A Method for Predicting Icing Time at Tunnel Entrances in Cold Regions Based on Molecular Dynamics Simulation

Abstract

This application provides a method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation. Starting from the molecular dynamics simulation level, it uses a preliminarily constructed mineral-water-air three-phase interface model as a foundation to further simulate cooling and determine critical time nodes. Then, it considers the macroscopic icing time mapping of rocks to configure the final prediction model. Through this multi-scale integrated technical path consisting of microscopic simulation, parameter extraction, time mapping, and engineering prediction, the resulting prediction model can achieve in-depth and high-precision prediction of icing time at tunnel entrances in cold regions, providing high-quality data support for the prevention of icing at tunnel entrances in cold regions and showing promising application prospects.

Description

Technical Field

[0001] This application relates to the field of geology, specifically to a method for predicting the icing time at the entrance of a tunnel in a cold region based on molecular dynamics simulation. Background Technology

[0002] It is easy to understand that icing at the entrance of tunnels in cold regions may cause traffic accidents such as vehicles being unable to brake, skidding, or fishtailing. To address this safety hazard, appropriate preventative measures and de-icing procedures are necessary.

[0003] However, the inventors of this application have found that existing predictions of icing at tunnel entrances in cold regions mainly rely on manual methods, which combine past icing data with weather conditions for forecasting. Secondly, there are also some technologies that use basic meteorological factors such as temperature and humidity to predict icing at tunnel entrances in cold regions. However, based on actual research results, these methods, like manual methods, suffer from poor prediction accuracy. Summary of the Invention

[0004] This application provides a method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation. Starting from the molecular dynamics simulation level, it uses a preliminarily constructed mineral-water-air three-phase interface model as a foundation to further simulate cooling and determine critical time nodes. Then, it considers the macroscopic icing time mapping of rocks to configure the final prediction model. Through this multi-scale integrated technical path consisting of microscopic simulation, parameter extraction, time mapping, and engineering prediction, the resulting prediction model can achieve in-depth and high-precision prediction of the icing time of tunnel entrances in cold regions, providing high-quality data support for the prevention of icing at tunnel entrances in cold regions and showing promising application prospects.

[0005] Firstly, this application provides a method for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulations, the method comprising:

[0006] During the modeling stage, the main rock-forming minerals of the cold region rock mass are selected, and the corresponding stable crystal face structures of the main rock-forming minerals are constructed and vertically superimposed with the water molecule layer to form a mineral-water-air three-phase interface model with a thickness of 5-6 mm and a vacuum layer set on the top.

[0007] During the simulation execution phase, the molecular force field of the simulated clay mineral, water molecule and ion system corresponding to the mineral part is used, and the rigid planar four-point water model corresponding to the water molecule part is used. Under the ensemble conditions of keeping the number of particles, volume and temperature constant, the cooling simulation is carried out, and the critical time node for water molecules on the mineral surface to freeze is determined by the combined index of the sudden change in comprehensive potential energy, the surge in the number of hydrogen bonds and the transition of the mean square displacement curve from linear to plateau.

[0008] In the macroscopic rock freezing time mapping stage, the conversion relationship between the constructed freezing simulation time and the actual rock freezing time is used to map the microscopic time scale to the macroscopic freezing duration. Combined with the freezing time multi-factor response database, a prediction model is configured. Specifically, the prediction model adopts the support vector regression model, the random forest model, or the extreme gradient boosting model.

[0009] During the engineering prediction phase, the prediction model is used to predict the freezing time at the tunnel entrance in cold regions.

[0010] Secondly, this application provides a device for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation. The device includes:

[0011] The building unit is used in the modeling stage to select the main rock-forming minerals of the cold region rock mass and construct the corresponding stable crystal face structure of the main rock-forming minerals. It is then vertically superimposed with the water molecule layer to form a mineral-water-air three-phase interface model. The model thickness is 5-6 mm, and a vacuum layer is set on the top.

[0012] The simulation unit is used to simulate the molecular force field of clay minerals, water molecules and ion systems corresponding to the mineral part, and the rigid four-point water model corresponding to the water molecule part during the simulation execution phase. Under the ensemble conditions of keeping the number of particles, volume and temperature constant, the unit performs cooling simulation and uses the combined indices of potential energy mutation, hydrogen bond number surge and mean square displacement curve transition from linear to plateau to determine the critical time node for water molecules to freeze on the mineral surface.

[0013] The configuration unit is used to map the micro-timescale to the macro-freezing duration in the macro-freezing time mapping stage of rock by constructing the conversion relationship between the simulated freezing time and the actual rock freezing time, and to configure the prediction model in combination with the multi-factor response database of freezing time. The prediction model specifically adopts the support vector regression model, the random forest model, or the extreme gradient boosting model.

[0014] The prediction unit is used in the engineering prediction phase to call the prediction model to predict the icing time at the tunnel entrance in cold regions.

[0015] Thirdly, this application provides a processing device, including a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the method provided in the first aspect of this application when it invokes the computer program in the memory.

[0016] Fourthly, this application provides a computer-readable storage medium storing a plurality of instructions adapted for loading by a processor to execute the method provided in the first aspect of this application.

[0017] From the above, it can be concluded that this application has the following beneficial effects:

[0018] To predict the icing time at tunnel entrances in cold regions, this study starts with molecular dynamics simulations. Based on a preliminary mineral-water-air three-phase interface model, it continues to simulate cooling and determine critical time points. Then, it considers the macroscopic icing time mapping of rocks to configure the final prediction model. In this way, through a multi-scale integrated technical path consisting of microscopic simulation, parameter extraction, time mapping, and engineering prediction, the resulting prediction model can achieve in-depth and high-precision prediction of icing time at tunnel entrances in cold regions. This provides high-quality data support for the prevention of icing at tunnel entrances in cold regions and has good application prospects. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a method for predicting icing time at tunnel entrances in cold regions based on molecular dynamics simulations, as described in this application.

[0021] Figure 2 This is a schematic diagram of a device for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation, as described in this application.

[0022] Figure 3 This is a schematic diagram of one type of processing equipment used in this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to those explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices. The naming or numbering of steps appearing in this application does not imply that the steps in the method flow must be performed in the chronological / logical order indicated by the naming or numbering. The execution order of named or numbered process steps can be changed according to the desired technical purpose, as long as the same or similar technical effect is achieved.

[0025] The module division described in this application is a logical division. In practical applications, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the coupling or direct coupling or communication connection between modules shown or discussed may be through some interfaces, and the indirect coupling or communication connection between modules may be electrical or other similar forms, none of which are limited in this application. Furthermore, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed in multiple circuit modules. Some or all of the modules may be selected to achieve the purpose of the solution in this application according to actual needs.

[0026] Before introducing the method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation provided in this application, we will first introduce the background content involved in this application.

[0027] The method, apparatus, and computer-readable storage medium for predicting icing time at tunnel entrances in cold regions based on molecular dynamics simulation provided in this application can be applied to processing equipment. Starting from the molecular dynamics simulation level, it uses a preliminarily constructed mineral-water-air three-phase interface model as a foundation to further simulate cooling and determine critical time nodes. Then, it considers the macroscopic icing time mapping of rocks to configure the final prediction model. Thus, through a multi-scale integrated technical path consisting of microscopic simulation, parameter extraction, time mapping, and engineering prediction, the resulting prediction model can achieve a deep and high-precision prediction effect on the icing time at tunnel entrances in cold regions, providing high-quality data support for the prevention of icing at tunnel entrances in cold regions and showing promising application prospects.

[0028] The method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation mentioned in this application can be implemented by a device for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation, or by different types of processing devices such as servers, physical hosts, or user equipment (UEs) that integrate such a device. The device for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation can be implemented in hardware or software. The UE can be a terminal device such as a smartphone, tablet, laptop, desktop computer, or personal digital assistant (PDA). The processing devices can be configured in a device cluster.

[0029] It is understandable that the solution proposed in this application is mainly based on existing data. Therefore, in practical applications, the processing equipment that implements the method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation, or the processing equipment that carries the corresponding application service of the method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation, usually only needs to meet the required data processing capabilities. The specific equipment type and equipment deployment form are quite flexible and can be flexibly configured according to the actual situation.

[0030] If the direct collection of existing data (mainly referring to the relevant parameters that need to be collected from the site) mentioned above is involved, then the existing data acquisition equipment needs to be integrated into the processing equipment, or the existing data acquisition equipment needs to be included in the equipment cluster of the processing equipment, or the processing equipment can call it in the form of a third-party device.

[0031] In addition, if there are corresponding content display requirements for the solution processing results or process, the processing device can also be equipped with a corresponding display screen (including touch screen) to display the content, or it can be connected to an external display device or call other devices with display screens to display the content.

[0032] The following section introduces the method for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation, which is provided in this application.

[0033] First, refer to Figure 1 , Figure 1 This paper presents a flowchart illustrating a method for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulations, as described in this application. The method specifically includes the following steps S101 to S104:

[0034] Step S101: In the modeling stage, select the main rock-forming minerals of the cold region rock mass and construct the corresponding stable crystal surface structure of the main rock-forming minerals. Vertically superimpose them with the water molecule layer to form a mineral-water-air three-phase interface model with a model thickness of 5-6 mm and a vacuum layer on top.

[0035] It is understandable that the microscopic simulation here refers more specifically to the processing stage of constructing the mineral-water interface model. The goal is to initially construct the corresponding molecular-level / scale three-phase interface model of mineral-water-air, laying a good foundation for more detailed dynamic processing in the subsequent simulation execution stage.

[0036] The main rock-forming minerals are representative minerals found in common rock masses at the entrance of tunnels in cold regions, such as granite, gneiss, and schist. Specifically, quartz, feldspar, and mica can be selected.

[0037] For the main rock-forming minerals selected for the current tunnel entrances in cold regions, based on the most common stable crystal face orientation of each type of mineral, such as the (001) face of quartz, the (001) face of potassium feldspar, and the (001) face of muscovite, stable crystal structures / crystal structure models are extracted respectively.

[0038] The selected crystal facets have good surface stability and can truly reflect the interaction mechanism between mineral surfaces and water molecules. They are used to simulate the typical water-mineral interface structure of the rock mass surface at the entrance of a tunnel in a cold region after being wetted by surface water. This provides a molecular simulation basis for predicting the freezing behavior and onset time of the water layer under low temperature conditions.

[0039] Furthermore, the thickness of the water film is controlled at approximately 5-6 nanometers (nm) to ensure that water molecules can both undergo adsorption and reconstruction at the interface and have sufficient volume to participate in the nucleation and crystallization process.

[0040] Meanwhile, to avoid mutual interference caused by periodic boundary conditions during the simulation process, a vacuum layer (typically 10-15 nm thick) is set on the top of the model / simulation box to construct a mineral-water-air three-phase interface model, simulate the actual state of water occurrence in the pores, and form a typical molecular-scale simulation system.

[0041] Furthermore, after completing the geometric structure construction work above, the system initialization work can continue.

[0042] Correspondingly, as an exemplary embodiment, the modeling phase may further include:

[0043] 1.1) Energy Minimization

[0044] The initial model structure is optimized for energy using algorithms such as the conjugate gradient algorithm or the steepest descent algorithm.

[0045] Understandably, the purpose of this operation is to eliminate, in detail, any non-physical overlaps and stress accumulations that may exist in the current mineral-water-air three-phase interface model.

[0046] 1.2) Initial thermal equilibrium treatment (Equilibration) step

[0047] At room temperature, short-term equilibrium is achieved through ensemble conditions.

[0048] Understandably, the purpose of this operation is to, in terms of details, maintain the ensemble conditions (i.e., NVT ensemble) under normal temperature conditions such as 280-300K, corresponding to the subsequent particle number (N), volume (V), and temperature (T) being kept constant, to achieve a better stable state of the interface structure of the mineral-water-air three-phase interface model, thus providing a reliable configuration for subsequent temperature control.

[0049] Step S102: During the simulation execution phase, the molecular force field of the simulated clay mineral, water molecule and ion system corresponding to the mineral part is used, and the rigid planar four-point water model corresponding to the water molecule part is used. Under the ensemble conditions where the number of particles, volume and temperature are kept constant, a cooling simulation is carried out, and the critical time node for water molecules on the mineral surface to freeze is determined by the combined index of the sudden change in comprehensive potential energy, the surge in the number of hydrogen bonds and the transition of the mean square displacement curve from linear to plateau.

[0050] Understandably, the parameter extraction here refers more specifically to the processing steps of icing process simulation and critical time extraction. Based on the mineral-water-air three-phase interface model obtained / constructed earlier, dynamic cooling simulation is carried out. Thus, the critical time node / critical time marker for water molecule freezing on the mineral surface can be further determined by monitoring the cooling simulation.

[0051] Specifically, the corresponding mineral part uses the molecular force field of the simulated clay mineral, water molecule and ion system, namely the CLAYFF force field. This force field has good cross-mineral versatility and is suitable for modeling mineral materials mainly composed of silicon, aluminum and oxygen, including quartz, feldspar and mica.

[0052] Meanwhile, for the water molecule part, a rigid planar four-point water model, namely the TIP4P / Ice model, was specifically adopted. This model is a four-point water model developed for ice phase systems, which can accurately describe the intermolecular hydrogen bond structure and thermodynamic properties during the water-ice phase transition.

[0053] In addition, to achieve a deeper and more refined cooling simulation effect, this application may further introduce other constraints and optimization settings.

[0054] Correspondingly, as an exemplary embodiment, during the simulation execution phase, van der Waals interactions, electrostatic interactions, integral algorithms, and thermal control mechanisms can also be used, combined with enhanced sampling methods to increase the probability of nucleation events, in order to perform cooling simulation.

[0055] Specifically, the settings here include:

[0056] Van der Waals interaction: described using the Lennard-Jones 12-6 potential function, with the cutoff radius set to 10 Å (the unit is angstrom, a commonly used unit of length in crystallography, atomic physics, and ultrastructures).

[0057] Electrostatic interaction treatment: For long-range electrostatic interactions between water molecules and charged atoms in minerals, the Particle-Particle-Particle-Mesh (PPPM) method was used for calculation, with the cutoff radius of the electrostatic potential set to 8.5 Å.

[0058] Integration algorithm: The system dynamic equations are integrated using the Velocity Verlet integrator with a time step of 1 fs (femtosecond).

[0059] Thermal control mechanism: The Nosé–Hoover temperature controller is used during the simulation to keep the system temperature stable at the preset target temperature and ensure the energy and dynamic balance under the NVT ensemble.

[0060] It should be noted that simulating the freezing process of water on mineral surfaces, especially heterogeneous nucleation events, within the framework of all-atomic molecular dynamics usually faces significant timescale obstacles.

[0061] Specifically, the spontaneous formation of ice embryos is a low-probability and rare event, and it often cannot occur naturally under normal simulation conditions (picosecond to nanosecond level), while the actual freezing process in engineering environments usually occurs on the order of seconds to hours.

[0062] Enhanced sampling strategy assistance: For systems where nucleation delay is significant under specific conditions and freezing events are difficult to observe in standard simulation time, this application can combine enhanced sampling methods such as Metadynamics and Forward Flux Sampling to increase the probability of rare nucleation events, thereby capturing ice nucleus formation behavior within a limited simulation time range and providing support for the extraction of freezing time points. This strategy does not change the physical nature of the system, but only improves efficiency in the sampling space, ensuring that the extracted freezing times are comparable and representative.

[0063] At this point, the specific cooling simulation process can be advanced by combining the ensemble conditions where the number of particles, volume, and temperature remain constant, i.e., the NVT ensemble.

[0064] As an exemplary embodiment, the specific cooling simulation process adopted here may include:

[0065] During the simulated cooling process, the system temperature is adjusted, specifically from the initial temperature to any set temperature range below freezing point, either slowly or in stages.

[0066] Understandably, during the simulated cooling process, the system temperature is adjusted to slowly or gradually decrease from the initial temperature to any set temperature range below the freezing point, such as gradually decreasing from 273K to 260K, 250K, etc.

[0067] The temperature control method adopts constant temperature (NVT) conditions and is combined with Nose-Hoover temperature controller to achieve stable cooling of the system, thereby capturing the microstructure evolution process of water molecules at a specific temperature.

[0068] The model construction strategy can realistically reproduce the freezing process of water film on the surface of tunnel rock in cold regions under natural cooling, and lay a microscopic foundation for subsequent extraction of key physical criteria, determination of freezing initiation point and prediction of freezing time.

[0069] Based on the cooling simulation, the process of determining the critical time point for water molecule freezing on the mineral surface by combining the combined indices of sudden change in comprehensive potential energy, surge in the number of hydrogen bonds, and the transition of the mean square displacement curve from linear to plateau can specifically involve the freezing criterion definition shown below.

[0070] As the temperature gradually decreases below the freezing point, the structure and dynamics of water molecules within the system will undergo significant changes with temperature. This application determines whether freezing occurs in the system by real-time monitoring of the dynamic evolution of the following three indicators:

[0071] (1) Change in total potential energy of the system

[0072] If an ice-forming phase transition occurs during the cooling process, the system's potential energy will drop rapidly, reflecting the latent heat released during the transition from a disordered liquid state to an ordered solid structure.

[0073] After the ice phase is formed, the potential energy curve tends to stabilize, indicating that the system has reached phase transition equilibrium.

[0074] (2) Changes in the number of hydrogen bonds

[0075] The number of hydrogen bonds in liquid water fluctuates greatly and lacks regularity;

[0076] As freezing progresses, water molecules arrange themselves in an orderly manner, the number of hydrogen bonds increases dramatically, and tends to stabilize after crystallization, reflecting a stable hydrogen bond network in the ice phase structure.

[0077] (3) Mean Square Displacement (MSD) of water molecules

[0078] In the liquid state, the MSD of water molecules increases linearly over time, reflecting their diffusivity and fluidity.

[0079] When the system freezes, water molecules are positioned in the crystal lattice, the MSD curve tends to stabilize and no longer increases over time, indicating that it has lost its fluidity.

[0080] Based on this, this application proposes an "ice formation criterion": including the rapid decrease in the potential energy of water molecules, the sudden increase in the number of hydrogen bonds, and the transition of the mean square displacement (MSD) curve from linear to plateau, etc., to determine the critical time point for the initial formation of the ice phase. This determination method does not rely on the identification of crystal features of a single structure, but rather accurately and indirectly identifies the occurrence of phase transition through the coupled changes of multiple thermodynamic and kinetic signals.

[0081] By simulating different mineral-water interface systems under the same cooling path, recording the changes in the aforementioned characteristic parameters, and extracting the time point when the freezing criteria are first simultaneously met, the time point when water molecules on the mineral surface freeze can be obtained. This method provides a high-precision, quantifiable path for extracting microscopic freezing behavior, and has good versatility and engineering guidance value.

[0082] Step S103: In the macroscopic rock freezing time mapping stage, the conversion relationship between the constructed freezing simulation time and the actual rock freezing time is used to map the microscopic time scale to the macroscopic freezing duration. Combined with the freezing time multi-factor response database, a prediction model is configured. Specifically, the prediction model adopts the Support Vector Regression (SVR) model, the Random Forest (RF) model, or the Extreme Gradient Boosting (XGBoost) model.

[0083] Understandably, the time mapping here refers more specifically to the micro-to-macro mapping and influencing factor fitting process. After determining the critical time node for water molecules on the mineral surface to freeze / condense, i.e., the critical freezing time node, based on the cooling simulation, we can continue to perform the conversion between the molecular scale and the actual application scale, from micro to macro, to help configure a predictive model that can be put into practical use.

[0084] In order to realize the quantifiable output of molecular simulation icing results in the actual application of tunnels in cold regions, this application proposes a mapping method for the macroscopic icing time of rock mass. By constructing the conversion relationship between the molecular simulation time scale and the actual geological-environmental process time scale, or the "simulation-actual" time scaling model, a time prediction model with practical application value is formed, which is the aforementioned conversion relationship between the constructed icing simulation time and the actual rock mass icing time.

[0085] Specifically, in molecular-scale simulations, the evolution of the freezing process is measured in picoseconds (ps) to nanoseconds (ns), which can only reflect the rapid phase transition behavior of water molecules under ideal conditions induced by the crystal lattice. In contrast, the actual freezing process of rock mass at the tunnel entrance is affected by the following coupled factors: sluggish heat conduction in the rock mass; gradual temperature gradient; migration and redistribution of capillary water or accumulated water; development of surface ice nuclei and evolution of pore freezing; and local freezing lag caused by surface micromorphology.

[0086] To bridge the aforementioned timescale differences, this application addresses this by constructing a scaling mapping model of "simulated time - actual freezing time." As an exemplary embodiment, the conversion relationship between simulated freezing time and actual rock mass freezing time can specifically include the following construction content:

[0087] 3.1) Select representative mineral-water interface systems and obtain their freezing initiation time points under the same cooling path;

[0088] In this processing step, the representative mineral-water interface system can be selected from specific systems such as quartz-water, feldspar-water, and mica-water.

[0089] 3.2) Based on existing experimental or monitoring data, establish a typical temperature drop-freezing response curve;

[0090] In this process, experiments such as indoor cooling and freezing tests of water-bearing rock samples and monitoring data such as on-site temperature measurements and rock mass freezing records have been conducted.

[0091] 3.3) By matching the temperature drop path with the freezing dynamic features through normalization, the scaling factor is extracted to construct a time mapping function that adapts to different lithologies and temperature drop conditions.

[0092] In this processing step, the time mapping function is used to map the "time point when the freezing criterion is first established" in the simulation to the time window when free water or capillary water in the actual rock mass begins to freeze, thereby realizing the conversion from microscopic dynamic time to macroscopic freezing time.

[0093] After mapping the micro-timescale to the macro-freezing duration to obtain the critical freezing time point for practical application, the specific configuration and processing of the prediction model can be further advanced by combining the multi-factor response database of freezing time.

[0094] It is easy to understand that this application may also involve the construction and processing of the multi-factor response database for freezing time.

[0095] Understandably, this application further constructs a multi-factor response database of freezing time, or a freezing time-environment response database, to help form a general prediction function model, targeting the differences in freezing behavior under different combinations of lithology, porosity characteristics and temperature environment conditions.

[0096] Specifically, as an exemplary embodiment, the icing time multi-factor response database may include the following components:

[0097] The system collects and processes parameters including rock mass type, water-bearing state, ambient temperature field, and water accumulation state at the tunnel entrance;

[0098] The freezing time nodes obtained from simulations under various conditions (or different combinations of conditions) are paired with parameters to form sample pairs, creating "freezing time-multi-factor" data pairs, thus forming a structured database, namely the freezing time multi-factor response database.

[0099] Specifically, the parameters for the above four aspects can be:

[0100] (1) Rock mass type:

[0101] Specific parameters include lithological classification (such as granite, gneiss, diabase, etc.) and main mineral types (such as quartz, potassium feldspar, biotite, etc.).

[0102] Parameters including porosity, saturation, water content, volume / depth of water accumulation, external temperature gradient, rate of temperature drop, and minimum ambient temperature;

[0103] (2) Moisture content:

[0104] Specific parameters include pore water content, saturation, and pore structure characteristics (which can be simplified to the saturated-unsaturated range).

[0105] (3) Ambient temperature field:

[0106] This includes specific parameters such as temperature-time change curves measured on-site / foreign weather forecasts (especially the rate of temperature drop and the minimum temperature threshold).

[0107] (4) Water accumulation at the entrance of the tunnel:

[0108] This includes specific parameters such as the initial water depth, water film thickness, and water coverage of the surface or shallow free water body in the area near the tunnel entrance. It should be noted that these parameters are highly correlated with the freezing start time, meaning they have a significant impact.

[0109] By using the specific parameters mentioned above, the actual working conditions / prediction scenarios corresponding to the current prediction task can be clearly described / defined. This helps the prediction model to cope with combinations of different rock types, different moisture conditions, and non-constant temperature drop fields, and has good working condition adaptability and generalization ability. This allows the prediction model to perform detailed prediction processing of the icing time at the tunnel entrance in cold regions.

[0110] Having obtained the two main datasets for configuring the prediction model—the critical freezing time node and the multi-factor response database for freezing time—we can then perform normalization and standardization processing on the obtained datasets (such as unit unification and range determination) to further improve the data quality of the training samples used to train the model through data preprocessing.

[0111] Next, the prediction model can be trained using multiple regression, interpolation fitting, or machine learning methods (which may involve initialization machine learning models such as random forest regression model and support vector machine regression).

[0112] It is understandable that the specific model structure, model training scheme, and loss function involved in the prediction model involved in this application are not the focus of this application. Therefore, relevant existing technologies can be directly adopted. Of course, in actual situations, it is not ruled out that existing technologies can be further optimized or even novel self-developed schemes can be used.

[0113] Furthermore, it can be noted that the prediction model in this application can be a nonlinear prediction model or an interpolation fitting function, or it can be an AI model.

[0114] A well-trained predictive model that can be put into practical use can, in essence, provide / output estimates of the freezing start time and freezing completion time of the corresponding rock mass under the current working conditions, based on the relevant input parameters from the site, thus meeting the data usage needs of related projects.

[0115] It is easy to understand that this prediction method is particularly suitable for determining whether there is a risk of icing in the tunnel entrance area in cold regions within a specific time period, especially in rock mass areas with significant water accumulation or high water content, and has important safety and scheduling guidance significance.

[0116] Step S104: In the engineering prediction stage, the prediction model is called to predict the icing time at the tunnel entrance in cold regions.

[0117] More specifically, the engineering prediction here refers to the intelligent prediction and output of freezing time. As mentioned earlier, after the prediction model has been trained in advance, it can be invoked locally or remotely during the actual use stage, i.e., the engineering prediction stage. By inputting a parameter set corresponding to the multi-factor response database of freezing time, the prediction model can perform the corresponding freezing time prediction processing, analyze the key physical processes of water freezing on the surface of rocks and minerals at the molecular scale, determine the critical time for the initial freezing, and help identify high-risk freezing scenarios and freezing evolution trends.

[0118] In practical applications, the model input can be configured either directly according to the parameter set in the previous model configuration process or by matching the corresponding target parameter set obtained from the parameter set in the previous model configuration process. This means matching the closest simulated working condition. This corresponds to the different model types that the prediction model of this application can involve, in order to meet the diverse application needs of the scheme.

[0119] For the model input obtained, automatic recognition / extraction processing can be introduced during the data entry process so that the corresponding application can enter the data. At the same time, preprocessing such as normalization and standardization (e.g., unit unification, range judgment) can also be involved to improve data quality.

[0120] After obtaining the prediction results, namely the prediction results of the icing time at the entrance of tunnels in cold regions, through the prediction model, further data processing may be involved to meet the specific data application needs in actual situations.

[0121] For example, the prediction results of icing time at tunnel entrances in cold regions can be stored locally or remotely, displayed, pushed to users, and accompanied by notifications of completed predictions or further data analysis.

[0122] Understandably, the specific data processing can be flexibly adjusted according to the pre-configured and real-time data processing strategies.

[0123] As an example, the output of the prediction model can also be presented in an engineering-friendly format, specifically supporting the following output formats:

[0124] Basic output format: The output uses the quadruple "rock mass type-water content-water accumulation state-temperature drop rate" as the main index to output the corresponding predicted critical freezing time point (unit: hours / day).

[0125] Additional information: It can also output auxiliary parameters such as the freezing duration range, estimated freezing rate, and freezing risk level suggestions (such as low / medium / high levels);

[0126] Interface support: Supports embedding prediction results into construction organization design systems and monitoring and early warning platforms in the form of tables, visualization charts or application programming interfaces (APIs), providing decision support for decision-making links / processing systems such as support sequence arrangement, lining structure antifreeze design or construction stoppage plan during the construction phase.

[0127] It is worth noting that, in this embodiment, the output of the prediction model of this application can be specifically configured to include several aspects such as the critical freezing time point, freezing duration range, freezing rate estimate and freezing risk level suggestion, thereby forming in-depth and information-rich prediction content, which is more visual and practical.

[0128] In conclusion, regarding the above solutions, for the goal of predicting the icing time at tunnel entrances in cold regions, starting from the molecular dynamics simulation level, and based on the initially constructed mineral-water-air three-phase interface model, further cooling simulations and critical time node determination are carried out. Then, the macroscopic icing time mapping of rocks is considered to fit the final prediction model. In this way, under the multi-scale integrated technical path of microscopic simulation, parameter extraction, time mapping, and engineering prediction, the obtained prediction model can achieve a deep and high-precision prediction effect on the icing time at tunnel entrances in cold regions, providing high-quality data support for the prevention of icing at tunnel entrances in cold regions, and has good application prospects.

[0129] The above is an introduction to the method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation provided in this application. To facilitate better implementation of the method for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation provided in this application, this application also provides a device for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation from the perspective of functional modules.

[0130] See Figure 2 , Figure 2 This is a schematic diagram of a device for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation, as described in this application. Specifically, the device 200 for predicting the icing time of tunnel entrances in cold regions based on molecular dynamics simulation may include the following structure:

[0131] Construction unit 201 is used in the modeling stage to select the main rock-forming minerals of the cold region rock mass and construct the corresponding stable crystal surface structure of the main rock-forming minerals, which are vertically superimposed with the water molecule layer to form a mineral-water-air three-phase interface model. The model thickness is 5-6 mm, and a vacuum layer is set on the top.

[0132] Simulation unit 202 is used to simulate the molecular force field of clay minerals, water molecules and ion systems corresponding to the mineral part, and the rigid four-point water model corresponding to the water molecule part during the simulation execution phase. Under the ensemble conditions of keeping the number of particles, volume and temperature constant, the cooling simulation is performed, and the critical time node for water molecules on the mineral surface to freeze is determined by the combined index of the sudden change in comprehensive potential energy, the surge in the number of hydrogen bonds and the transition of the mean square displacement curve from linear to plateau.

[0133] Configuration unit 203 is used to map the micro-timescale to the macro-ice duration in the macro-ice time mapping stage of rock macro-ice time by constructing the conversion relationship between the simulated ice time and the actual rock ice time, and to configure the prediction model in combination with the multi-factor response database of ice time. Specifically, the prediction model adopts the support vector regression model, the random forest model, or the extreme gradient boosting model.

[0134] Prediction unit 204 is used to call the prediction model to predict the icing time at the tunnel entrance in cold regions during the engineering prediction phase.

[0135] As an exemplary embodiment, the modeling phase also includes:

[0136] Energy optimization of the initial model structure is performed using the conjugate gradient algorithm or the steepest descent algorithm.

[0137] At room temperature, short-term equilibrium is achieved through ensemble conditions.

[0138] As another exemplary embodiment, the simulated cooling process includes:

[0139] During the simulated cooling process, the system temperature is adjusted, specifically from the initial temperature to any set temperature range below freezing point, either slowly or in stages.

[0140] As another exemplary embodiment, during the simulation execution phase, van der Waals interactions, electrostatic interactions, integral algorithms, and thermal control mechanisms are also employed, along with an enhanced sampling method to increase the probability of nucleation events, in order to simulate cooling.

[0141] As another exemplary embodiment, the conversion relationship between the simulated freezing time and the actual rock mass freezing time includes the following components:

[0142] Representative mineral-water interface systems were selected, and their freezing initiation time points were obtained under the same cooling path.

[0143] Based on existing experimental or monitoring data, establish a typical temperature drop-freezing response curve;

[0144] By matching the temperature drop path with the freezing dynamic features and extracting the scaling factor, a time mapping function adapted to different lithologies and temperature drop conditions is constructed.

[0145] As another exemplary embodiment, the freezing time multifactor response database includes the following components:

[0146] The system collects and processes parameters including rock mass type, water-bearing state, ambient temperature field, and water accumulation state at the tunnel entrance;

[0147] The icing time points obtained from simulations under various conditions are paired with parameters to form a structured database.

[0148] As another exemplary embodiment, the output of the prediction model includes the critical freezing time point, freezing duration range, freezing rate estimate, and freezing risk level recommendation.

[0149] This application also provides a processing device from a hardware architecture perspective, see [link / reference]. Figure 3 , Figure 3 This diagram illustrates a structural schematic of the processing device of this application. Specifically, the processing device may include a processor 301, a memory 302, and an input / output device 303. The processor 301 executes the computer program stored in the memory 302 to implement, for example... Figure 1 The corresponding steps of the method for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation in the embodiment; or, when the processor 301 executes the computer program stored in the memory 302, it implements as follows: Figure 2 Corresponding to the functions of each unit in the embodiment, the memory 302 is used to store the functions executed by the processor 301 as described above. Figure 1The corresponding embodiment includes the computer program required for the method of predicting icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation.

[0150] For example, a computer program may be divided into one or more modules / units, one or more of which are stored in memory 302 and executed by processor 301 to complete this application. One or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a computer device.

[0151] The processing device may include, but is not limited to, processor 301, memory 302, and input / output device 303. Those skilled in the art will understand that the illustrations are merely examples of the processing device and do not constitute a limitation on the processing device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the processing device may also include network access devices, buses, etc., and processor 301, memory 302, input / output device 303, etc., are connected via a bus.

[0152] Processor 301 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the processing device, connecting various parts of the device through various interfaces and lines.

[0153] The memory 302 can be used to store computer programs and / or modules. The processor 301 implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory 302 and by calling data stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function, etc.; the data storage area may store data created according to the use of the processing device, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0154] When processor 301 executes a computer program stored in memory 302, it can specifically perform the following functions:

[0155] During the modeling stage, the main rock-forming minerals of the cold region rock mass are selected, and the corresponding stable crystal face structures of the main rock-forming minerals are constructed and vertically superimposed with the water molecule layer to form a mineral-water-air three-phase interface model with a thickness of 5-6 mm and a vacuum layer set on the top.

[0156] During the simulation execution phase, the molecular force field of the simulated clay mineral, water molecule and ion system corresponding to the mineral part is used, and the rigid planar four-point water model corresponding to the water molecule part is used. Under the ensemble conditions of keeping the number of particles, volume and temperature constant, the cooling simulation is carried out, and the critical time node for water molecules on the mineral surface to freeze is determined by the combined index of the sudden change in comprehensive potential energy, the surge in the number of hydrogen bonds and the transition of the mean square displacement curve from linear to plateau.

[0157] In the macroscopic rock freezing time mapping stage, the conversion relationship between the constructed freezing simulation time and the actual rock freezing time is used to map the microscopic time scale to the macroscopic freezing duration. Combined with the freezing time multi-factor response database, a prediction model is configured. Specifically, the prediction model adopts the support vector regression model, the random forest model, or the extreme gradient boosting model.

[0158] During the engineering prediction phase, the prediction model is used to predict the freezing time at the tunnel entrance in cold regions.

[0159] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described molecular dynamics simulation-based device for predicting icing time at tunnel entrances in cold regions, its processing equipment, and its corresponding units can be found in the following reference: Figure 1The description of the method for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation in the corresponding embodiment will not be repeated here.

[0160] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0161] Therefore, this application provides a computer-readable storage medium storing a plurality of instructions that can be loaded by a processor to execute the present application. Figure 1 The steps of the method for predicting icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation in the corresponding embodiment can be found in the following example. Figure 1 The description of the method for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation in the corresponding embodiment will not be repeated here.

[0162] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0163] Because of the instructions stored in the computer-readable storage medium, the present application can be executed as described above. Figure 1 The steps of the method for predicting icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation in the corresponding embodiment can be implemented as described in this application. Figure 1 The beneficial effects of the molecular dynamics simulation-based method for predicting icing time at tunnel entrances in cold regions, as described in the corresponding embodiments, are detailed in the preceding description and will not be repeated here.

[0164] The above provides a detailed description of the method, apparatus, processing equipment, and computer-readable storage medium for predicting icing time at tunnel entrances in cold regions based on molecular dynamics simulations, as provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of this application. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation, characterized in that, The method includes: During the modeling stage, the main rock-forming minerals of the cold region rock mass are selected, and the corresponding stable crystal face structure of the main rock-forming minerals is constructed and vertically superimposed with the water molecule layer to form a mineral-water-air three-phase interface model with a model thickness of 5-6 mm and a vacuum layer set on the top. During the simulation execution phase, the molecular force field of the simulated clay mineral, water molecule and ion system corresponding to the mineral part is used, and the rigid planar four-point water model corresponding to the water molecule part is used. Under the ensemble conditions of keeping the number of particles, volume and temperature constant, the cooling simulation is carried out, and the critical time node for water molecules on the mineral surface to freeze is determined by the combined index of the sudden change in comprehensive potential energy, the surge in the number of hydrogen bonds and the transition of the mean square displacement curve from linear to plateau. In the macroscopic rock freezing time mapping stage, the conversion relationship between the constructed freezing simulation time and the actual rock freezing time is used to map the microscopic time scale to the macroscopic freezing duration. Combined with the freezing time multi-factor response database, a prediction model is configured. Specifically, the prediction model adopts a support vector regression model, a random forest model, or an extreme gradient boosting model. During the engineering prediction phase, the prediction model is used to predict the freezing time at the tunnel entrance in cold regions.

2. The method according to claim 1, characterized in that, The modeling phase also includes: Energy optimization of the initial model structure is performed using the conjugate gradient algorithm or the steepest descent algorithm. At room temperature, short-term equilibrium is achieved using the ensemble conditions.

3. The method according to claim 1, characterized in that, The simulated cooling process includes: During the simulated cooling process, the system temperature is adjusted and gradually or stepwise reduced from the initial temperature to any set temperature range below the freezing point.

4. The method according to claim 1, characterized in that, In the simulation execution phase, van der Waals interactions, electrostatic interactions, integral algorithms, and thermal control mechanisms are also employed, along with enhanced sampling methods to increase the probability of nucleation events, to perform the cooling simulation.

5. The method according to claim 1, characterized in that, The conversion relationship between the simulated freezing time and the actual rock mass freezing time includes the following components: Representative mineral-water interface systems were selected, and their freezing initiation time points were obtained under the same cooling path. Based on existing experimental or monitoring data, establish a typical temperature drop-freezing response curve; By matching the temperature drop path with the freezing dynamic features and extracting the scaling factor, a time mapping function adapted to different lithologies and temperature drop conditions is constructed.

6. The method according to claim 1, characterized in that, The freezing time multi-factor response database includes the following components: The system collects and processes parameters including rock mass type, water-bearing state, ambient temperature field, and water accumulation state at the tunnel entrance; The icing time points obtained from simulations under various conditions are paired with the parameters to form a structured database.

7. The method according to claim 1, characterized in that, The output of the prediction model includes the critical freezing time point, freezing duration range, estimated freezing rate, and freezing risk level recommendation.

8. A device for predicting the icing time at the entrance of a cold-region tunnel based on molecular dynamics simulation, characterized in that, The device includes: The building unit is used in the modeling stage to select the main rock-forming minerals of the cold region rock mass and construct the corresponding stable crystal face structure of the main rock-forming minerals. It is then vertically superimposed with the water molecule layer to form a mineral-water-air three-phase interface model with a model thickness of 5-6 mm and a vacuum layer on top. The simulation unit is used to simulate the molecular force field of clay minerals, water molecules and ion systems corresponding to the mineral part, and the rigid four-point water model corresponding to the water molecule part during the simulation execution phase. Under the ensemble conditions of keeping the number of particles, volume and temperature constant, the unit performs cooling simulation and uses the combined indices of potential energy mutation, hydrogen bond number surge and mean square displacement curve transition from linear to plateau to determine the critical time node for water molecules to freeze on the mineral surface. The configuration unit is used to map the micro-timescale to the macro-ice duration in the macro-ice time mapping stage of rock macro-ice time by constructing the conversion relationship between the simulated ice time and the actual rock ice time, and to configure the prediction model in combination with the multi-factor response database of ice time. Specifically, the prediction model adopts the support vector regression model, the random forest model, or the extreme gradient boosting model. The prediction unit is used to call the prediction model during the engineering prediction phase to predict the icing time at the tunnel entrance in cold regions.

9. A processing device, characterized in that, The method includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the method as described in any one of claims 1 to 7 when it invokes the computer program in the memory.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions adapted for loading by a processor to perform the method of any one of claims 1 to 7.

Images