针对电动车动力总成悬置系统(PMS)参数同时存在不确定性和相关性的复杂情形，本文开展了考虑概率参数相关性的电动车悬置系统固有特性分析研究。首先，分别采用相关系数矩阵和概率不确定变量描述PMS参数的相关性和不确定性；然后，基于蒙特卡洛抽样提出了一种求解概率参数相关情形下PMS固有特性统计矩的蒙特卡洛法(MCM)；接着，基于稀疏网格数值积分(SGNI)提出了一种求解PMS固有特性统计矩响应的高效方法；最后，通过某电动车PMS算例验证方法的有效性，分析结果表明：以MCM为参考，SGNI方法在求解PMS固有频率、解耦率的统计矩及响应边界方面具有良好的求解精度和较高的求解效率；系统概率参数的相关性对解耦率响应边界具有较大影响，计及概率参数的相关性可获得更加合理的分析结果。

Aiming at the complex situation where the parameters of powertrain mounting system (PMS) of electric vehicle are both uncertain and correlated, this paper investigated the inherent characteristics of electric vehicle PMS considering the correlation of probabilistic parameters. Firstly, the correlation matrix and probabilistic parameters were employed to describe the correlation and uncertainty of PMS parameters. Then, the Monte-Carlo method (MCM) was developed to calculate the PMS inherent characteristics with correlated probabilistic parametersbased on Monte-Carlo sampling. Then, an efficient method for calculating the statistical moments of PMS inherent characteristics was put forward based on sparse grid numerical integration (SGNI). Finally, the effectiveness of the proposed approach was demonstrated by the numerical example of an electric vehicle PMS. The analysis results show that, the SGNI method has good accuracy and efficiency in solving statistical moments and bounds of PMS natural frequencies and decoupling ratios, compared with MCM. The correlation of probabilistic parameters has a great influence on the upper and lower bounds of decoupling ratios. More reasonable analysis results can be obtained by taking the correlation of the probabilistic parameters into consideration.