列车 PHM 系统的容器化部署策略及其性能优化研究

打开文本图片集

DOI：10.19981/j.CN23-1581/G3.2026.12.033

中图分类号：U296

文献标志码：A

文章编号：2095-2945（2026）12-0132-04

Abstract： In view of the many limitations in architecture deployment of traditional train fault prediction and health management（PHM） systems， such as complex environment， rigid resources， slow service start-up and poor isolation， a microservice deployment solution based on containerization technology is proposed. By splitting core components， introducing Docker to achieve lightweight and portable containerized encapsulation， and designing and orchestrating deployment solutions based on Kubernetes to improve system elasticity， flexibility and maintainability. Performance optimization is achieved through refined resource management， horizontal automatic expansion and reduction， sequential data processing， and multi-level caching mechanisms. Research confirms that all key performance indicators of this deployment strategy are better than traditional solutions， verifying the effectiveness of containerization technology in train PHM system deployment， and providing a reliable technical path for intelligent system operation and maintenance.

Keywords： PHM system; container-based deployment; microservice architecture; performance optimization; intelligent operation and maintenance

在轨道交通系统发展越来越高速、智能的当下，列车运营安全与维护效率不断成为行业核心议题，因此，列车故障预测与健康管理（Prognostics and Health Management， PHM）系统的重要性越发凸显[1]。（剩余6044字）