Using Characterized Grids Resolving Method for GPR on Investigating Railway Mud-pumping

• Main Authors: HSU, CHIH-HAO, 許志豪
• Other Authors: KUO, CHIH-PING
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/b7vfsw

碩士 === 明新科技大學 === 土木工程與環境資源管理系碩士班 === 105 === The strength and stability of filled foundation under railway would be effected by nature factors such as change of water content and soil density. Mud-pumping is one very commonly worldwide phenomenon. The majorly inducing factors are soil situations, saturated degree, and repeated loading from trains. Remediation for fouled railway foundations by mud-pumping is a wide and various issue. Effective regulating and controlling dealing method require study in detail. The traditionally inspecting method is by digging and dynamic touching. Although that is the most direct way, the result is very limited in local area. Much time and effort require as well but not equally conditions can be obtained. Moreover, that may request the operation of train service halting for the excavating. It is difficult to have a comprehensive and detail controlling of whole railway foundation. Therefore, the traditional method is incapable to reflect the overall information of foundation under above mentioned. Existing investigating method for mud-pumping is by visual patrol and only surfacing will be observed. To deal with previous problem, Ground-penetrating radar (GPR) inspecting method is adapting in this research. The result of in-situ experiment shows that GPR provides briefly precise, fast, continuous, and non-destructive inspection for the flaws of railway foundations. The structures of layered weakness, differences in vertical, and distribution of mud-pumping can be obtained by the scanned images from GPR. That can assist the design for improving foundation. However, GPR is just only an assistant utility for traditional method. Now GPR technology and its interpretation can be used for investigating the integrity of lower part of railway foundation. The distribution of mud-pumping also can be illustrated by searching interfaces of strongly and weakly reflecting energy and characterizing-gridding the images. The positive discriminating ability of the proposed method can reach over 82.1% for entire area and 92.7% for effective area after comparing with the sectional excavation. The result by characterized and gridded images is fitting well to actual situation and for furthered usage on replacement and improving engineering.