车辆底盘故障诊断的多智能体技术研究  

Research on Multi-agent Technology of Vehicle Chassis Fault Diagnosis

作　　者：牛礼民[1] 许吉禅[1] 刘超[1] 

机构地区：[1]安徽工业大学机械工程学院,安徽马鞍山243032

出　　处：《安徽工业大学学报（自然科学版）》2013年第4期419-424,共6页Journal of Anhui University of Technology（Natural Science）

基　　金：国家自然科学基金项目(51275002)

摘　　要：半主动悬架系统(SAS)和电动助力转向系统(EPS)的故障诊断对保障整车安全性及其重要。为快速准确诊断SAS和EPS的子系统故障,基于多智能体理论构建出半主动悬架智能体、电动助力转向智能体、系统智能体、显示智能体等,构架底盘悬架子系统与转向子系统通用故障树,建立故障征兆库、专家规则库等数据库,利用智能体的智能行为决策问诊、快速分析判别原因并显示维修建议。通过故障实例诊断模拟,验证了智能体技术在底盘故障诊断中的可行性与有效性。Fault diagnosis of semi-active suspension （SAS） and electric power steering （EPS） plays a vital role in vehicle safety.An intelligent diagnosis method based on multi-agent theory was proposed in order to diagnose the faults of suspension and steering subsystems rapidly.SAS agent,EPS agent,system agent （SYS agent）,display agent （DSP agent） were built.And generalized fault tree of SAS and EPS was framed out as well.Database such as fault symptom,expert rule,and so on were set up,intelligent behaviors of agents was emploged to make decision and judge fault reasons quickly.Proper maintenance suggestions were proposed synchronously.With the fault example simulation the feasibility and validity of multi-agent technology for chassis fault diagnosis were proved.

关 键 词：故障诊断 多智能体 半主动悬架 电动助力转向 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]