基于贝叶斯网络的城轨车辆空调系统故障预测模型研究
电子技术应用
于海飞1,梁晓珂1,贾英武2
1.中车南京浦镇车辆有限公司;2.石家庄国祥运输设备有限公司
摘要: 城轨车辆空调系统结构复杂、运行工况多变,传统故障预测方法难以有效应对其多源耦合与动态变化的特性。提出一种基于贝叶斯网络的故障预测模型。该模型首先利用自编码器对正常运行状态下的多源传感器数据进行无监督学习,通过重构误差实现异常检测;进而结合贝叶斯网络进行故障因果推理,实现故障定位与根因追溯。实验结果表明,该模型对制冷剂泄漏、冷凝风机异常和通风机异常等典型故障的诊断准确率均超过92%,验证了其有效性与实用性,为城轨车辆空调系统的智能运维提供了可靠的技术支持。
中图分类号:TP181;U279.3 文献标志码:A DOI: 10.16157/j.issn.0258-7998.257172
中文引用格式: 于海飞,梁晓珂,贾英武. 基于贝叶斯网络的城轨车辆空调系统故障预测模型研究[J]. 电子技术应用,2026,52(8):99-105.
英文引用格式: Yu Haifei,Liang Xiaoke,Jia Yingwu. Fault prediction model for urban rail vehicle air conditioning systems based on Bayesian network[J]. Application of Electronic Technique,2026,52(8):99-105.
中文引用格式: 于海飞,梁晓珂,贾英武. 基于贝叶斯网络的城轨车辆空调系统故障预测模型研究[J]. 电子技术应用,2026,52(8):99-105.
英文引用格式: Yu Haifei,Liang Xiaoke,Jia Yingwu. Fault prediction model for urban rail vehicle air conditioning systems based on Bayesian network[J]. Application of Electronic Technique,2026,52(8):99-105.
Fault prediction model for urban rail vehicle air conditioning systems based on Bayesian network
Yu Haifei1,Liang Xiaoke1,Jia Yingwu2
1.CRRC Nanjing Puzhen Co., Ltd.;2.Shijiazhuang Guoxiang Transportation Equipment Co., Ltd.
Abstract: The air conditioning system in urban rail vehicles features a complex structure and variable operating conditions, making it difficult for traditional fault diagnosis methods to effectively address its characteristics of multi-source coupling and dynamic changes. This paper proposes a fault prediction model based on a Bayesian network. The model first utilizes the auto-encoder to perform unsupervised learning on multi-source sensor data under normal operating states, achieving anomaly detection through reconstruction error. Subsequently, it integrates the Bayesian network for causal fault reasoning to accomplish fault localization and root cause tracing. Experimental results demonstrate that the diagnosis accuracy of this model for typical faults, such as refrigerant leakage, condenser fan abnormality, and ventilator abnormality, exceeds 92%, verifying its effectiveness and practicality. This provides reliable technical support for the intelligent maintenance of urban rail vehicle air conditioning systems.
Key words : urban rail vehicles;air conditioning system;auto-encoder;anomaly detection;Bayesian network;fault prediction
引言
城市轨道交通的蓬勃发展对其运营安全性与可靠性提出了极高要求。城轨车辆空调系统作为维持车厢环境舒适与设备正常运行的核心装备,其质量与乘坐体验受到了广泛关注[1],其在高密度运营与复杂环境下易发生冷凝风机异常、制冷剂泄漏、通风机性能退化等故障,影响乘客舒适与行车安全,亟需从事后维修走向预测维护[2-4]。近年城轨车辆空调系统故障检测诊断快速发展,数据驱动与知识引导融合成为趋势,其中贝叶斯网络因能处理不确定性与因果推断,被用于故障定位与归因[5-8]。国内在轨道交通与空调领域也展开了针对空调传感器,多联机与城轨车辆智慧空调的研究,验证了贝叶斯网络方法的可行性,但是仍存在一些问题:(1)城轨车辆工况差异大、数据稀疏且标注不足;(2)跨系统耦合故障的解释性与可视化不足;(3)模型在全寿命周期中的在线更新与不平衡样本鲁棒性不足等[9-15]。本文结合城轨车辆空调的系统结构与运维知识,构建融合先验机理与数据学习的贝叶斯网络故障预测模型,面向跨层级因果链路建模与可解释诊断;在充足的样本情况下通过结构学习与参数学习提升泛化,并利用实际运营数据与数据集流程进行验证,兼顾准确性、可解释性与可部署性。
本文详细内容请下载:
https://www.chinaaet.com/resource/share/2000007217
作者信息:
于海飞1,梁晓珂1,贾英武2
(1.中车南京浦镇车辆有限公司,江苏 南京 210031;
2.石家庄国祥运输设备有限公司,河北 石家庄 050035)

此内容为AET网站原创,未经授权禁止转载。
