A NMR-based Prediction Of Soil-water Characteristic Curve

Soil is widely distributed in nature, the vast majority of soil are not saturated in nature. With the development of science, society and technology. The study of unsaturated soil have become popular in the Rock and Soil Mechanics academia. For the study of unsaturated soil will inevitably come into contact with soil-water characteristic curve, The soil-water characteristic curve is very important in the study of the basis for the mechanical theory of non-saturated soils. Not only this, The soil-water characteristic curve have a very important position in the study of the engineering properties of unsaturated soil. The soil-water characteristic curve measuring need to consume a lot of time, and the test in laboratory can only be obtained in a particular state of the soil-water characteristic curve. However, when the study of unsaturated soil, it is usually required under any deformation conditions of the soil-water characteristic curve. So it is very necessary to proposed a method that could forecasting soil-water characteristic curve fast, efficient and economic.The main conclusions of the paper are following:(1) Select the soil in Wuhan Han-Street to take the research, experiments mainly include: soil basic physical indicators taken, sample preparation, measuring the soil-water characteristic curve of different degree of compaction, measuring the nuclear magnetic resonance curve of different degree of compaction.(2) Research and analysis the factors of soil-water characteristic curve. Verified that the different degree of compaction of the soil-water characteristic curve law.(3) Introduce and analysis a variety of classic soil-water characteristic curve models. Using Matlab and other analysis software to study the experimental data. Analyzed the advantages and disadvantages and applicability of each models. Put forward opinions of each models and select the optimal model for subsequent studies.(4) Study of the NMR experimental data under different degree of compaction. Verified the initial dry density of the sample that primarily affects the content of the larger diameter pores, and affects the content of the small diameter pores is secondary. Found that there is a correlation between Nuclear magnetic resonance curve and soilwater characteristic curve. Because they are similar after conversion. At last, the author suggest a mathematical model that by nuclear magnetic resonance experiments to predict soil-water characteristic curve. It is provides a new way of research for the soilwater characteristic curve.(5) Introduce the theory of fractal, Verified that transverse relaxation time and the quality of the water content has obvious fractal behavior. It reflects the soil pore distribution has a strong characteristic of fractal. Suggest a fractal model of nuclear magnetic resonance curve that represented by the mass water content. Proposed a method that the initial dry density is known and using this model to predict soil-water characteristic curve.