列车防滑器主机故障诊断系统设计

Fault Diagnosis System for the Train Anti-Skid Device Host

摘要:

SAB型列车防滑制动系统应用于国内T型列车,数量多,为了提高列车防滑器主机故障检测和维修效率,针对SAB WABCO公司的SWKAS20C型列车防滑器主机,设计了元件级故障检测与定位专家系统。通过研究模拟电子电路故障诊断技术,设计了基于支持向量机的多故障分类器,用于主机模拟电路故障样本数据的训练和测试,并通过改进算法寻优向量机参数,提高了故障诊断的准确度。搭建了模拟列车运行环境的测试平台,用以模拟主机在多种状态下的故障模式；使用VS2010开发环境编写了MFC专家系统。

Abstract:

SAB type train anti-skid braking system which has a high market share is applied to the domestic T type train.In order to improve the efficiency of fault detection and maintenance for the host of train anti-skid device, the expert system of fault detection and localization in component level is designed for the SAB WABCO company's SWKAS20C type train anti-skid device.A multiple fault classifier based on support vector machine is designed by studying analog electronic circuit fault diagnosis technology,The classifier is used to train and test the fault sample data of the host analog circuit, and improve the accuracy of the fault diagnosis by the improved algorithm.A test platform for simulating the running environment of the train is set up to simulate the failure mode of the anti-skid device host in a variety of conditions.The MFC expert system was prepared by using VS2010 development environment.