The reliability assessment of train control and management system (TCMS) is essential for the condition monitoring of high-speed train. Different from other general complex systems, the TCMS has the characteristics of multi-system unit, strong coupling and multiple factors. Considering the special system operating environment and high safety requirements of high-speed train. In this paper, for the reliability assessment of TCMS, we propose a new quantitative model based on the evidential reasoning rule and covariance matrix adaptation evolution strategy algorithm, the proposed model offers the following advantages: it has a strong modeling capability for the TCMS reliability, it has an interpretable model assessment process, it can describe the assessment result under probabilistic uncertainty and ignorance uncertainty, and it possesses considerable robustness. To make the model interpretable, an assessment hierarchy is established for the TCMS; to improve model robustness, weights interval is applied to replace the trained weights as the model weights. Several traditional methods are compared with the proposed model to demonstrate its performance, the results show that the proposed model has a better training accuracy. Moreover, a case study is conducted to verify the model’s functional feasibility.

列车控制与管理系统（TCMS）的可靠性评估对于高速列车的状态监测至关重要。与其他一般复杂系统不同，TCMS具有多系统单元、强耦合、多因素的特点。考虑到高速列车特殊的系统运行环境和较高的安全要求。在本文中，针对TCMS可靠性评估，我们提出了一种基于证据推理规则和协方差矩阵自适应演化策略算法的新量化模型，该模型具有以下优点：对TCMS可靠性具有很强的建模能力，具有可解释的模型评估过程，可以描述概率不确定性和无知不确定性下的评估结果，并且具有相当的鲁棒性。为了使模型具有可解释性，为 TCMS 建立了一个评估层次；为了提高模型的鲁棒性，应用权重区间来代替训练的权重作为模型权重。将几种传统方法与所提出的模型进行比较以证明其性能，结果表明所提出的模型具有更好的训练精度。此外，还进行了案例研究以验证模型的功能可行性。结果表明，所提出的模型具有更好的训练精度。此外，还进行了案例研究以验证模型的功能可行性。结果表明，所提出的模型具有更好的训练精度。此外，还进行了案例研究以验证模型的功能可行性。