Improving The Reliability Of Infrared Temperature Detection System

The infrared hotbox detecting system is the important device to find out the hotbox of running train and ensure the safety of runway transport. The rapid development of China's runway transport industry requires the system to be not only intelligent, network-based, modular, and more importantly a very high reliability. But the system used currently generally exist the problems such as the high failure rates of rail side equipments, the low hotbox exchange rates. And these problems greatly affected the reliability of the system. Therefore, the means of analyzing these problems are studied to improve the reliability of the system.The research contents are as follows:Firstly, the basic working principles and structure of the system are analyzed. As the important part, the structure of the detecting station is introduced mainly. And it is considered that improving the reliability of detecting station is the root to keep long-term and reliable operation of the infrared hotbox detecting system. Secondly, for the common fault of the rail side equipments in the course of practical work, the reasons are analyzed and the suggestions are given. Thirdly, in order to enhance the hotbox exchange rate, the processing and classifying technique of the bearing temperature wave of infrared ray are chiefly studied. First of all, the method which bases on genetic neural network is proposed to identify the abnormal wave. Then several means of hotbox estimation is studies, including traditional fix quantify analysis such as comparing absolute value, comparing relative value, and estimation method base on genetic neural network, classification method base on fuzzy classification system. According to the analysis of the simulation results of hotbox data, these methods are proved to be able to improve the accuracy of hotbox estimation, and can increase the exchange rate.