Chu, Xuanxuan
(2023)
Road and Railway Foundation Response to Moisture Content and Drainage.
PhD thesis, University of Nottingham.
Abstract
Attenuation and loss of service capability are closely related to subsurface water in a pavement/track foundation. The strength of the foundation decreases with increasing moisture content. Thus, it is important to investigate foundation soil behaviour. It is beyond the scope of this project to thoroughly cover both pavements and railway tracks, and the analysis of pavement foundations was concentrated on. The findings from the pavement analysis are expected to provide a reference for understanding the foundation response of railway tracks since railways are similar except for the upper structures. This research aims to reveal the significance of drainage and broaden the understanding of foundation behaviour at various moisture contents. The research was performed from various aspects (including simulation analysis and laboratory tests), thus comprehensively demonstrating the effect of moisture content and drainage. Prediction models of resilient modulus were developed for unsaturated soils using data from literature sources. Furthermore, moisture effects and/or drainage benefits were demonstrated on three levels, i.e. simulation analysis, triaxial tests and subgrade box tests. A selected series of studies were analysed and multi-layered elastic simulations were carried out for sensitivity analysis. The sensitivity results provide a reference for parameter selection in the pavement analysis following triaxial tests. Triaxial tests were performed to investigate the stress-strain behaviour and modulus of silty sand at various laboratory-controlled moisture contents and drainage conditions. Subgrade box tests were conducted to study the deformation behaviour under cyclic traffic loading in a more realistic way compared to triaxial tests.
Testing data were collected from previous studies and analysed, and a general trend was revealed among the data. A relationship of resilient modulus with confining pressure, stress states and moisture content was developed, namely the consistency index model and the stress-modified consistency index model. In terms of the model parameters, their relationship with clay content and plasticity index was proposed through regression analysis. These models showed fairly good prediction results of resilient modulus with a wider variety of soil types. Thus, the consistency index as a normalized soil property can extend sensitivity analysis results to different soils. These models were proposed for the first time. The proposed models initially correlated consistency index with resilient modulus and provided alternative ways for resilient modulus prediction at various moisture contents.
Based on multi-layered elastic theory, a sensitivity analysis was conducted to investigate the variation of pavement response at different moisture contents. The sensitivity analysis involved two aspects, data analysis using predetermined subgrade modulus from model calculations and data analysis using subgrade modulus from literature sources. The variables adopted for the analysis included the thickness and elastic modulus of pavement layers (asphalt concrete, base layer and subbase layer), axle load and subgrade modulus at different moisture contents. Results from the sensitivity analysis included fatigue and rutting life and their sensitivity indices to moisture variation. A design of a polymer drainage layer was also proposed. The critical factors influencing pavement response were identified (e.g. the thickness of asphalt concrete), thus providing a basis for the pavement analysis using unloading modulus from multistage triaxial tests. The comprehensive analysis of the sensitivity of pavement response to moisture content could expand the understanding of moisture effects. This may also help to establish a way to evaluate in-situ pavement structures and pavement design.
Triaxial tests included saturated consolidated undrained and drained triaxial compression tests and unsaturated constant water content triaxial tests. A series of single-stage (monotonic) and multistage loading tests were performed under various confining pressures for saturated and unsaturated tests. Drainage was allowed between loading stages. Results were investigated, including stress-strain behaviour, unloading modulus, cohesion, friction angle, volumetric behaviour (for consolidated drained tests) and matric suction (for unsaturated soils tests). The unloading modulus was adopted as an input into KENPAVE in order to study foundation response in a flexible pavement structure. Effects of drying, wetting and drainage were revealed. The drainage was quantitatively related to pavement response based on pore water pressure reduction. The effects of drainage and moisture content on pavement foundation soils were further comprehensively investigated in a laboratory-controlled manner (i.e. triaxial tests) under different drainage conditions, and the results were compared so as to obtain a sensitivity analysis.
Subgrade box tests were carried out to simulate a more realistic condition. Cyclic loading was applied at various cyclic stresses, moisture contents and frequencies. The settlement of the subgrade surface was analysed. Moisture distribution with depth before and after loading was checked. It was found that the soil settlement increased with cyclic stress and moisture content. In contrast, with limited repetitions, the settlement did not show a significant increase with frequency. The box tests further revealed the effects of moisture content on foundation soils and were more representative of actual pavement foundations compared with triaxial tests.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
