Abstrakt:
Railways are considered as a cost-effective, reliable and safe form of transport. Ensuring the timely maintenance of this transport mode is the prioritized task of the railway operators. Railway ballast is the governing element of track infrastructure, which plays an immense role in the overall track stability. Therefore, assessing the condition of ballast reveals very significant information about the overall track condition. Detecting the level of a phenomenon called ballast fouling, i.e. contamination of ballast within the ballast structure provides beneficial information about the status of the ballast. Ballast fouling occurs when air voids in the ballast are filled with finer materials. It is of crucial importance to diagnose the level and the type of fouling which may jeopardize the all-around safety of the track and its economic life. Another significant threat to track stability is the trapped water within track infrastructure which is mostly due to ballast fouling. Ground Penetrating Radar (GPR), is a powerful non-destructive geophysical diagnosis method providing cost-efficient, time-saving and continuous railway ballast surveys. To that effect, this dissertation addresses evaluation of railway ballast condition by using GPR. Performance compliance tests for the available GPR set have been performed to evaluate the performance of available GPR equipment. Numerous laboratory measurements have been undertaken to analyze ballast fouling condition with two different granite ballast types and three different fouling materials with 2 GHz air-coupled antenna. Strong linear trends have been found between ballast fouling percentage and Relative Dielectric Permittivity (RDP) values with a coefficient of determination greater than 0.9. Moisture content influence on GPR signal was also assessed through laboratory measurements. Relationships between water content and RDP values were determined. Conducted field surveys validated the laboratory measurements where ground truth data and GPR data using the laboratory results were in good agreement. Field survey results were analyzed in terms of antenna types, antenna frequencies ranging from 400MHz to 2GHz, antenna orientation, and type of sleepers under survey. Characteristics of EM signal reflected from limestone railway ballast has been achieved as a function of ballast fouling condition by silty sand, antenna types, antenna frequencies ranging from 1GHz to 2GHz, antenna orientation and presence of sleeper(s) and rails under controlled laboratory conditions. RDP values of clean and fouled ballast were determined. Overall, all the laboratory and field measurements proved the effectiveness and eligibility of GPR method to diagnose the condition of ballast in terms of fouling level and type, moisture content, and to identify the layering of clean and fouled ballast, hence improving the evaluation and monitoring efficiency of railway infrastructure site investigations. Moreover, sleeper influence on GPR reflected signal was assessed. Optimum type, frequency, and orientation of GPR antennas on GPR railway ballast surveys were determined.
