| The frozen ground area of China rankes third in the world.In high and cold regions,frozen soil has a strong impact on the long-term stability of engineering structures.With the global warming and the increasing frequency of extreme climate events,engineering problems caused by the freezing-thawing instability of soil will become more and more prominent.The premise to solve this problem lies in mastering characteristics of frost heave and thaw settlement of soil.Based on the research and development of temperature control system with adaptive control ability,this paper designed a series of new freezing-thawing test devices,constructed a multi-field coupling model,proposed a new engineering structure suitable for the shallow instability protection of frozen soil slopes and carried out model tests.The main achievements and conclusions are as follows:1.The temperature loss and hysteresis during the variable temperature adjustment,which is the main problem of the mechanical refrigerating system,have been solved by the solid-state refrigeration module using temperature control algorithm combing neural network with PID algorithm.The SN-PID(Single Neuron)algorithm has the advantages of simple structure and strong robustness,and its control accuracy reaches ± 0.002 ℃ in the state of constant temperature control,which can meet the temperature control requirements of frozen soil tests.This provides a prerequisite for the research and development of new test equipments of frozen soils.On this basis,the linear cooling mode was combined with the ultrasonic excitation system to solve the problems of "size effect" and "hypercooling" in the freezing temperature test,and the soil microscopic pore characteristics were obtained by the particle accumulation morphology algorithm,called the FV(Fourier-Voronoi)algorithm.The test equipment of soil porosity rate was developed by combining the solid-state cooling module with the high-precision displacement sensing system.A data smoothing algorithm using a roughness penalty was presented.This provided the conditions for studying the hysteretic response characteristics of porosity rate during freezing and thawing process.An binarization method,called the ACB algorithm,was proposed,which determined the binarization threshold by combining the local and global gray intensity information.This algorithm was used to process the imagines obtained by the linear CCD scanning imaging module to microscopicly identify the pore ice during the freezing and thawing test of soil.2.The key equation of the multi-field coupling model based on the water-heat equilibrium at the microscopic ice-water phase transition interface is the Clapeyron equation or modified Clapeyron equation.The surface radius of the ice-water phase transition interface is an essential parameter for these equations.But as a highly heterogeneous porous medium,the microscopic pore characteristics of soil are very complex,and it is difficult to obtain effective pore radius to describe the overall ice-water interface.Therefore,the Clapeyron equation may fail to describe the hydrothermal equilibrium of the ice-water interface.From the perspective of soil freezing and thawing process,the formation and melt of pore ice is the key to the multi-field coupling of frozen soil,which is macroscopically manifested as frost heave and thaw settlement of soils and microscopically affects the pore geometry.This process can be quantified by the change rate of porosity.Therefore,the addition of porosity rate function as the coupling medium in multi-field coupling model can improve the coupling effect and simplify the calculation process of the model.On the basis of the porosity rate test,this paper proposed a macro-multifield coupling model using the porosity rate function,and the finite element model was developed using ABAQUS subroutines which could identify the freezing state and calculate parameters of porosity rate curve of elements automaticly.This multifield coupling model was used to simulate the porosity rate test,the freezingthawing test of soil column and the model test of frozen soil slope.Results showed that this model had a high computational efficiency and good convergence.3.The thermal anchor-frame protection structure can effectively prevent the shallow instability of frozen soil slope.Compared with the thermosiphon,the thermal anchor conducted heat through the anchor body rather than the heat transfer medium,which simplifies the anchor structure and makes it suitable for the protection of shallow instability of frozen soil.Under the influence of thermal anchors,soil can be accelerated to freeze,thus reducing the frost heave and thawing settlement.The protection structure of gravel backfill has the effect of insulation on the surface.Under the protection this structure,the penetration of frozen front and the depth of soil affected by freeze-thaw decrease.Although the protection structure of interlocking block does not have the ability of temperature control,it can adapt to the large deformation of soil during freezing-thawing.The model with the protection structure of interlocking block is still in a stable state when shallow slump occurs in the control group(bare slope)in the slope model test. |