具有学习机制的列车运行调度Agent  

Train Operation Dispatch Agent With Learning Machinery

作　　者：李梅月[1] 王宏刚[1] 

机构地区：[1]太原科技大学计算机学院,山西太原030024

出　　处：《电脑知识与技术》2009年第5X期3984-3986,共3页Computer Knowledge and Technology

基　　金：轨道交通控制及安全国家重点实验室(北京交通大学)开放课题资助(2007K009)

摘　　要：为提高列车运行调度的智能性,分析了列车运行调度活动,采用agent技术建立了具有学习机制的列车运行调度agent结构,学习机制包含自学习机制和它学习机制,并详细论述了学习机制。通过采用原京沪高速铁路数据建立仿真平台对具有学习机制的列车运行调度agent进行了验证,结果表明,采用具有学习机制的列车运行调度agent可提高列车运行调度系统的智能性,调度决策所用时间缩短一半左右。In order to improve the intelligence of train operation dispatch,by analyzing train operation dispatch activity and using agent technology,the agent architecture with learning machinery for train operation dispatch was built,which includes self-learning machinery and other-learning machinery,and the learning machinery was described in detail. The agent architecture was verified in the simulation system using the Jing-Hu high-speed railway data. The result indicates that the agent architecture with learning machinery could improve the intelligence of train operation dispatch,and the time to make dispatch decision is about one half of the time to make dispatch decision without the learning machinery.

关 键 词：AGENT 列车运行调度 学习机制 

分 类 号：U284.59[交通运输工程—交通信息工程及控制]