Abstract:  Considering the limitation of the nonlinear parameter identification method based entirely on the time domain or the frequency domain， a parameter identification method for nonlinear hydro-pneumatic suspensions was proposed based on feature extraction in time-frequency domain. The vehicle dynamics simulation models with nonlinear hydro-dynamic suspensions were established， the IFFT method was used to obtain the road excitation with the level of the standard road surface， the feature quantities in the time-frequency domain of the experimental results were extracted by combining with the wavelet analysis and filtering， and an optimization model for identification of nonlinear hydro-pneumatic suspension parameters was constructed. By minimizing the feature quantities of simulation and experimental results in the time-frequency domain， the nonlinear parameters of the hydro-pneumatic suspensions were accurately identified under actual conditions. The quantitative and qualitative analyses of the identification show that the identification results are accurate and reliable. By combining road time domain modeling， wavelet filtering technology and construction method for the optimization model with simulated annealing algorithm， an effective and reliable identification method was provided for parameter identification of nonlinear suspension systems. 

Key words: hydro-pneumatic suspension, parameter identification, time-domain model on road, inverse fast Fourier transform（IFFT） method, wavelet analysis, simulated annealing algorithm

摘要： 针对完全基于时域或完全基于频域的非线性参数识别方法的局限性，提出了一种兼顾时频域特征量提取的非线性油气悬架参数识别方法。在建立含非线性油气悬架车辆动力学仿真模型的基础上，采用快速傅里叶逆变换法获得标准路面不平度等级的输入激励，结合小波分析和滤波处理提取实验结果的时频域特征量，构建非线性油气悬架参数识别的优化模型，通过最小化仿真结果与实验结果在时频域的特征量，实现了非线性油气悬架参数在实际工况下的准确识别。识别的定量与定性分析表明了识别结果的准确可靠。通过将路面时域建模技术、小波滤波技术和优化模型构造方法与模拟退火算法的有机结合，为非线性悬架系统的参数识别提供了一种有效可靠的方法。

关键词: 油气悬架, 参数识别, 路面时域建模, 快速傅里叶逆变换法, 小波分析, 模拟退火算法