行了重构精度分析.结果表明,基于固定门限准则的经验模态分解小波硬阈值去噪后的功率谱方差最小,利用AR模型重构道路谱与实测道路谱重合度高,为道路谱的进一步应用提供参考.%A method of two-dimensional equivalent reconstruction for the measured road surface spectrum was proposed in order to provide more accurate road incentives for automotive test and dynamic simulation. Combined with the data characteristics of a level 3 highway test surface, some methods including the empirical mode decomposition method were used to remove the trend items, and the EMD wavelet threshold method for filtering and denoising. The two-dimensional AR reconstruction method for the measured road surface was studied systematically. Then, the reconstruction precision was analyzed by the correlation coefficient method. The results showed that the variance of power spectrum after denoising by the empirical mode decomposition wavelet hard threshold based on the fixed threshold criterion was the minimum, and the relevant degree of the reconstructed road spectrum by AR model with the measured one was high. The method can provide a basis for the further application of road spectrum.
展开▼
