The intensified climate warming has led to frequent settlement problems in permafrost regions, posing a serious threat to the construction and safe operation of subgrades, which require strict control of settlement deformation. However, previous studies have insufficient understanding of the long-term deformation mechanisms of frozen soil-rock mixture (FSRM) used as subgrade fill materials, directly limiting the accuracy of deformation prediction for engineering projects. In this study, uniaxial creep tests were conducted to investigate the creep characteristics of FSRM under different temperatures (-8, -10, -12 degrees C), block sizes and gradations. It was found that at -8 degrees C, smaller block sizes under the same gradations resulted in shorter cumulative creep times, whereas the opposite trend was observed at -10 degrees C and - 12 degrees C, highlighting the regulatory effect of temperature on creep behavior. On this basis, the study attempts to introduce resistivity monitoring technology to analyze and quantify the evolution of creep damage. Due to the strong correlation between resistivity and creep strain, an innovative nonlinear viscoelastic plastic creep damage model based on resistivity was developed. Finally, the intrinsic mechanism of temperature's control over creep characteristics of FSRM was revealed to be the variation in ice crystal quantity, and the creep settlement mechanism of subgrades in cold regions under the background of climate warming and increased humidity was explored. This study provides novel insights into the damage representation and characterization of accelerated creep behavior in frozen geotechnical materials, aiming to provide theoretical support for creep settlement prediction and disaster prevention in cold-region subgrades and other infrastructure.

The cumulative plastic deformation and damage evolution of frozen soil-rock mixtures under cyclic loading was studied by a dynamic triaxial instrument with real-time resistivity measurement function. A series of low- temperature cyclic triaxial tests were conducted under varying confining pressures (200 kPa, 500 kPa, 800 kPa), block proportions (0, 30 %, 40 %, 50 %), and dynamic stress ratios (0.4, 0.6, 0.8). The results reveal that the cumulative plastic deformation process can be divided into three stages, such as microcrack closure as the initial stage, crack steady growth as the middle stage, and rapid crack propagation until it fails as the final stage. Under the same number of cycles, the greater the dynamic stress is, the greater the cumulative plastic deformation is. Furthermore, a strong correlation is identified between the resistivity and the cumulative plastic deformation. With the increase of the number of cycles, the cumulative plastic deformation leads to the accumulation of internal damage, and the resistivity gradually increases. Thus, a damage evolution model based on resistivity damage variables is proposed. The model demonstrated an average fitting accuracy of 97.36 % with the experimental data.

Machine learning is extensively used in constitutive modeling of geomaterials and is critical for enhancing computational analyses in geotechnical engineering. In the area of geomechanical modeling, the primary application of machine learning techniques is to use discrete data points for model training and testing. However, this methodology frequently fails to accurately capture the stress-strain behavior of geomaterials because of imbalanced training data sets, which undermines model performance and accuracy. In addition, the limited interpretability of these models hampers their practical application and understanding. To address these shortcomings, a machine learning framework specifically developed for the analysis of frozen soil-rock mixtures is introduced. The proposed model uses an autoencoder to extract features directly from individual stress-strain curves. The extracted features serve as training outputs for the multilayer perceptron, and the test conditions are employed as inputs to train the model. The decoding function of the autoencoder is employed to reconstruct the stress-strain data. The average fit between the model predictions and the actual results exceeds 93%. Shapley additive explanations are further incorporated to quantify the impact of the test conditions, revealing that temperature exerts the most substantial influence, followed by block proportion and confining pressure. Notably, an analysis reveals that the proportion of block significantly affects postpeak stress-strain responses, highlighting its comparable importance to temperature. This study identifies key test conditions influencing the mechanical behavior of frozen soil-rock mixtures, providing geotechnical engineers and researchers with a reference tool for predicting their properties.

The China-Nepal transportation corridor, a vital link, includes the western route of the China-Nepal Highway from Lhasa to the Jilong Border. This route has faced persistent engineering disturbance hazards since its construction in 1965, significantly compromising its safety. The planned China-Nepal Railway is set to be built largely in parallel to the existing highway. Therefore, a systematic study of the existing types, characteristics, triggering factors, and susceptibility zones of engineering disturbance hazards along the China-Nepal Highway can provide valuable insights for mitigating such hazards in future railway construction. Field investigation along the highway have cataloged over 160 hazard sites, classified into six main types: earth slides (14 cases), debris-bedrock binary slides (20 cases), earth-rock mixture slides (37 cases), earth-rock mixture falls (31 cases), rock falls (46 cases), and rock topples (13 cases). The susceptibility prediction model has been refined by applying the First Law of Geography to the hazards identified in field investigations. A comparison with ANN and Simple CNN models demonstrates that the proposed sample optimization strategy effectively enhances model performance. This strategy addresses the limitation in susceptibility prediction where point-format hazards fail to adequately represent the environmental characteristics of the entire hazard body. Results indicate two areas of extremely high susceptibility to engineering disturbance hazards along the China-Nepal Highway. The first area features rocky slopes with unloading fractures in broad river valleys, particularly where slate is exposed. The second area includes slopes of Quaternary loose deposits that have been disturbed by slope excavation in alpine valleys with heavy rainfall.

Recent advancements have seen a pervasive application of machine learning methodologies in assessing the susceptibility of geological hazards. A pivotal element influencing the accuracy of model predictions resides in the prudent selection of model parameters within machine learning frameworks. The objective of this study is to develop a robust landslide susceptibility assessment model by refining the support vector machine (SVM) model through the employment of the Bayesian algorithm for hyperparameter optimization. The southern part of the Qinghai-Tibet Plateau, focusing on major highways, is selected as the study area. Nine influencing factors, namely the elevation, slope, aspect, profile curvature, lithology, topographic wetness index, normalized difference vegetation index, distance to faults, and distance to rivers, are selected as the conditioning variables instrumental in evaluating the likelihood of collapse occurrences. Secondly, data from field surveys involving 351 landslides and randomly generated non-landslide data are utilized in a balanced 1:1 ratio to construct the training and testing datasets. Next, the cross-validation loss rate of the SVM model is selected as the objective function, and the Bayesian algorithm is used to optimize the BoxConstraint and KernelScale parameters of the SVM model, resulting in a Bayesian optimization-based SVM model. The results show that, within a five-fold cross-validation framework, the model yields 99.15

Expansive soil, characterized by widespread fissures, is a special type of soil prone to localized fissure propagation, leading to failure and instability, often resulting in landslides. Numerous studies have applied fracture mechanics theory to analyze soil failure along fissures, but no standardized testing method has been established. This paper reviews existing soil fracture toughness testing methods, designs an integrated system combining digital image correlation(DIC) technique and electrical resistance testing, and employs the Cracked ChevronNotched Brazilian Disc (CCNBD) method to assess the fracture toughness of expansive soil. The deformation and internal damage accumulation processes of the specimens were monitored. The K & Iukcy;c and K & Iukcy;& Iukcy;c of expansive soil were tested at moisture contents of 15 %, 20 %, and 25 % using specimens with three different crack length ratios (a/R). The role of water was explored by injecting water into the fissures. The results showed that K & Iukcy;c of expansive soil ranged from 10 to 36 kPa center dot m0.5, with a linear negative correlation to moisture content and a proportional relationship to tensile strength, having a proportionality coefficient of 0.331. The K & Iukcy;& Iukcy;c ranged from 20 to 70 kPa center dot m0.5,and it is greater than K & Iukcy;c under the same conditions. Based on the load, internal damage, strain, and fissure area during the tests, the failure process of expansive soil along fissures was divided into four stages: initial deformation stage(I), quasi-elastic deformation stage(II), fissure extension stage(III), and failure stage(IV). Water injection into the fissures reduced the soil's fracture toughness, with a more significant reduction as a/R increased and the failure showed progressive behavior. The CCNBD method, combined with DIC technique and electrical resistance testing, effectively measures the fracture toughness of soil, aiding in understanding the failure mechanism along fissures and providing a basis for preventing and controlling landslides and other hazards in expansive soils.

Expansive soils are widely distributed and the landslides happened in these area cause significant economic losses. The sliding surface records crucial information about landslides, providing insight into the mechanisms behind their occurrences. This research delves into the microstructural features of landslide surfaces in expansive soils. Various sliding surfaces were created through direct shear tests under different water contents (10%, 20%, 30%) and normal stresses (50 kPa, 100 kPa, 200 kPa, 400 Pa). Scanning electron microscopy (SEM) was used to observe the surfaces and based on their structural morphology and formation mechanisms, we categorized the surfaces’ structure into scratch, tension crack, and residual pore. Four parameters, porosity, average pore area, probability entropy, and area probability distribution index, were calculated and used to quantitatively characterized the surface structures. By correlating field landslide surface features with geological survey data, it was inferred that the triggering moisture content was 25.3–26.8% at least, the sliding surface was situated 10 to 12 meters deep within the slope, and multiple sliding events occurred. The water infiltration from the slope top and precipitation along fractures coupled with rising irrigation water levels that saturated the slope base ultimately leading to fracture layer breakthrough and slope failure.

The strength damage and deformation failure of frozen soil–rock mixture (FSRM) often restrict the safety of the major engineering construction in cold areas or the spatial development of urban underground water-rich rock and soil masses. To investigate the uniaxial strength damage evolution and failure characteristics of FSRM under different loading rates (0.3, 0.6, 3, 6, 30, and 60 mm·min −1 ) in the quasi-static range, resistivity monitoring and image recognition technology were used to study the time-stress-volumetric strain-resistivity changes. The results indicate that the peak stress, peak strain, initial yield modulus, and tangential modulus of FSRM increase rapidly before increasing slowly as the loading rate increases, and there are critical loading rates and post-peak failure phenomenon. Three distinct types of failure modes, bulge failure, oblique shear failure, and fragmentation failure were observed at low (0.3–0.6 mm·min −1 ), medium (3–6 mm·min −1 ), and high loading rates (30–60 mm·min −1 ), respectively. The macroscopic failure of the FSRM at different loading rates arises from a combination of strain rate hardening of the strength and damage softening of the structure. To predict the stress-strain characteristics at various loading rates, a damage prediction model with a damage variable correction factor considering residual strength was employed based on the improved Duncan–Chang model and damage theory of electrical resistivity, and the predicted results were in good agreement with the experimental data.

The variation of ultrasonic parameters is closely linked to the mechanical properties and damage evolution of rock and soil mass. In this paper, uniaxial compression tests and real-time ultrasonic monitoring technology were used to explore the strength, deformation and damage characteristics of frozen soil-rock mixture (FSRM) with different block sizes and gradations, as well as the law of ultrasonic wave propagation. The results indicate that: (1) A wider gradation of rock blocks corresponds to a higher specimen strength and a lower breakage degree of rock blocks. Within the same gradation, specimens with smaller block sizes have higher strength. Wider gradation and smaller particle size of rock blocks exerts a delayed effect on damage and failure of FSRM. (2) The particle size of rock blocks plays a key role in the variation of ultrasonic parameters. Within the same gradation, the specimens with larger block sizes possess higher wave velocity and lower first wave amplitude. The effect of gradation on ultrasonic parameters is attributed to the variation of block size. (3) With the first wave amplitude as damage state variable, a damage prediction model based on the improved Duncan–Chang model is established, demonstrating superior prediction potential on the stress–strain curves of FSRM. The research confirms and promotes the quantitative correlation between ultrasonic parameters and mechanical properties of geotechnical materials, which may provide theoretical support for testing and evaluating the mechanical properties of roadbed filling in cold mountainous areas.

GHIDSS is studied and developed on the platform of windows by means of VC and VB language. It is a all Chinese interface geological hazards computer intelligence decision system, which has a friendly user interface and ability of space analysis. The system consists of five subsystem of information resources database, Map and Image base (GIS) auxiliary analysis, modelling and prediction, economic evaluation and expert system of policy decision analysis for prevention and treatment and modules of core exchange and user interface. The idea of its overall design and the design of function modules of the system fully embodied the consideration of "four bases in one", i.e. to incorporate the quantitative calculation using data and models, and the qualitative analysis using knowledge and inference, and make them infiltrate into each other, to combine the geographical information system (GIS) and intelligence decision support system (IDSS) in order to give GHIDSS the ability of richer and more figurative space analysis. By use of this system we can predict the evolution of geological hazards in time and space, delimit dangerous area, evaluate economically the hazards and choose the suitable countermeasure for prevention and reduction of disasters.

It is an important way to accurately grasp the development mechanism and image characteristics of cracks and deduce the formation process of expansive landslide. In this paper, the expansive soil slope in Pishihang irrigation area of Anhui Province was taken as the research object. The in-situ test under the condition of rainfall plus dead weight was carried out to systematically study the image characteristics of soil surface cracks during the instability process. The emphasis was on the analysis of the relationship between the variation of crack degree, crack geometric characteristic parameters and slope instability. The results show that it is feasible to evaluate the slope risk through comprehensive analysis of soil surface crack images. In view of the expansive soil slope in this area, when the transverse cracks along the soft structural plane develop into the main control cracks, the slope would appear abnormal, and the early warning should be put forward.

River floods, dammed lake flood discharge, reservoir discharge, seawater recession, etc. all cause the water level in front of a slope to drop, which changes the original steady-state seepage field in the soil, leading to harmful slope instability. To study this phenomenon, a numerical model was established through theoretical analysis combined with the coupling of the Seep/W and Slope/W modules of the GeoStudio finite element software, and the numerical model was verified by the model test results of indoor medium sand and silt. This paper focuses on the effects when the water level in front of a slope drops at different speeds, different drop ratios, different initial water levels, different filling materials, and matrix suction on the seepage field and slope stability. The conclusions are as follows: (1) the greater the speed at which the water level in front of a slope falls, the greater the downward seepage force formed by the seepage field of the slope to the slope body; (2) the change curve of the safety factor at a higher speed is steeper when the water level falls at different speeds, and the safety factor value when the water level in front of the slope is constant is smaller; (3) the safety factor of the slope decreases with an increase in the drop ratio; when the drop ratio is the same, the loss of stability is worse if the initial water level is lower; (4) when there is a drawdown of water levels in front of the slope, the non-cohesive medium sand slope is more prone to instability failure than the cohesive silt slope; and (5) when this modeling method is applied to matrix suction, the effect of matrix suction increases the safety factor of the slope.

As a new rock breaking method, CO2 transient cracking has been widely used in rock excavation projects in recent years. However, in the actual construction process, there are often situations where the fracturing effect varies due to different rock mass structures. Through theoretical analysis and on-site cracking tests, this article studies the effect of CO2 transient cracking under the control of different rock mass structures. The results show that: (1) the dynamic compressive strength of rock directly determines the number and range of dynamic impact fractures; the original fractures of rock mass and those caused by dynamic impact in the first stage jointly determine the effect of high-pressure gas expansion in the second stage. (2) The arrangement of holes along the strata is conducive to the action of high-pressure expanding gas along the soft structural plane in the rock mass, which is conducive to the fracturing of the rock mass; the amount of crack formation is small, but the influence range is large. (3) The cracking effect of carbon dioxide transient cracking applied to massive rock mass is better than that of monolithic rock mass, while the cracking effect of layered rock mass with soil interlayer is poor. The research results are of great significance for improving the effectiveness of carbon dioxide transient-induced cracking excavation and guiding actual construction.

In practical engineering applications, silt is prone to liquefaction and quicksand. This paper mainly studies the improvement effects of urease, lignin and their mixture on the strength and liquefaction resistance of silt. Based on the results and phenomena of an unconfined compressive strength and dynamic triaxial test, the improvement effects of the compressive strength, deformation resistance and liquefaction resistance of silt under different improvement schemes are analyzed, and the optimal values of the cement or lignin when enzyme-induced calcium carbonate precipitation (EICP) technology, lignin alone or EICP and lignin are obtained. The results show that the optimum concentration of the constant temperature and humidity sample (referred to as the constant humidity sample) and the constant temperature immersion sample (referred to as the soaking sample) of urease in the unconfined compressive strength test is 1.0 mol/L, and the compressive strength of the soaking sample is 4.9 MPa, which is 1.56 times that of the constant humidity sample; the optimum addition ratio of the lignin-improved constant humidity sample is 3%, and its compressive strength is 2.07 Mpa; the optimum addition ratio of the samples immersed at constant temperature is 4%, and the compressive strength is 3.05 MPa; when urease combines with lignin to improve silt, 4% is the best lignin addition ratio, the compressive strength of the constant humidity sample reaches 1.57 Mpa and the compressive strength of the soaking sample reaches 3.75 MPa; in the dynamic triaxial multi-stage cyclic load test, all samples were cured at constant humidity sample, and in the urease modified silt scheme, 1.0 mol/L was the optimal cement concentration; in the scheme of improving silt with lignin, 3% is the optimal addition ratio; when 1.25 mol/L cementation solution plus urease crude extract is combined with different ratios of lignin in the experimental scheme, 3% is the best lignin addition ratio.

The slope angle of bedrock surface plays an important role in deformation and failure of the overlying rock-soil mixture（debris）. Therefore, it is of great importance to quantify the effect of bedrock slope on the evolution of landslides in order to obtain a comprehensive evaluation of the mechanical behavior and stability of landslides; however, such quantitative studies are currently lacking. To address this gap we carried out an experimental study on deformation and failure of rock-soil mixture under different slope angles of the underlying bedrock using a home-built large size push-shear apparatus. The results show that with increasing slope angle the included angle between the shearing and bedrock surfaces increases, and so does the distance between the two surfaces. At 25° inclination the overall movement of the rock-soil mixture in the shear zone is most stable; whereas at 45° inclination occurs the most serious debris rupture in the shear zone while the overall movement of debris is most unstable, and the slip zone is most difficult to manage. With increasing slope angle, the maximum shear stress first increases and then starts to decrease at 45° inclination—at this point, the maximum shear stress reaches maximum in the deformation process, and the peaking time of pore-water/soil pressure changes（from their initial values） reach maximum as well. The results and parameters obtained in this study provide a reference basis for evaluating landslide stability and other geological hazards based on the determination of bedrock slope.

Expansive soil is one of the most widely distributed special soils in the world. It is widely developed in Henan, Anhui, Guangxi and other places in China, and highly overlaps with densely populated and economically active areas. Expansive soil is considered a typical “problematic soil” because its mechanical behaviour is very sensitive to water content changes; such behaviour mainly manifests as swelling upon wetting and shrinking upon drying, so the presence of expansive soil is an important factor in mountain landslide disasters in southern China. Because the particularities of its constituent materials are related to typical physical and mechanical properties, forecasting the failure times of expansive soil slopes remains a global problem. In this study, a series of in situ artificial rainfall experiments were conducted on an excavated expansive soil slope; then, the digital image correlation (DIC) method was applied to monitor the slope surface deformation and crack development. Finally, the failure time of the slope was forecasted using the inverse velocity (INV) and slope (SLO) models. The study results show that the deformation and failure processes of the analysed expansive soil slope had an obvious crack control effect, and the displacement–time curve derived by the DIC method had an obvious “phased change law”. The data points calculated by the INV method were discrete and had high linear fitting requirements, resulting in large failure time forecasts. When the SLO method was used to forecast the failure time, because the values derived in the stable deformation stage were relatively concentrated in the calculation process, an obvious linear relationship was found in only the accelerated deformation stage, so the prediction results were more accurate. Therefore, the SLO method should be preferentially used to forecast the failure of expansive soil slopes with “step-like” displacement. These results enabled us to characterize slide processes and identify the mechanism responsible for the movement of a rainfall-induced expansive soil landslide. The stage deformation and failure mode of expansive soil landslide under rainfall infiltration: “slow deformation—stable deformation—accelerated deformation—instability failure” was revealed. This study is helpful for determining the deformation and failure mechanism of rainfall-induced expansive soil landslide and forecasting expansive soil landslides and providing guidance for controlling landslide hazards in expansive soil areas.

Due to rainfall infiltration, slope instability becomes frequent, which is the main reason for landslide disasters. In this study, the stability of slope affected by rainfall was analyzed using an indoor model test and geo-studio simulation method, and the variation law of phreatic line, seepage field, the most dangerous sliding surface, and safety factor with time were studied under rainfall infiltration. Research results showed that under the effect of rainfall, the slope failure presented a typical traction development mode. With the increase of time, the phreatic line of the slope kept rising, the water head keeps increasing, the seepage depth in the slope became deeper, and the slope stability worsened until the slope was damaged. The water head height decreased gradually from the slope left boundary to the right, and the water head width decreased gradually. The soil at the slope back edge was damaged, and the sliding soil accumulated at the slope foot, forming a gentle slope, which increased the shear strength of the slope, making the slope finally reach a stable state. In this process, the overlying soil changed from an unsaturated state to a saturated state, the pore water pressure and soil pressure increased, and then the slope was damaged, both of which decreased. Under high rainfall intensity, the slope was damaged, the soil in the slope was rapidly saturated, and the time required to produce the sliding area was short. When the rainfall intensity was the same, the smaller the slope angle was, the smaller the safety factor was. When the slope angle was the same, the greater the rainfall intensity was, the smaller the safety factor was.

Quantitative risk assessment of landslides has always been the focus and difficulty in the field of landslide research. In this paper, taking Mayang County, Hunan Province as an example, the risk assessment of rainfall-induced landslides was carried out from the regional and individual scales. On the regional scale, the risk factors of geological disasters were analyzed. Based on the slope unit, the risk analysis of slope geological disasters and the vulnerability risk assessment of hazard-bearing bodies were carried out to form the block plan. On an individual scale, based on the analysis of rainfall extreme value, the variation law of landslide seepage field and stability under different rainfall recurrence periods was simulated. Then, the vulnerability of the disaster-bearing body was studied according to the analysis of the impact range and the field investigation. Combined with the evaluation results of landslide hazard and vulnerability of the disaster-bearing body, the life and economic risks under different working conditions were further obtained. Therefore, the research results could provide not only a reference for the risk assessment of rainfall-induced landslides in other regions but also a theoretical basis for the early warning and prediction of landslide disasters.

In order to determine the applicability of liquid CO2 phase-transition fracturing technology in rock mass excavations, the principles of CO2 phase-transition fracturing were analyzed, and field tests of liquid CO2 phase-transition fracturing were performed. An “Unmanned Aerial Vehicle (UAV) camera shooting + Microstructure Image Processing System (MIPS) analyzing” method was used to acquire the rock mass characteristics. Further, the Hilbert–Huang Transform (HHT) energy analysis principle was adopted to analyze the characteristics of fracturing vibration waves. The experimental results showed that during the process of fracturing, there were both dynamic actions of rock breakage due to excitation stress wave impacts, and quasi-static actions of rock breakage caused by gasification expansion wedges. In semi-infinite spaces, rock-breakage zones can mainly be divided into crushing zones, fracture zones, and vibration zones. At the same time, under ideal fracturing effects and large volumes, the fracturing granularity will be in accordance with the fractal laws. For example, the larger the fractal dimensions, the higher the proportion of small fragments, and vice versa. Moreover, the vibration waves of the liquid CO2 phase-transition fracturing have short durations, fast attenuation, and fewer high-frequency components. The dominant frequency band of energy will range between 0 and 20 Hz. The liquid CO2 phase-transition fracturing technology has been observed to overcome the shortcomings of traditional explosive blasting methods and can be applied to a variety of rock types. It is a safe and efficient method for rock-breaking excavations; therefore, the above technology effectively provides a new method for the follow-up of similar engineering practices.