Railway snow-blowing disaster management method, system and equipment

Abstract

The invention provides a railway wind-blown snow disaster management method, system and device, and relates to the technical field of railway wind-blown snow, and the method comprises the steps: carrying out the grid unit segmentation of a railway line, and calculating the topographic factor and shielding factor of each grid unit according to the topographic distribution data and forest belt canopy data of the railway line; calculating a snowfall accumulation prediction index of each grid unit in the expected prediction window according to the real-time meteorological input, the topographic factor and the shielding factor; constructing a local accumulation feature model according to the snow blowing equipment array, and performing local accumulation analysis on the snowfall accumulation prediction index to obtain a snowfall local accumulation prediction index; and performing control optimization on the snow blowing equipment array according to the snowfall local accumulation prediction index to obtain a control parameter set, and issuing the control parameter set to the snow blowing equipment array to execute snow blowing control. According to the invention, the technical problem of insufficient railway snow disaster management precision in the prior art can be solved, and the technical effect of improving the management precision is achieved.

Description

Technical Field

[0001] This application relates to the field of railway blowing snow technology, and in particular to railway blowing snow disaster management methods, systems and equipment. Background Technology

[0002] With the impact of factors such as climate change and environmental alteration, railway operations face increasingly severe threats from natural disasters. Winter snowfall brings numerous problems to railway operations. Railway blowing snow disaster refers to the situation where, under the threat of strong winds and heavy snow, snow accumulation covers critical components such as switch points and frogs, causing switch switching to become inflexible or even impossible, increasing the risk of train delays and safety accidents. Simultaneously, snowfall can also cause frost damage to switch components, affecting their service life.

[0003] Currently, snow removal on railway switches mainly relies on traditional methods such as manual sweeping and mechanical snow removal. Manual snow removal is inefficient, labor-intensive, and difficult to guarantee timely snow removal in severe weather. Mechanical snow removal may cause some damage to switch components and may also result in incomplete snow removal. In addition, the lack of fine-grained analysis of local wind fields, terrain features, and the shading effect of forest belts often leads to delayed or overly broad snow removal decisions, resulting in low resource utilization efficiency and the possibility of untimely snow removal, excessive energy consumption, or insufficient snow removal due to misjudgment of snow accumulation risks.

[0004] In summary, existing technologies suffer from technical problems due to the lack of a joint modeling mechanism and control optimization strategy based on terrain, forest belts, and real-time meteorological data. These problems result in insufficient accuracy in predicting snow accumulation along railway lines, untimely control response of snow blowing equipment, and unreasonable energy consumption management. Consequently, these issues further affect the operational safety, snow removal efficiency, and reliability of equipment collaborative control in high-altitude and cold regions of railways under severe weather conditions. Summary of the Invention

[0005] The purpose of this application is to provide a method, system, and equipment for railway snowstorm disaster management, in order to solve the technical problems in the existing technology that, due to the lack of a joint modeling mechanism and control optimization strategy based on terrain, forest belt, and real-time meteorological data, the prediction accuracy of snow accumulation along the railway line is insufficient, the control response of snow blowing equipment is untimely, and the energy consumption management is unreasonable, which further affects the operational safety of railways in high-altitude and cold areas under severe weather conditions, the efficiency of snow removal operations, and the reliability of equipment collaborative control.

[0006] In view of the above problems, this application provides methods, systems and equipment for railway snowstorm disaster management.

[0007] Firstly, this application provides a railway snowstorm disaster management method, implemented through a railway snowstorm disaster management system, comprising: collecting topographic distribution data and forest canopy data along the railway line; dividing the railway line into grid units; calculating the topographic factor and shading factor of each grid unit according to the topographic distribution data and forest canopy data; connecting to a meteorological data management system; calculating the snow accumulation prediction index of each grid unit in the expected prediction window based on real-time meteorological input, topographic factor, and shading factor; constructing a local accumulation feature model based on the snowblowing equipment array distributed along the railway line; performing local accumulation analysis on the snow accumulation prediction index using the local accumulation feature model to obtain the local snow accumulation prediction index; optimizing the control of the snowblowing equipment array according to the local snow accumulation prediction index to obtain a set of control parameters; and distributing the set of control parameters to the snowblowing equipment array to execute snowblowing control.

[0008] Preferably, the railway wind-blown snow disaster management method further includes: wherein the terrain distribution data and the forest canopy data are data acquired based on lidar scanning; the terrain factor of each grid unit is calculated according to the terrain distribution data, wherein the terrain factor is a comprehensive weighted factor of elevation normalization, slope component, concavity and convexity, and wind direction coupling degree; and the shading factor of each grid unit is calculated according to the forest canopy data, wherein the shading factor is a comprehensive weighted factor of canopy height, leaf area index, and local opening angle.

[0009] Preferably, the railway snowstorm disaster management method further includes: the real-time meteorological input includes snowfall rate, ground wind speed profile, real-time wind direction, temperature, and relative humidity; the input real-time meteorological input, terrain factors, and shading factors are predicted by a discretized snow accumulation prediction model to obtain subsidence snow accumulation prediction index, lateral transport snow accumulation prediction index, and local snow loss prediction index; the snow accumulation prediction index of each grid cell in the desired prediction window is calculated using the subsidence snow accumulation prediction index, lateral transport snow accumulation prediction index, and local snow loss prediction index.

[0010] Preferably, the railway snowfall disaster management method further includes: the discretized snow accumulation prediction model includes a wind speed transport deposition table, which is obtained by simulating different test wind speed samples using input terrain factor samples and shading factor samples to obtain lateral transport deposition simulation data; the wind speed transport deposition table is obtained according to the correspondence between the test wind speed samples, transport simulation data and deposition simulation data included in the lateral transport deposition simulation data; and the lateral transport snow accumulation prediction index is output by querying the wind speed transport deposition table.

[0011] Preferably, the railway snowfall disaster management method further includes: the discretized snow accumulation prediction model includes a correction function, which is used to calculate the error between the model prediction result sample output by the discretized snow accumulation prediction model and the actual detection result sample to obtain a prediction error sample; and to perform model feedback optimization according to the prediction error sample to obtain an optimized discretized snow accumulation prediction model.

[0012] Preferably, the railway snow blowing disaster management method further includes: collecting the laying position of each snow blowing device in the snow blowing equipment array; extracting the position boundary features and track laying features of each snow blowing device according to the laying position; matching the snow accumulation prediction index corresponding to the grid cell to which each snow blowing device belongs according to the laying position of each snow blowing device; and the local accumulation feature model performing local accumulation analysis on the snow accumulation prediction index corresponding to the grid cell to which each snow blowing device belongs according to the position boundary features and track laying features of each snow blowing device, to obtain the local snow accumulation prediction index corresponding to each snow blowing device in the snow blowing equipment array.

[0013] Preferably, the railway snowstorm disaster management method further includes: collecting location boundary feature samples and track laying feature samples of multiple types of laying locations, as well as snow accumulation prediction index samples and historical snow accumulation detection indexes corresponding to the multiple types of laying locations; parameterizing the location boundary feature samples and track laying feature samples of the multiple types of laying locations, and outputting parameterized feature samples; using the historical snow accumulation detection index as a supervision label, and using the snow accumulation prediction index samples and the parameterized feature samples corresponding to the multiple types of laying locations as inputs for labeled supervised training, until the error between the historical snow accumulation detection index and the trained output local snow accumulation prediction index is less than a preset threshold, and outputting a local accumulation feature model.

[0014] Preferably, the railway snowfall disaster management method further includes: establishing an optimization objective, which includes snow accumulation risk and total energy consumption control, wherein the snow accumulation risk is obtained by calculating the sum of squares of the excess amounts between the local snow accumulation prediction index and the local snow accumulation safety threshold; obtaining the control parameters of each snow blowing device in the snow blowing equipment array, including the jet angle, on / off status, and power level; optimizing the control parameters of each snow blowing device using the optimization objective, and calculating the real-time snow accumulation risk of each candidate control parameter set in real time until a candidate control parameter set below the local snow accumulation safety threshold is obtained and output as the control parameter set of the snow blowing equipment array.

[0015] Secondly, this application also provides a railway snowstorm disaster management system for implementing the railway snowstorm disaster management method as described in the first aspect, comprising: a grid cell segmentation module for collecting topographic distribution data and forest canopy data along the railway line, segmenting the railway line into grid cells, and calculating the topographic factor and shading factor of each grid cell according to the topographic distribution data and forest canopy data; a prediction index calculation module for accessing a meteorological data management system, calculating the snow accumulation prediction index of each grid cell in the desired prediction window based on real-time meteorological input, topographic factor, and shading factor; a prediction index acquisition module for constructing a local accumulation feature model based on the snowblowing equipment array distributed along the railway line, performing local accumulation analysis on the snow accumulation prediction index using the local accumulation feature model, and obtaining the local snow accumulation prediction index; and a snowblowing control module for optimizing the control of the snowblowing equipment array according to the local snow accumulation prediction index to obtain a set of control parameters, and sending the set of control parameters to the snowblowing equipment array to execute snowblowing control.

[0016] Thirdly, this application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the railway wind and snow disaster management method described in any of the first aspects above.

[0017] The technical solution provided in this application has at least the following technical effects or advantages: by achieving the technical goal of refined prediction and dynamic response control of snowfall disasters along railway lines, it achieves the technical effects of improving the prediction accuracy of snow accumulation through multi-source data fusion modeling, realizing the coordinated regulation and optimal energy consumption allocation of snow blowing equipment arrays through intelligent optimization algorithms, thereby improving the safe passage capability of railway lines under extreme weather conditions, reducing the energy consumption cost of snow removal systems, and enhancing the adaptive operation capability of equipment groups.

[0018] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

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

[0020] Figure 1 This is a flowchart illustrating the railway wind and snow disaster management method proposed in this application.

[0021] Figure 2 This is a schematic diagram of the railway wind and snow disaster management system of this application.

[0022] Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application.

[0023] Figure labeling: 1. Grid cell segmentation module; 2. Prediction index calculation module; 3. Prediction index acquisition module; 4. Snow blowing control module; 300. Bus; 301. Receiver; 302. Processor; 303. Transmitter; 304. Memory; 305. Bus interface. Detailed Implementation

[0024] This application provides a method, system, and equipment for railway snowfall disaster management. It addresses the technical problems in existing technologies where the lack of a joint modeling mechanism and control optimization strategy based on terrain, forest belts, and real-time meteorological data leads to insufficient accuracy in predicting snowfall accumulation along railway lines, untimely control response of snow-blowing equipment, and unreasonable energy consumption management. These issues further affect the operational safety, snow removal efficiency, and reliability of equipment collaborative control in high-altitude and cold regions under severe weather conditions. The application achieves the technical goal of refined prediction and dynamic response control of snowfall disasters along railway lines. It improves the accuracy of snowfall accumulation prediction through multi-source data fusion modeling and utilizes intelligent optimization algorithms to achieve collaborative control and optimal energy consumption allocation of snow-blowing equipment arrays. This results in improved safe passage of railway lines under extreme weather conditions, reduced energy costs of snow removal systems, and enhanced adaptive operation capabilities of equipment groups.

[0025] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0026] Example 1, please refer to the appendix. Figure 1 This application provides a railway snowstorm disaster management method, which is applied to the railway snowstorm disaster management system, and specifically includes the following steps: S1: Collect topographic distribution data and forest canopy data along the railway line, divide the railway line into grid cells, and calculate the topographic factor and shading factor of each grid cell according to the topographic distribution data and forest canopy data.

[0027] Furthermore, this application also includes: wherein the topographic distribution data and the forest canopy data are data acquired based on lidar scanning; the topographic factor of each grid cell is calculated according to the topographic distribution data, wherein the topographic factor is a comprehensive weighted factor of elevation normalization, slope component, concavity and convexity, and wind direction coupling; and the shading factor of each grid cell is calculated according to the forest canopy data, wherein the shading factor is a comprehensive weighted factor of canopy height, leaf area index, and local opening angle.

[0028] Specifically, topographic distribution data and forest canopy data along the railway line will be collected. The topographic distribution data and forest canopy data will be acquired based on lidar scanning, meaning they will be precisely obtained through a lidar system that actively transmits and receives reflected signals. The topographic distribution data reflects changes in ground elevation, slope morphology, and surface undulation, while the forest canopy data reflects the height, density, and structural characteristics of protective forests or natural vegetation on both sides of the railway. By acquiring this data, a spatial information foundation reflecting the natural conditions along the railway line can be established, providing realistic environmental input for subsequent snow accumulation and wind field simulations.

[0029] Grid segmentation along a railway line involves dividing the entire railway area into several adjacent smaller regions, each called a grid cell. This discretizes the continuous spatial environment, facilitating computation and analysis. Each grid cell can be viewed as an independent analysis unit, with its terrain and forest features approximated as homogeneous.

[0030] The topographic factor and shading factor for each grid cell are calculated based on topographic distribution data and forest canopy data, representing two key environmental indicators reflecting the impact of wind and snow in each grid. Specifically, the topographic factor for each grid cell is calculated based on topographic distribution data. This topographic factor is a weighted average of elevation normalization, slope component, concavity / convexity, and wind direction coupling, quantifying the degree to which different topographic features affect wind and snow transport in each grid. Elevation normalization balances the overall topographic differences across different elevation zones; the slope component describes the effect of ground tilt direction and angle on snow flow slippage and wind field guidance; concavity / convexity reflects the interference of surface undulation on snow particle retention or accumulation; and wind direction coupling measures the degree of matching between topography and the prevailing wind direction. Then, the shading factor for each grid cell is calculated based on forest canopy data. This shading factor is a weighted average of canopy height, leaf area index, and local opening angle, representing the weakening effect of the forest belt on the wind field and its blocking effect on snow flow. The higher the canopy height, the stronger the wind shading effect of the forest belt; the larger the leaf area index, the denser the leaf coverage per unit area, and the lower the airflow permeability; the local opening angle represents the angle of the gap between the forest belt canopy. When the opening angle is large, the wind can form a local acceleration channel through the gap, reducing the overall shading efficiency.

[0031] S2: Connect to the meteorological data management system and calculate the snow accumulation prediction index for each grid cell in the expected prediction window based on real-time meteorological input, terrain factors, and shading factors.

[0032] Furthermore, this application also includes: the real-time meteorological input includes snowfall rate, surface wind speed profile, real-time wind direction, temperature, and relative humidity; the input real-time meteorological input, terrain factors, and shading factors are predicted by a discretized snow accumulation prediction model to obtain subsidence snow accumulation prediction index, lateral transport snow accumulation prediction index, and local snow loss prediction index; the snow accumulation prediction index of each grid cell in the desired prediction window is calculated using the subsidence snow accumulation prediction index, lateral transport snow accumulation prediction index, and local snow loss prediction index.

[0033] Furthermore, this application also includes: the discretized snow accumulation prediction model includes a wind speed transport deposition table, which is obtained by simulating lateral transport deposition simulation data under different test wind speed samples by inputting topographic factor samples and shading factor samples; the wind speed transport deposition table is obtained according to the correspondence between the test wind speed samples, transport simulation data and deposition simulation data included in the lateral transport deposition simulation data; and the lateral transport snow accumulation prediction index is output by querying the wind speed transport deposition table.

[0034] Furthermore, this application also includes: the discretized snow accumulation prediction model includes a correction function, which is used to calculate the error between the model prediction result sample output by the discretized snow accumulation prediction model and the actual detection result sample to obtain a prediction error sample; and to perform model feedback optimization according to the prediction error sample to obtain an optimized discretized snow accumulation prediction model.

[0035] Specifically, accessing the meteorological data management system refers to the interaction of data between environmental monitoring and meteorological forecasting systems along the railway line, acquiring meteorological monitoring data in real time through interfaces and combining it with existing geographic information. The meteorological data management system integrates multi-source meteorological information from automatic weather stations, satellite remote sensing, and numerical weather prediction models, providing dynamic meteorological input.

[0036] Real-time meteorological inputs, including snowfall rate, surface wind speed profile, real-time wind direction, temperature, and relative humidity, collectively determine the formation, transport, and deposition of snow particles in the air. Snowfall rate describes the intensity of snowfall per unit time and is a direct factor affecting the rate of snow accumulation; the surface wind speed profile indicates the variation in wind speed at different altitudes and can be used to infer the wind's ability to transport snow particles; real-time wind direction determines the main direction of snow particle movement; temperature affects the melting and refreezing behavior of snow particles; and relative humidity reflects the water vapor content in the air, thus affecting the density and cohesiveness of snow particles.

[0037] By using a discretized snow accumulation prediction model to predict real-time meteorological inputs, topographic factors, and shading factors, the model discretizes continuously changing meteorological and environmental variables into computable feature units and uses mathematical equations or data-driven algorithms to infer snow accumulation trends. The discretized snow accumulation prediction model comprehensively considers the dynamic effects of meteorological factors on snow particle formation and transport, as well as the blocking and guiding effects of topography and forest belts, thus outputting snow accumulation prediction results reflecting different mechanisms. This leads to the acquisition of subsidence snow accumulation prediction indices, lateral transport snow accumulation prediction indices, and local snow loss prediction indices, indicating that the discretized snow accumulation prediction model decomposes the snow accumulation process into three main physical mechanisms: subsidence, transport, and loss. The subsidence snow accumulation prediction index reflects the rate and thickness of snow accumulation formed by snow particles directly falling to the ground; the lateral transport snow accumulation prediction index describes the trend of snow particles migrating along the surface and redepositing in certain areas due to wind; and the local snow loss prediction index considers snow reduction caused by evaporation, sublimation, or wind erosion.

[0038] Calculating the snow accumulation prediction index for each grid cell within the expected prediction window using subsidence snow accumulation prediction index, lateral transport snow accumulation prediction index, and local snow loss prediction index involves weighting or combining these three sub-indicators to obtain the total snow accumulation value for each grid cell within a certain time range. The expected prediction window is for the next few hours or days, used to predict the snow accumulation trend during this period. Through comprehensive calculation, areas more likely to accumulate snow in the future can be identified, thus providing an early warning basis for railway operation safety.

[0039] Furthermore, the discretized snow accumulation prediction model includes a wind speed transport and deposition table, which represents an empirical relationship table embedded in the model, derived from simulation, to describe the quantitative relationship between wind speed and snow grain transport and deposition. The wind speed transport and deposition table obtains lateral transport and deposition simulation data by inputting different topographic factor samples and shading factor samples, and conducting numerical simulations or wind tunnel tests under multiple test wind speed conditions. This data reflects the impact of wind speed on snow flow paths and deposition thickness under different environmental combinations.

[0040] Based on the correspondence between lateral transport and deposition simulation data, including test wind speed samples, transport simulation data, and deposition simulation data, a wind speed transport and deposition table is obtained. This means establishing a corresponding mapping relationship between wind speed inputs and corresponding transport and deposition outputs. Furthermore, by simply using real-time wind speeds and local topographic parameters, the corresponding lateral transport snow accumulation prediction indicators can be quickly obtained through table lookup, significantly improving computational efficiency and real-time performance.

[0041] Using the wind speed transport deposition table to query and output the lateral transport snow accumulation prediction index refers to the process in which, after obtaining real-time wind speed data and the corresponding grid's topography and forest belt parameters, the snow volume change estimate under the influence of lateral transport is quickly obtained by matching the data pairs in the wind speed transport deposition table.

[0042] Furthermore, the discretized snow accumulation prediction model includes a correction function, which is used to calculate the error between the model prediction results and the actual detection results. The correction function calculates the prediction error sample by comparing the deviation between the model prediction results and the field detection or historical monitoring data, reflecting the sources and distribution characteristics of the deviation of the discretized snow accumulation prediction model under different environmental conditions.

[0043] The model is optimized through feedback based on prediction error samples, resulting in an optimized discretized snow accumulation prediction model. This represents an iterative update process that continuously improves prediction accuracy. The model feedback optimization process can employ methods such as gradient descent, least squares, or Bayesian updates to gradually reduce the error between predictions and actual measurements. As the amount of training data increases, the model's adaptability to different terrain and meteorological conditions also improves, leading to higher reliability and generalization performance in future predictions.

[0044] S3: Construct a local accumulation feature model based on the snow blowing equipment array distributed along the railway line, and use the local accumulation feature model to perform local accumulation analysis on the snow accumulation prediction index to obtain the local snow accumulation prediction index.

[0045] Furthermore, this application also includes: collecting the laying position of each snow blowing device in the snow blowing equipment array; extracting the position boundary features and track laying features of each snow blowing device according to the laying position; matching the snow accumulation prediction index corresponding to the grid cell to which each snow blowing device belongs according to the laying position of each snow blowing device; and performing local accumulation analysis on the snow accumulation prediction index corresponding to the grid cell to which each snow blowing device belongs according to the position boundary features and track laying features of each snow blowing device, thereby obtaining the local snow accumulation prediction index corresponding to each snow blowing device in the snow blowing equipment array.

[0046] Furthermore, this application also includes: collecting location boundary feature samples and track laying feature samples of multiple types of laying locations, as well as snow accumulation prediction index samples and historical snow accumulation detection indexes corresponding to the multiple types of laying locations; parameterizing the location boundary feature samples and track laying feature samples of the multiple types of laying locations, and outputting parameterized feature samples; using the historical snow accumulation detection index as a supervision label, and using the snow accumulation prediction index samples and the parameterized feature samples corresponding to the multiple types of laying locations as inputs for labeled supervised training, until the error between the historical snow accumulation detection index and the trained output local snow accumulation prediction index is less than a preset threshold, and outputting a local accumulation feature model.

[0047] Specifically, collecting location boundary feature samples and track laying feature samples for various snow blowing locations refers to identifying the installation locations of different types of snow blowing equipment along the railway line, such as straight sections, curved sections, tunnel entrances, and bridge sections. Location boundary feature samples include the distance of the laying location relative to the track centerline, the height of surrounding terrain undulations, and the distribution range of nearby structures; these characteristics determine the airflow pattern in that area. Track laying feature samples reflect engineering parameters such as sleeper type, track gauge, and track surface height difference, which influence the way snow adheres to and accumulates on the track. Simultaneously, it is necessary to collect snow accumulation prediction index samples and historical snow accumulation detection indexes corresponding to different laying locations.

[0048] Parameterizing the location boundary feature samples and track laying feature samples of various laying locations transforms the originally complex and unstructured geometric and physical information into computable numerical feature vectors. The parameterization process includes quantizing, normalizing, and encoding the features; for example, representing terrain slope as an angle value, sleeper density as a quantity per meter, and track gauge deviation in millimeters. The output parameterized feature samples can serve as input for subsequent model training, enabling the model to identify the implicit patterns and interactions between the features.

[0049] Using historical snow accumulation detection metrics as supervisory labels means using historical measured data as a true reference for the local snow accumulation feature model's learning, guiding its training direction. The supervisory labels measure the accuracy of the model's output; by comparing it with predicted values, the model can gradually adjust its internal weights to minimize error. Labeled supervised training uses snow accumulation prediction metric samples and parameterized feature samples corresponding to multiple paving locations as input. This means considering both environmental features and model predictions at the input, allowing the local snow accumulation feature model to learn not only the relationships between features but also the distribution of prediction errors during training. Labeled supervised training employs machine learning algorithms, such as gradient descent or random forests, iteratively updating and gradually optimizing prediction performance. The local snow accumulation feature model is output when the error between the historical snow accumulation detection metrics and the trained output snow accumulation prediction metrics falls below a preset threshold. This means the training process will continue until the deviation between the local snow accumulation feature model's output and the actual detection results is within an acceptable range. The preset threshold is set according to safety standards or control accuracy, such as an error of less than 5% or a snow thickness error of less than 2 cm. When the error meets this requirement, a local accumulation feature model is output, which can then predict the possible local snow accumulation in the future based on the structural and environmental characteristics of different laying locations, thereby achieving pre-optimization of the snow blowing equipment control strategy.

[0050] Furthermore, the installation location of each snow-blowing device in the snow-blowing equipment array is collected. Based on this location, the positional boundary features and track laying characteristics of each snow-blowing device are extracted. This refers to obtaining the specific installation location of each snow-blowing device along the track line in a railway snow control system using satellite positioning systems or track mapping data. Positional boundary features describe the spatial distribution of the snow-blowing equipment, such as its distance from the track centerline, the spacing between adjacent devices, and its relative relationship with structures like sleepers and ballast. Track laying characteristics refer to the structural information of the track itself, including track type, laying method, rail gradient, and gauge. Together, these factors determine the spatial relationship between the equipment and the track environment, providing a geometric basis for subsequent snow accumulation prediction.

[0051] By matching the placement location of each snowblower with the corresponding snow accumulation prediction index of its assigned grid cell, each snowblower is spatially mapped to a pre-defined grid cell. Each grid cell has a snow accumulation prediction index, calculated based on multiple factors such as terrain, weather, and wind transport, reflecting the potential future snow accumulation at that location. Location matching allows each device to be correlated with the snow accumulation trend of its environment, enabling personalized snow removal strategies. For example, if a grid cell predicts a large snowfall, the corresponding device can enter high-power operation in advance, while devices in areas with less snowfall can remain in low-energy standby mode, thereby improving the overall system's energy efficiency.

[0052] The local snow accumulation feature model performs local snow accumulation analysis on the snow accumulation prediction index corresponding to the grid cell of each snowblower based on its location boundary characteristics and track laying characteristics. This obtains the local snow accumulation prediction index for each snowblower in the array, meaning that by using machine learning or regression analysis models, the geometric features of the snowblower are combined with environmental prediction data to analyze the snow accumulation characteristics of the local area where the snowblower is located. The core of the local snow accumulation feature model lies in fitting the microclimate and topographic effects of the area where each snowblower is located, thereby outputting a more refined local snow accumulation estimate than the global prediction. Furthermore, it can not only identify the overall snow accumulation trend but also capture local snow accumulation differences caused by wind direction deviations, track elevation differences, or shading structures.

[0053] S4: The control parameter set is obtained by optimizing the snow blowing equipment array according to the snow accumulation prediction index, and the control parameter set is sent to the snow blowing equipment array to perform snow blowing control.

[0054] Furthermore, this application also includes: establishing an optimization objective, which includes snow accumulation risk and total energy consumption for control, wherein the snow accumulation risk is obtained by calculating the sum of squares of the excess between the snow accumulation prediction index and the snow accumulation safety threshold; obtaining the control parameters of each snow blowing device in the snow blowing equipment array, including jet angle, on / off state, and power level; optimizing the control parameters of each snow blowing device with the optimization objective, calculating the real-time snow accumulation risk of each candidate control parameter set in real time, until a candidate control parameter set below the snow accumulation safety threshold is obtained and output as the control parameter set of the snow blowing equipment array.

[0055] Specifically, an optimization objective is established, encompassing snow accumulation risk and total energy consumption control. This indicates that a balance between safety and energy efficiency must be considered simultaneously during system control. The optimization objective is a multi-objective function in a mathematical sense, aiming to minimize snow accumulation risk while reducing the overall energy consumption of snowblowing equipment. Snow accumulation risk measures the degree to which snow accumulation in a given area may exceed safety limits, serving as a crucial parameter for assessing railway operational safety. Total energy consumption control reflects the energy consumed by snowblowing equipment during operation, a key indicator of system economics. By combining these two indicators into a single optimization objective, energy waste can be minimized while ensuring safety. For example, in low wind speeds and high temperatures, equipment power can be automatically reduced to decrease energy consumption; however, in blizzard conditions, priority will be given to ensuring that risk indicators meet targets, even if it requires greater energy consumption to maintain safety.

[0056] Among them, the risk of snow accumulation is obtained by calculating the sum of squares of the excess between the predicted local snow accumulation index and the local snow accumulation safety threshold. This means that the risk quantification adopts the form of squared loss, that is, when the predicted local snow accumulation exceeds the set safety threshold, the excess part will be squared and accumulated to highlight the risk of a large excess.

[0057] Obtaining the control parameters of each snow blower in the snow blowing equipment array, including jet angle, on / off status, and power level, means that the operating status and adjustable parameters of each snow blower need to be collected before executing optimized control. The jet angle determines the airflow direction and has a significant impact on the snow removal path; the on / off status indicates whether the equipment is activated and is the most basic control signal; the power level indicates the strength of the equipment's output energy and directly affects snow removal efficiency and energy consumption.

[0058] The control parameters of each snowblowing device are optimized to achieve the desired effect. The real-time risk of snow accumulation for each candidate control parameter set is calculated continuously until a set of candidate control parameters below the local snow accumulation safety threshold is obtained. This set of candidate control parameters is then output as the control parameter set for the snowblowing device array. This represents the dynamic adjustment of control parameters using iterative optimization algorithms such as genetic algorithms, particle swarm optimization, or gradient descent to simultaneously meet safety and energy consumption objectives. The optimization process is cyclical. In each round, a set of candidate control parameters is generated, and the corresponding real-time risk value is predicted using a model. If the risk is still higher than the threshold, the parameters are adjusted further. Finally, when the predicted local snow accumulation values ​​of all snowblowing devices are lower than the safety threshold, the current optimal control parameter set is output as the execution plan.

[0059] Furthermore, the set of control parameters is sent to the snow-blowing equipment array for snow-blowing control. This means that after optimization calculations, an optimal set of control parameters is obtained and sent to all snow-blowing equipment to synchronously execute the snow removal task. The sending process is completed through industrial communication networks, such as Ethernet, wireless control links, or fieldbus systems, to achieve real-time data interaction between the central control system and the snow-blowing equipment. The snow-blowing equipment array refers to multiple snow-blowing devices distributed along the railway line, located in key sections such as bridges, tunnel entrances, curves, and areas prone to snow accumulation, forming a coordinated control network. Executing snow-blowing control means that the snow-blowing equipment starts or adjusts its operating status according to the received parameter commands, such as changing the jet angle to adapt to wind direction or adjusting the power level to control airflow intensity, ultimately achieving snow removal and ensuring track safety.

[0060] In summary, the railway snowstorm disaster management method provided in this application has the following technical effects: by achieving the technical goal of refined prediction and dynamic response control of snowstorm disasters along railway lines, it achieves the technical effects of improving the prediction accuracy of snow accumulation through multi-source data fusion modeling, realizing the coordinated regulation and optimal energy consumption allocation of snow blowing equipment arrays through intelligent optimization algorithms, thereby improving the safe passage capability of railway lines under extreme weather conditions, reducing the energy consumption cost of snow removal systems, and enhancing the adaptive operation capability of equipment groups.

[0061] Example 2: Based on the same inventive concept as the railway snowstorm disaster management method in the foregoing examples, this application also provides a railway snowstorm disaster management system. Please refer to the appendix. Figure 2The system includes: a grid cell segmentation module 1, used to collect topographic distribution data and forest canopy data along the railway line, segment the railway line into grid cells, and calculate the topographic factor and shading factor of each grid cell according to the topographic distribution data and forest canopy data; a prediction index calculation module 2, used to connect to the meteorological data management system, and calculate the snow accumulation prediction index of each grid cell in the expected prediction window according to real-time meteorological input, topographic factor, and shading factor; a prediction index acquisition module 3, used to construct a local accumulation feature model based on the snow blowing equipment array distributed along the railway line, and use the local accumulation feature model to perform local accumulation analysis on the snow accumulation prediction index to obtain the local snow accumulation prediction index; and a snow blowing control module 4, used to optimize the control of the snow blowing equipment array according to the local snow accumulation prediction index to obtain a set of control parameters, and send the set of control parameters to the snow blowing equipment array to execute snow blowing control.

[0062] Furthermore, the railway wind-blown snow disaster management system is also used for: wherein the terrain distribution data and the forest canopy data are data acquired based on lidar scanning; calculating the terrain factor of each grid unit according to the terrain distribution data, wherein the terrain factor is a comprehensive weighted factor of elevation normalization, slope component, concavity and convexity, and wind direction coupling degree; and calculating the shading factor of each grid unit according to the forest canopy data, wherein the shading factor is a comprehensive weighted factor of canopy height, leaf area index, and local opening angle.

[0063] Furthermore, the railway snowstorm disaster management system is also used for: the real-time meteorological input including snowfall rate, ground wind speed profile, real-time wind direction, temperature, and relative humidity; predicting the input real-time meteorological input, terrain factors, and shading factors through a discretized snow accumulation prediction model to obtain subsidence snow accumulation prediction indicators, lateral transport snow accumulation prediction indicators, and local snow loss prediction indicators; and calculating the snow accumulation prediction indicators for each grid unit within the desired prediction window using the subsidence snow accumulation prediction indicators, lateral transport snow accumulation prediction indicators, and local snow loss prediction indicators.

[0064] Furthermore, the railway snowstorm disaster management system is also used for: the discretized snow accumulation prediction model including a wind speed transport deposition table, which is obtained by simulating different test wind speed samples by inputting terrain factor samples and shading factor samples to obtain lateral transport deposition simulation data; according to the correspondence between the lateral transport deposition simulation data including test wind speed samples, transport simulation data and deposition simulation data, a wind speed transport deposition table is obtained; and the lateral transport snow accumulation prediction index is output by querying the wind speed transport deposition table.

[0065] Furthermore, the railway snowstorm disaster management system is also used for: the discretized snow accumulation prediction model includes a correction function, which is used to calculate the error between the model prediction result sample output by the discretized snow accumulation prediction model and the actual detection result sample to obtain a prediction error sample; and to perform model feedback optimization according to the prediction error sample to obtain an optimized discretized snow accumulation prediction model.

[0066] Furthermore, the railway snowstorm disaster management system is also used for: collecting the laying position of each snowblowing device in the snowblowing equipment array; extracting the position boundary features and track laying features of each snowblowing device according to the laying position; matching the snow accumulation prediction index corresponding to the grid unit to which each snowblowing device belongs according to the laying position of each snowblowing device; and performing local accumulation analysis on the snow accumulation prediction index corresponding to the grid unit to which each snowblowing device belongs according to the position boundary features and track laying features of each snowblowing device, thereby obtaining the local snow accumulation prediction index corresponding to each snowblowing device in the snowblowing equipment array.

[0067] Furthermore, the railway snowstorm disaster management system is also used for: collecting location boundary feature samples and track laying feature samples of multiple types of laying locations, as well as snow accumulation prediction index samples and historical snow accumulation detection indexes corresponding to the multiple types of laying locations; parameterizing the location boundary feature samples and track laying feature samples of the multiple types of laying locations, and outputting parameterized feature samples; using the historical snow accumulation detection index as a supervision label, and using the snow accumulation prediction index samples and the parameterized feature samples corresponding to the multiple types of laying locations as inputs for labeled supervised training, until the error between the historical snow accumulation detection index and the trained output local snow accumulation prediction index is less than a preset threshold, and outputting a local accumulation feature model.

[0068] Furthermore, the railway snowstorm disaster management system is also used for: establishing optimization objectives, which include snow accumulation risk and total energy consumption control, wherein the snow accumulation risk is obtained by calculating the sum of squares of the excess amounts between the local snow accumulation prediction index and the local snow accumulation safety threshold; obtaining the control parameters of each snowblowing device in the snowblowing equipment array, including jet angle, on / off status, and power level; optimizing the control parameters of each snowblowing device according to the optimization objectives, calculating the real-time snow accumulation risk of each candidate control parameter set in real time, until a candidate control parameter set below the local snow accumulation safety threshold is obtained and output as the control parameter set of the snowblowing equipment array.

[0069] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The railway snowstorm disaster management method and specific examples in the aforementioned embodiment one are also applicable to the railway snowstorm disaster management system in this embodiment. Through the foregoing detailed description of the railway snowstorm disaster management method, those skilled in the art can clearly understand the railway snowstorm disaster management system in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0070] Example 3: Based on the inventive concept of the railway snowstorm disaster management method in the foregoing examples, this application also provides an electronic device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the railway snowstorm disaster management method described in any one of Examples 1 above.

[0071] Appendix Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application. Figure 3 In this document, the bus architecture is represented by bus 300. Bus 300 may include any number of interconnected buses and bridges, and bus 300 connects various circuits including one or more processors represented by processor 302 and memory represented by memory 304. Bus 300 may also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, i.e., a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 can be used to store data used by processor 302 during operation.

[0072] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0073] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for managing railway wind-blowing snow disasters, characterized by, The method comprises: Collecting terrain distribution data and forest belt canopy data along the railway line, dividing the railway line into grid units, and calculating terrain factors and shelter factors of each grid unit according to the terrain distribution data and the forest belt canopy data; Accessing a meteorological data management system, and calculating snow accumulation prediction indexes of each grid unit in a desired prediction window according to real-time meteorological input, terrain factors and shelter factors; Building a local accumulation feature model according to an array of snow blowing equipment distributed along the railway line, performing local accumulation analysis on the snow accumulation prediction indexes by using the local accumulation feature model, and obtaining snow local accumulation prediction indexes; Controlling and optimizing the array of snow blowing equipment according to the snow local accumulation prediction indexes to obtain a control parameter set, and issuing the control parameter set to the array of snow blowing equipment to perform snow blowing control.

2. The method of claim 1, wherein, The method for calculating terrain factors and shelter factors of each grid unit according to the terrain distribution data and the forest belt canopy data comprises: The terrain distribution data and the forest belt canopy data are data obtained based on laser radar scanning; The terrain factors of each grid unit are calculated according to the terrain distribution data, and the terrain factors are comprehensive weighted factors of elevation normalization, slope component, concave-convex degree and wind direction coupling degree; The shelter factors of each grid unit are calculated according to the forest belt canopy data, and the shelter factors are comprehensive weighted factors of canopy height, leaf area index and local opening angle. The method for calculating snow accumulation prediction indexes of each grid unit in a desired prediction window according to real-time meteorological input, terrain factors and shelter factors comprises:

3. The method of claim 1, wherein, The real-time meteorological input includes snowfall rate, ground wind speed profile, real-time wind direction, temperature and relative humidity; The input real-time meteorological input, terrain factors and shelter factors are predicted by a discrete snow accumulation prediction model to obtain deposition snow accumulation prediction indexes, lateral transport snow accumulation prediction indexes and local snow loss prediction indexes; The snow accumulation prediction indexes of each grid unit in the desired prediction window are calculated by using the deposition snow accumulation prediction indexes, the lateral transport snow accumulation prediction indexes and the local snow loss prediction indexes. The discrete snow accumulation prediction model includes a wind speed transport deposition table, the wind speed transport deposition table is simulated under different test wind speed samples by inputting terrain factor samples and shelter factor samples to obtain lateral transport deposition simulation data; 4. The method of claim 3, wherein, According to the correspondence relationship of the lateral transport deposition simulation data including test wind speed samples, transport simulation data and deposition simulation data, the wind speed transport deposition table is obtained; The wind speed transport deposition table is used for query output of the lateral transport snow accumulation prediction indexes. The discrete snow accumulation prediction model includes a correction function, the correction function is used for error calculation of model prediction result samples and actual detection result samples output by the discrete snow accumulation prediction model to obtain prediction error samples; 5. The method of claim 3, wherein the wind-blow snow hazard management method is characterized by, According to the prediction error samples, model feedback optimization is performed to obtain an optimized discrete snow accumulation prediction model. ​ 6. The method of claim 1, wherein, The local accumulation characteristic model is used for local accumulation analysis on the snow accumulation prediction index, and the method comprises the following steps: Collecting the laying position of each snow blowing device in the snow blowing device array, and extracting the position boundary characteristic and track laying characteristic of each snow blowing device according to the laying position; Matching the snow accumulation prediction index corresponding to the grid unit according to the laying position of each snow blowing device; The local accumulation characteristic model is used for local accumulation analysis on the snow accumulation prediction index corresponding to the grid unit according to the position boundary characteristic and track laying characteristic of each snow blowing device, and the snow accumulation prediction index corresponding to each snow blowing device in the snow blowing device array is obtained.

7. The method of claim 6, wherein, The method for constructing the local accumulation characteristic model comprises the following steps: Collecting the position boundary characteristic sample and track laying characteristic sample of multiple types of laying positions, and the snow accumulation prediction index sample and historical snow accumulation detection index corresponding to the multiple types of laying positions; Parameterizing the position boundary characteristic sample and track laying characteristic sample of the multiple types of laying positions, and outputting the parameterized characteristic sample; Taking the historical snow accumulation detection index as a supervision label, taking the snow accumulation prediction index sample corresponding to the multiple types of laying positions and the parameterized characteristic sample as input for labelization supervision training until the error between the historical snow accumulation detection index and the snow accumulation prediction index output by the training is less than a preset threshold, and outputting the local accumulation characteristic model.

8. The method of claim 1, wherein, The snow blowing device array is controlled and optimized according to the snow accumulation prediction index to obtain a control parameter set, and the method comprises the following steps: An optimization target is established, and the optimization target comprises a snow accumulation risk and a total control energy consumption, wherein the snow accumulation risk is obtained by calculating the sum of squares of the excess amount between the snow accumulation prediction index and the snow accumulation safety threshold; Control parameters of each snow blowing device in the snow blowing device array are obtained, including a jet angle, an on-off state and a power level; The control parameters of each snow blowing device are controlled and optimized according to the optimization target, the snow accumulation real-time risk of each candidate control parameter set is calculated in real time until a candidate control parameter set less than the snow accumulation safety threshold is obtained and output as the control parameter set of the snow blowing device array.

9. A railway wind-blowing snow disaster management system characterized by The steps for implementing the railway wind snow disaster management method in any one of claims 1 to 8 comprise: A grid unit segmentation module is configured to collect terrain distribution data and forest canopy data along a railway line, segment the railway line into grid units, and calculate terrain factors and shielding factors of each grid unit according to the terrain distribution data and forest canopy data; A prediction index calculation module is configured to access a meteorological data management system, calculate a snow accumulation prediction index of each grid unit in a desired prediction window according to real-time meteorological input, terrain factors and shielding factors; A prediction index acquisition module is configured to construct a local accumulation characteristic model according to a snow blowing device array distributed on the railway line, use the local accumulation characteristic model to perform local accumulation analysis on the snow accumulation prediction index, and obtain a snow accumulation prediction index. A snow blowing control module is configured to control and optimize the snow blowing device array according to the snow accumulation local prediction index to obtain a control parameter set, and to issue the control parameter set to the snow blowing device array to perform snow blowing control.

10. An electronic device, comprising: The method comprises the following steps: at least one processor; a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the railway wind and snow disaster management method according to any one of claims 1 to 8.

Images