Related Research Of Disease And Safety Evaluation For Railway Tunnel Lining

Diseases of tunnel lining have plagued the railway management. Because of the particularity of structure and location, those diseases are direct threat to the tunnel safety. China becomes one of the largest railway tunnels in the world, however there is a big gap in the quality of the railway tunnel in comparison with other countries. In recent years, the proportion of China's railway tunnel with disqualification has reached to 60%, according to autumn inspection statistics from the ministry of railways. Each year, tunnel railway authorities have conducted a number of tunnel testing, and accumulated a large number of tunnel disease data, which have largely been idle. Although some of these traditional management systems could supply some simple query and statistical work for the data collected, however, these data do not apply in the deep-level research, Such as the potential association rule mining, security assessment, decision supporting and so on.This article aims to provide an efficient disease detection method.The management can plan the tunnel maintenance and repair work more reasonable by this method. Defect detection of tunnel lining is mainly manual inspection with experience and data processing just only fill in the form. There is no systematic treatment. In order to solve these problems, we should treat the radar maps and build the image database by image segmentation and feature extraction. We can translate the image information into digital information by disease recognition.Obtained the data for diseases, we can run associated analysis with SQL Server 2005. After setting different support and size, we can get dependence of various diseases network and find frequent item sets. According to the results, we can infer the future diseases by occurred ones. At the same time, we can find the reason of diseases by historical analysis. This would help us to plan preventive work.There are so much uncertain information in tunnel safety evaluation index, so it's more appropriate and feasible to make the safety assessment of the tunnel on fuzzy pattern recognition theory. Based on fuzzy pattern recognition theory, the assessments of railway tunnel safety indicators are reduced to two levels, by the low level of indicators to evaluate the up level indicators. Followed step by step evaluation, the first assessment is conducted on the segment tunnel, and finally to the entire tunnel. Thus a detailed algorithm for evaluating security is designed. By compare to the weight of various diseases, safe driving and threshold point for close the tunnel or not, we can have the answer whether the tunnel is continue opening or close to maintenance as quick as possible.