Cold Regions Science and Technology最新文献

{"title":"Experimental investigation on dynamic characteristics of hydrophobic modified silt after freezing-thawing under cyclic loading","authors":"Shang Gao ,&nbsp;Xinzhuang Cui ,&nbsp;Xiongying Ma ,&nbsp;Qing Jin ,&nbsp;Xiaoning Zhang ,&nbsp;Hao Zeng","doi":"10.1016/j.coldregions.2025.104606","DOIUrl":"10.1016/j.coldregions.2025.104606","url":null,"abstract":"<div><div>Silt subgrades in seasonally frozen regions are subjected to progressive deterioration under the combined influence of freeze-thaw cycles and traffic loads, resulting in pavement cracking and uneven settlement. This study introduces nano-type hydrophobic material (NT-HM) into silt to enhance silt freeze-thaw resistance. The dynamic stability and freeze-thaw resistance of NT-HM modified silt were studied by dynamic triaxial tests. Based on the shakedown theory, the critical dynamic stress equation of hydrophobic silt under the shakedown limit state is established. The experimental results show that NT-HM wraps the surface of silt particles to form a covalent bond structure to give them hydrophobicity. When the content of NT-HM in silt reaches 0.5 %, the hydrophobic performance reaches super hydrophobic state. After 7 freeze-thaw cycles, the contact angle of the silt with 0.5 % NT-HM content surface only decreased by 2.3 %. The modification effect provided by NT-HM and the confinement effect provided by confining pressure have a coupled superposition effect. Moreover, the addition of NT-HM to silt reduces the sensitivity of the accumulated plastic strain to the cyclic stress amplitude. After seven freeze-thaw cycles, the cumulative plastic strain of the silt with 0.5 % NT-HM content was reduced by 47.7 %–53.1 % compared with the unmodified silt. Hydrophobic modification effectively extends both the plastic shakedown and plastic creep boundaries, with increases of approximately 1.7-fold and 1.4-fold, respectively, compared to the unmodified silt. A cumulative plastic strain model considering the number of freeze-thaw cycles, NT-HM content and stress state was established, which can accurately reproduce the test results. The silt with 0.5 % NT-HM content can still maintain a good skeleton structure after 7 freeze-thaw cycles. This effectively slows down the silt volume changes during the freeze-thaw cycle. This study providing a theoretical basis for hydrophobic material application in seasonally frozen road engineering.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104606"},"PeriodicalIF":3.8,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144604996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Influences of sub-zero temperature on dynamic tensile behavior of rhyolite porphyry","authors":"Zilong Zhou ,&nbsp;Fanjunhui Mo ,&nbsp;Xin Cai ,&nbsp;Chu Wang ,&nbsp;Chunping Lin ,&nbsp;Yuanping Lai ,&nbsp;Shaohui Tang ,&nbsp;Zhongkang Wang","doi":"10.1016/j.coldregions.2025.104604","DOIUrl":"10.1016/j.coldregions.2025.104604","url":null,"abstract":"<div><div>The mechanical behavior of rock materials in cold regions undergoes significant alterations under sub-zero temperatures. To understand the dynamic tensile properties of rhyolite porphyry (RP) under cryogenic conditions, a series of dynamic Brazilian disc tests were carried out on RP specimens within a temperature range from 10 °C to −40 °C covering a wide range of loading rate utilizing a split Hopkinson pressure bar. The test results demonstrate that, under sub-zero temperature conditions, the dynamic tensile strength (DTS) of RP is dependent on loading rate, increasing exponentially with loading rate across all temperatures. Additionally, at a given loading rate, the DTS initially increases and subsequently declines with decreasing temperature. Scanning electronic microscope (SEM) analysis reveals that the extreme cooling (−20 °C to −40 °C) induces inconsistent mineral shrinkage, generating microcracks that degrade macroscopic strength. The skeleton contraction stress theory is incorporated to computationally quantify freezing-induced contraction stress among distinct mineral constituents within RP specimens, theoretically confirming significant contraction stress differentials between compositional phases. Furthermore, a predictive model for rock strength, integrating temperature and loading rate effects, was developed through response surface methodology. The predicted values from this model showed good agreement with the experimental data, indicating its reliability for rock strength prediction.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104604"},"PeriodicalIF":3.8,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144597123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on the mechanism and theoretical model of water vapor freezing in unsaturated soil subgrade","authors":"Haojin Zhang ,&nbsp;Haihua Zhang ,&nbsp;Xianfeng Ma ,&nbsp;Jiangu Qian ,&nbsp;Guanlin Ye ,&nbsp;Huibo Zhang","doi":"10.1016/j.coldregions.2025.104597","DOIUrl":"10.1016/j.coldregions.2025.104597","url":null,"abstract":"<div><div>In seasonally frozen soil regions, changes in relative humidity (RH) within the pores of unsaturated soil subgrade induce the migration and freezing of water vapor (water in the form of a gas resulting from heating water or ice). This phenomenon significantly affects the deformation and strength of the soil. Investigating the freezing characteristics of water vapor in unsaturated soil subgrade and establishing mathematical models are essential for uncovering the mechanisms behind structural disasters in these formations. This study addresses the unique four-phase nature of unsaturated frozen soils. A new formula for the surface tension of water vapor ice interface has been proposed, which solves the key blank of the interface assumption. This enables accurate simulation of water vapor freezing dynamics in unsaturated frozen soil. Modifying the Kelvin equation based on the model of variable pore diameter results in a more accurate representation of the relationship between changing porosity, pore volume, and equilibrium RH during freezing. Regression analysis is conducted using data from the International Association for the Properties of Water and Water vapor. The calculation formula for the saturated vapor pressure of frozen soil below zero degrees is revised. Dynamic water vapor freezing experiments and numerical simulations are performed on sandy soils with varying pore parameters (porosity and pore volume) to verify the necessity of the model of variable pore diameter in water vapor freezing theory. The results demonstrated that the amount of frozen water vapor increases with an increase in pore size and RH. This research primarily lays the foundation for the frost heave theory of water vapor in unsaturated frozen soils and provides a theoretical basis for further multifield coupling studies.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104597"},"PeriodicalIF":3.8,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144571317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic behaviour of geopolymer-based ultra-high-performance concrete at low and cryogenic temperature","authors":"Kaiyi Chi ,&nbsp;Jun Li ,&nbsp;Ruizhe Shao ,&nbsp;Chengqing Wu","doi":"10.1016/j.coldregions.2025.104600","DOIUrl":"10.1016/j.coldregions.2025.104600","url":null,"abstract":"<div><div>This study investigates the mechanical behaviour of geopolymer-based ultra-high performance concrete (GUHPC) under coupled extreme low temperatures (−160–20 °C) and dynamic loads. The research utilised a <em>Φ</em>100-mm split Hopkinson pressure bar (SHPB) technique for dynamic testing. Dynamic impact experiments were conducted at temperatures of 20, −70 and − 160 °C, with strain rates ranging from 40 to 160 s⁻¹ for compression tests and 20 to 80 s⁻¹ for splitting tensile tests. Results reveal that both compressive and splitting tensile strengths of GUHPC increase as temperature decreases and strain rate increases. The dynamic increase factor (DIF) for both compression (<span><math><msub><mi>DIF</mi><msubsup><mi>f</mi><mi>c</mi><mo>′</mo></msubsup></msub></math></span>) and tension (<span><math><msub><mi>DIF</mi><msub><mi>f</mi><mi>t</mi></msub></msub></math></span>) was found to be higher at low temperatures and cryogenic condition. The <span><math><msub><mi>DIF</mi><msubsup><mi>f</mi><mi>c</mi><mo>′</mo></msubsup></msub><mspace></mspace></math></span>values at −70 °C and − 160 °C were approximately 12 % and 22 % higher, respectively, as compared to the values at 20 °C with the strain rate of 160 s⁻¹. The <span><math><msub><mi>DIF</mi><msub><mi>f</mi><mi>t</mi></msub></msub></math></span> at −70 °C and − 160 °C were 35 % and 47 % higher, respectively, in comparison with the room temperature values at a strain rate of 40 s⁻¹. The study also examined the material microstructural change, failure modes, energy absorption capacity, and developed empirical formulae for predicting the DIF at various temperatures. X-ray computed tomography (CT) scans revealed distinct microstructural changes in GUHPC under various temperatures, illustrating ice formation, pore structure alterations and microcracking, particularly at cryogenic temperatures. These findings contribute to the understanding of GUHPC behaviour in extreme environments and have implications for its application in cryogenic structures subjected to impact loading.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104600"},"PeriodicalIF":3.8,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144557632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In-cloud ice accretion and performance degradation of UAV propellers in forward flight: An experimental study","authors":"Manaf Muhammed ,&nbsp;Derek Harvey ,&nbsp;Hassan Abbas Khawaja ,&nbsp;Muhammad Shakeel Virk ,&nbsp;Gelareh Momen","doi":"10.1016/j.coldregions.2025.104599","DOIUrl":"10.1016/j.coldregions.2025.104599","url":null,"abstract":"<div><div>An experimental study of ice accretion on a rotating UAV propeller blade was conducted under diverse environmental conditions. This research aimed to study the effects of varying independent environmental parameters on the characteristics and morphology of the accreted ice as well as its influence on propeller's performance. These experiments were carried out at the Anti-icing Materials International Laboratory (AMIL) Icing Wind Tunnel (IWT) at the Université du Québec à Chicoutimi (UQAC), Canada. The icing conditions are determined in accordance with the 14 CFR Part 29 Appendix-C for rotorcraft operating at altitudes below 10,000 ft. The analysis of results revealed that increase in Liquid Water Content (LWC) values can significantly affect the ice accretion rates. Higher LWC intensifies ice accretion, leading to a sharp initial drop in thrust and a rapid rise in power demand; however, it is observed that these variations gradually saturate after the initial aggressive degradation phase. Increasing Median Volume Diameter (MVD) can significantly affect the nature, morphology, and mass of accreted ice. The thin propeller sections were highly sensitive to increase in droplet size, leading to increase collection efficiencies. In some cases, an increase in MVD could trigger a transition in the ice accretion regime from rime to glaze ice. Also, the ice transitioned from soft rime to hard glaze as atmospheric temperatures approached the freezing point. Such transitions resulted in significative increase in the severity of the aerodynamic performance degradation. Elevated values of LWC and MVD at temperatures close to the freezing point led to the development of severe ice formations characterized by ice horns along the leading edge, intricate ice structures near the blade tip and fast degradation of aerodynamic performance. During ice accretion, thrust decreases linearly, while input power increases quadratically with RPM. The 3D scans of the final ice shapes obtained in this research not only offered detailed insights into the ice morphology but will also serve to validate numerical ice accretion models in future work. Performance penalties were notably more significant during the first 50 s of ice accretion, indicating a necessity for ice protection systems with low reaction times in rotary wing UAVs.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104599"},"PeriodicalIF":3.8,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144571316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Retrieving snow depth distribution by downscaling ERA5 Reanalysis with ICESat-2 laser altimetry","authors":"Zhihao Liu ,&nbsp;Simon Filhol ,&nbsp;Désirée Treichler","doi":"10.1016/j.coldregions.2025.104580","DOIUrl":"10.1016/j.coldregions.2025.104580","url":null,"abstract":"<div><div>Estimating the variability of snow depth in remote areas poses significant challenges due to limited spatial and temporal data availability. This study uses snow depth measurements from the ICESat-2 satellite laser altimeter, which are sparse in both space and time, and incorporates them with climate reanalysis data into a downscaling-calibration scheme to produce monthly gridded snow depth maps at microscale (10 m). Snow surface elevation measurements from ICESat-2 along profiles are compared to a digital elevation model to determine snow depth at each point. To efficiently turn sparse measurements into snow depth maps, a regression model is fitted to establish a relationship between the retrieved snow depth and the corresponding ERA5 Land snow depth. This relationship, referred to as subgrid variability, is then applied to downscale the monthly ERA5 Land snow depth data. The method can provide timeseries of monthly snow depth maps for the entire ERA5 time range (since 1950). We observe that the generic output should be calibrated by a small number of localized control points from a one-time field survey to reproduce the full snow depth patterns. Results show that snow depth prediction achieved a <span><math><mrow><mi>R</mi><mn>2</mn></mrow></math></span> model fit value of 0.81 (post-calibration) at an intermediate scale (100 m × 500 m) using datasets from airborne laser scanning (ALS) in the Hardangervidda region of southern Norway, with still good results at microscale (<span><math><mrow><mi>R</mi><mn>2</mn></mrow></math></span> 0.34, RMSE 1.28 m, post-calibration). Bias is greatest for extremes, with very high/low snow depths being under- and overestimated, respectively. Modeled snow depth time series at the site level have a slightly smaller RMSE than ERA5 Land data, but are still consistently biased compared to measurements from meteorological stations. Despite such localized bias and a tendency towards average snow depths the model reproduces the relative snow distribution pattern very accurately, both for peak snow (Spearman’s <span><math><mi>ρ</mi></math></span> 0.77) and patchy snow meltout in late spring (Matthews correlation coefficient 0.35). The method relies on globally available data and is applicable to other snow regions above the treeline. Though requiring area-specific calibration, our approach has the potential to provide snow depth maps in areas where no such data exist and can be used to extrapolate existing snow surveys in time and over larger areas. With this, it can offer valuable input data for hydrological, ecological or permafrost modeling tasks.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104580"},"PeriodicalIF":3.8,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144523900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Compaction mechanics of unsintered snow layers based on particle image velocimetry techniques","authors":"Hao Wang ,&nbsp;Enzhao Xiao ,&nbsp;Yihe Wang ,&nbsp;Shengquan Li ,&nbsp;Xueyuan Tang ,&nbsp;Bo Sun","doi":"10.1016/j.coldregions.2025.104598","DOIUrl":"10.1016/j.coldregions.2025.104598","url":null,"abstract":"<div><div>The knowledge of snow response under applied loads, particularly the deformation process during compaction, is important for avalanche modeling and cold region engineering. This paper is devoted to analyzing the deformation process of unsintered snow layers under compaction loads. An experiment framework that allows the observation of the unsintered snow deformation process on the symmetric plane of indentation is utilized along with the particle image velocimetry (PIV) technique. The experiment results show that during deformation, the snow particle velocity field can be categorized into four types: Type A, initial velocity field; Type B, vertical velocity field; Type C, shear band velocity field; and Type D, irregular velocity field. The shear band in loose snow beneath the compacted zone in Type C is the main difference between the deformation processes of snow samples with small and large initial depths, and it is also the key driver of the “egg shell” compaction effects in snow samples with large initial depths. The 150 mm depth sample exhibited the most regular compaction zone, with the Type B velocity field dominating for over 90 % of the duration. In contrast, the 300 mm depth sample displayed the most pronounced “egg shell” structure, with a 21.6 % density decrease between the surface and underlying layer. The findings in this study provide guidelines for developing construction methods and procedures for snow runways and roads, and can be utilized as a database for the validation of snow compaction numerical models.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104598"},"PeriodicalIF":3.8,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144597122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advances in icephobic coatings: Concepts, mechanisms, classifications and prospects","authors":"Yanlong Zhan ,&nbsp;Zhenqian Pang ,&nbsp;Gang Tan","doi":"10.1016/j.coldregions.2025.104596","DOIUrl":"10.1016/j.coldregions.2025.104596","url":null,"abstract":"<div><div>Ice accumulation on critical infrastructure, such as power grids, wind turbines, aircraft, and offshore platforms, poses significant safety risks and economic challenges. Icephobic coatings, designed to reduce ice adhesion and enable natural ice removal, offer an effective solution to mitigate these hazards. This review examines the dangers of surface icing, evaluates current anti/de-icing methods, and explores the mechanisms of icing and ice-material interactions. Key topics include various icephobic coating strategies, such as low surface energy coatings, superhydrophobic coatings, slippery liquid-infused porous surfaces (SLIPS), sacrificial coatings, polyelectrolyte brush coatings, low-modulus elastomer coatings, stress localization coatings, low interfacial toughness coatings, and antifreeze proteins (AFPs) coatings, and their effectiveness and mechanisms. Methods for measuring ice adhesion and the role of computational simulations in advancing icephobic materials are also discussed. The review concludes with insights into future research directions for icephobic coatings.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104596"},"PeriodicalIF":3.8,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144523968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Numerical simulation of water migration in saturated soft clay induced by horizontal freezing","authors":"Hu Zhang ,&nbsp;Yu Liang ,&nbsp;Jintao Hu ,&nbsp;Shengjin Zhang ,&nbsp;Lijun Xing ,&nbsp;Zhe Jiang ,&nbsp;Yuxuan Dong ,&nbsp;Chongyi Liang ,&nbsp;Bo Zheng","doi":"10.1016/j.coldregions.2025.104595","DOIUrl":"10.1016/j.coldregions.2025.104595","url":null,"abstract":"<div><div>The application of artificial freezing methods in soft clay foundations faces a series of complex issues, especially in horizontal freezing processes, where the thermo-hydro-mechanical coupling mechanisms of soft clay are not well understood, leading to difficulties in predicting and controlling the related freezing processes. Investigating the thermo-hydro-mechanical coupling mechanisms of soft clay under horizontal freezing conditions is crucial for enhancing the reliability and effectiveness of freezing treatment techniques in soft clay foundations. This study formulates thermo-hydro-mechanical coupling control equations for saturated soft clay under horizontal freezing, grounded in energy and mass conservation principles, and validates the model's accuracy through laboratory experiments. The findings indicate that the model effectively characterizes the interactions among water migration, temperature distribution, and stress field variations during horizontal freezing. The study reveals the driving effect of temperature potential energy and gravitational potential energy on water migration, with temperature-induced changes dominating the process. Additionally, temperature variations significantly affect key parameters such as pore- water pressure, freezing front progression, and unfrozen water content. The research provides theoretical support for predicting the freezing process of soft clay foundations in horizontal freezing engineering, offering guidance for optimizing freezing construction techniques.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104595"},"PeriodicalIF":3.8,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modeling axial strain-deviatoric stress response of frozen sands with enhanced LSTM approach","authors":"Xinye Song ,&nbsp;Sai K. Vanapalli ,&nbsp;Junping Ren","doi":"10.1016/j.coldregions.2025.104594","DOIUrl":"10.1016/j.coldregions.2025.104594","url":null,"abstract":"<div><div>Data-driven approaches hold promise for modeling the highly nonlinear stress-strain behavior of frozen soils in comparison to standalone machine learning models which often overfit and lack generalizability. To address these issues, in this study an enhanced fundamental Long Short-Term Memory (LSTM) model with an iterative EXtreme Gradient Boosting (XGBoost) algorithm is proposed to predict the axial strain-deviatoric stress relationship of frozen sands. Using degree of saturation, mean particle size, test rate, initial void ratio, temperature, confining pressure, and axial strain as input parameters, the Local Interpretable Model-agnostic Explanations (LIME) feature-importance analysis identified initial void ratio as the most influential parameter. The adaptability of the fundamental LSTM shows better accuracy than the Random Forest (RF) model and the Multilayer Perception (MLP) model. To reduce discrepancies between predicted and measured results in the fundamental LSTM model, an uncertainty factor was introduced to improve residual accuracy, thereby facilitating the development of an XGBoost-optimized LSTM (LSTMXGBoost) model. The resulting LSTMXGBoost model demonstrated strong predictive performance for axial strain-deviatoric stress relationship across a range of triaxial shear test conditions. Analysis of results suggest that there is a good comparison between two key parameters; namely, modulus of elasticity and peak stress that were extracted from a separate triaxial test dataset not used in model training or testing. The results of this study are promising for constructing high-dimensional constitutive models that can be used in numerical simulations of frozen sands behavior for use in geotechnical engineering practice applications alleviating time consuming, cumbersome and expensive experimental test techniques.</div></div>","PeriodicalId":10522,"journal":{"name":"Cold Regions Science and Technology","volume":"239 ","pages":"Article 104594"},"PeriodicalIF":3.8,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535076","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}