A Classifier Model For Recognition Of Railway Infrastructure Abnormal State

2018 INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND INTELLIGENT SYSTEMS (CIIS 2018)(2018)

引用 1|浏览12
暂无评分
摘要
In recent year, the technologies of the fourth industrial revolution such as AI and IOT are emerging as the key means of railway industry. Moreover, disruptive technologies are expected to improve the safety and reliability of railway system through a gradual expansion of vehicle control which could be automated. In this paper, we propose a method to provide the infrastructure state with the information needed for the abnormal situation recognition model and the control technology for the autonomous train operation, which can affect the train running condition by utilizing various railway infrastructure data based on the IOT. The data required for the railway infrastructure situation are specified by rail temperature, intrusion detection, number of passenger, and line snowdrift. After applying the Sugeno fuzzy inference method of the proposed recognition model, the fuzzy rules are generated by operation standards of Korea Train Express(KTX). In the final task, the result value is calculated by fuzzy rule-base, then it uses the decision tree to classify the state of railway infrastructure. It was used to evaluate the recognition rate of the proposal model by using the railway infrastructure data set. The recognition rate result showed about 93 % accuracy. If the fuzzy inference-based proposed model is applied to the railway sector, it will be possible to recognize the situation of the railway infrastructure more efficiently than the conventional control based monitoring systems.
更多
查看译文
关键词
Abnormal state Recognition, Railway infrastructure, IOT Sensor, Autonomous Trains, Fuzzy inference, Classification
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要