车载燃油系统故障诊断智能算法开发及验证  

Development and Validation of an Intelligent Fault Diagnosis Algorithm for In-Vehicle Fuel Systems

作　　者：惠嘉赫 隗立国 黄英[1] 王健 李准 HUI Jiahe;WEI Liguo;HUANG Ying;WANG Jian;LI Zhun(School of Mechanical Engineering,Beijing Institute of Technology,Beijing 100081,China;China North Vehicle Research Institute,Beijing 100072,China)

机构地区：[1]北京理工大学机械与车辆学院,北京100081 [2]北方车辆研究所,北京100072

出　　处：《汽车工程学报》2024年第6期1025-1035,共11页Chinese Journal of Automotive Engineering

基　　金：国家级基础科研项目(202020329201)。

摘　　要：提出一种基于主成分分析与极限学习机的柴油机燃油系统多分类故障诊断智能算法,以实车数据为基础在Simulink中建立模型实现离线验证。然后参考OSA-CBM标准建立了在线故障诊断智能算法架构,基于此架构开发柴油机燃油系统在线故障诊断智能算法,并利用硬件在环平台完成故障诊断验证测试。验证结果表明,所开发的燃油系统多分类故障诊断智能算法,在离线与硬件在环平台中均具有较高的准确性与可靠性,可应用于车载。This paper proposes an intelligent multi-class fault diagnosis algorithm for the diesel engine fuel system based on PCA and ELM.Firstly,a fault diagnosis model is established in Simulink,utilizing real vehicle data to support offline verification.Subsequently,an architecture for the intelligent fault diagnosis algorithm is designed,following the OSA-CBM standard.Based on this architecture,the online fault diagnosis algorithm is developed in the Simulink and tested using a hardware-in-the-loop(HIL)platform.The verification results show that the proposed intelligent multi-class fault diagnosis algorithm achieves high accuracy and reliability in both offline simulations and HIL testing,indicating its potential for in-vehicle applications.

关 键 词：燃油系统 在线故障诊断 硬件在环 智能算法 

分 类 号：TK428[动力工程及工程热物理—动力机械及工程]