动车组列车安全状态检测网络时延建模  

Delay Modeling of Safety State Detection Network for EMUs

作　　者：李辉[1] 张和生[1,2] 潘成[1] 孙伟[1] 

机构地区：[1]北京交通大学电气工程学院,北京100044 [2]北京交通大学轨道交通控制与安全国家重点实验室,北京100044

出　　处：《机车电传动》2013年第6期75-81,共7页Electric Drive for Locomotives

基　　金：国家自然科学基金项目(61174179);铁道部重点项目(2011Z001-D)

摘　　要：针对动车组列车安全状态检测实时性要求较高的特点,对交换式工业以太网和CAN总线构成的异构网络应用于动车组列车安全状态检测进行了研究。综合分析了交换式工业以太网和CAN总线的周期性报文最大传输时延,建立了安全状态检测网络周期性报文最大传输时延的解析模型。基于该解析模型得出影响网络传输时延的主要因素为CAN节点、Ethernet/CAN网关和车厢级交换机的个数。最后分析了整个网络中周期性报文的最大传输时延随着CAN节点、Ethernet/CAN网关和车厢级交换机个数的变化趋势,并对所建立的解析模型进行了仿真验证。所建立的网络时延模型为定量分析网络实时性提供了理论依据。该建模方法可以推广到其他轨道交通列车中。According to the strictly real-time demands for safety state detection of EMUs, safety status detecting network of EMUs based on the heterogeneous network that consists of switched industrial Ethernet and CAN bus was studied. In order to satisfy the real-time performance of the network, the worst case network delay of the periodic messages was analyzed, and the analysis formula of delay in worst case was deduced. It was clear that the main factors affecting the worst case network time delay were the number of CAN nodes, Ethernet/ CAN gateways and vehicle level switches based on the analysis formula. The trends of the worst case network delay with the changes of CAN nodes, Ethernet/CAN gateways and vehicle level switches were analyzed, and the analytical model was validated by simulation. The worst case network delay model provided a theoretical basis for the quantitative analysis of network performance. The modeling methods and conclusions could be generalized to other rail traffics.

关 键 词：安全状态检测 异构网络 时延分析 ETHERNET CAN 动车组 

分 类 号：TP393[自动化与计算机技术—计算机应用技术] U266.2[自动化与计算机技术—计算机科学与技术]