Subway station health degree assessment method based on digital twinning and related equipment

Abstract

The invention discloses a subway station health degree assessment method based on digital twinning and related equipment, and relates to the technical field of digital twinning, and the method comprises the steps: obtaining subway station data and a health degree scoring standard system; mapping the subway station data to a preset digital twinborn model to obtain a subway station digital twinborn model; according to the health degree scoring standard system, performing normalization processing on the subway station data to obtain standardized evaluation data; inputting the standardized evaluation data into a pre-trained health degree evaluation model to obtain a health degree index of the subway station; when the health degree index is lower than a preset health degree threshold value, determining a hidden danger point of the subway station; and based on the hidden danger points, hidden danger area contour display is carried out on the subway station digital twin model. By implementing the method, potential risks can be found and processed in time, and the safety and reliability of subway station operation are improved.

Description

Technical Field

[0001] This application relates to the field of digital twin technology, and particularly to a method for evaluating the health of a subway station based on digital twin and related devices. Background Art

[0002] With the acceleration of the urbanization process, the subway, as an important part of public transportation, its safe operation is directly related to the travel safety of passengers and the efficient operation of transportation. Therefore, effective health assessment and maintenance management are crucial for ensuring the long-term stable operation of the subway system.

[0003] Currently, the operation status monitoring of subway stations mainly adopts a combination of sensor network layout and manual inspection. That is, vibration sensors, temperature and humidity sensors and other devices are arranged at key parts of the subway station to collect basic data. Staff regularly inspect the operation of the equipment, and manually analyze and evaluate the collected basic data.

[0004] However, the related technologies have problems of low data analysis efficiency and strong subjectivity of evaluation results in practical applications. Since there are many types of monitoring data involved in subway stations and there are complex correlations between various types of data, it is difficult to detect potential hidden dangers in time through manual analysis and evaluation. Summary of the Invention

[0005] This application provides a method for evaluating the health of a subway station based on digital twin and related devices, which is used to detect and handle potential risks in time and improve the safety and reliability of subway station operation.

[0006] In a first aspect, the present application provides a method for evaluating the health of a subway station based on digital twins, which is applied to a server. The method includes: obtaining subway station data and a health score standard system, where the subway station data includes structural vibration data, temperature and humidity data, passenger flow density data, and equipment operation status data, and the health score standard system includes structural integrity standards, electromechanical equipment standards, environmental suitability standards, and population density standards; mapping the subway station data to a preset digital twin model to obtain a subway station digital twin model, which is used to visually display the real-time operation status of the subway station; normalizing the subway station data according to the health score standard system to obtain standardized evaluation data, which includes structural status scores, equipment efficiency scores, environmental status scores, and population distribution scores; inputting the standardized evaluation data into a pre-trained health evaluation model to obtain the health index of the subway station; when the health index is lower than a preset health threshold, determining potential problem points of the subway station, which include structural damage points, equipment failure points, environmental anomaly points, and / or crowded points; and based on the potential problem points, displaying the outline of the potential problem area in the subway station digital twin model.

[0007] By adopting the above technical solution, the server maps the structural vibration data, temperature and humidity data, passenger flow density data, and equipment operation status data to a preset digital twin model to obtain a subway station digital twin model. The server combines the health score standard system and the health evaluation model, which can realize the comprehensive monitoring and evaluation of the real-time operation status of the subway station, and visually display various potential problem points in the subway station digital twin model, which helps to timely discover and handle potential risks and improve the safety and reliability of the subway station operation.

[0008] In combination with some embodiments of the first aspect, in some embodiments, before the step of mapping the subway station data to a preset digital twin model to obtain a subway station digital twin model, which is used to visually display the real-time operation status of the subway station, the method further includes: establishing a three-dimensional geometric model of the subway station, where the three-dimensional geometric model includes the main structure of the station and the layout of electromechanical equipment; and marking sensor acquisition points in the three-dimensional geometric model to obtain the preset digital twin model, where the sensor acquisition points include the positions of structural vibration sensors, temperature and humidity sensors, passenger flow counters, and equipment status monitoring points.

[0009] By adopting the above technical solution, before establishing the subway station digital twin model, the server constructs a three-dimensional geometric model of the subway station and marks the specific acquisition positions of various sensors in the three-dimensional geometric model, making subsequent data acquisition more targeted and the sensor layout more reasonable, so as to accurately reflect the actual operation status of the subway station.

[0010] In combination with some embodiments of the first aspect, in some embodiments, mapping the subway station data to a preset digital twin model to obtain a subway station digital twin model for visually displaying the real-time operation state of the subway station, which specifically includes: mapping the structural vibration data to the corresponding positions of structural vibration sensors through a preset vibration response model to obtain structural deformation data; mapping the temperature and humidity data to the in-station space grid points through a preset thermal diffusion model to obtain in-station temperature and humidity data; mapping the passenger flow density data to the in-station passageways and platform areas through a preset crowd flow model to obtain crowd density distribution data; mapping the equipment operation state data to the corresponding positions of equipment state monitoring points through a preset equipment performance model to obtain equipment working condition data; dynamically updating and rendering the preset digital twin model based on the structural deformation data, the in-station temperature and humidity data, the crowd density distribution data, and the equipment working condition data to obtain the subway station digital twin model.

[0011] By adopting the above technical solution, the server introduces a preset vibration response model, a preset thermal diffusion model, a preset crowd flow model, and a preset equipment performance model, thereby accurately mapping the original data into the subway station digital twin model, realizing the spatial display of data, truly reflecting the real-time changes in the operation state of the subway station, and effectively improving the intelligent level of subway station monitoring and management.

[0012] In combination with some embodiments of the first aspect, in some embodiments, normalizing the subway station data according to the health score standard system to obtain standardized evaluation data, which specifically includes: converting the structural vibration data into a structural state score according to the structural integrity standard, where the structural integrity standard includes a vibration amplitude threshold and a vibration frequency range; converting the equipment operation state data into an equipment efficiency score according to the electromechanical equipment standard, where the electromechanical equipment standard includes performance parameter requirements and operation state indicators; converting the temperature and humidity data into an environmental state score according to the environmental suitability standard, where the environmental suitability standard includes the in-station temperature range and the in-station humidity range; converting the passenger flow density data into a crowd distribution score according to the crowd density standard, where the crowd density standard includes the maximum allowable personnel density in each preset area in the station.

[0013] By adopting the above technical solutions, the server can evaluate the structural vibration data according to the structural integrity standard, timely detect potential structural safety hazards, evaluate the equipment operation status data according to the electromechanical equipment standard, ensure the efficient and reliable operation of the equipment, evaluate the temperature and humidity data according to the environmental suitability standard, ensure the comfort of the station environment, and evaluate the passenger flow density data according to the crowd density standard, effectively preventing congestion risks. This multi-dimensional health score standard system enables different types of monitoring data to be uniformly quantified and compared, providing a reliable data basis for subsequent comprehensive health assessment.

[0014] Combined with some embodiments of the first aspect, in some embodiments, before the step of inputting the standardized evaluation data into a pre-trained health assessment model to obtain the health index of the subway station, the method further includes: obtaining historical structural state scores, historical equipment efficiency scores, historical environmental state scores, historical population distribution scores, and historical health indices; using the historical structural state scores, the historical equipment efficiency scores, the historical environmental state scores, and the historical population distribution scores as inputs and the historical health index as the output to train a preset model; when the accuracy of the preset model exceeds a preset accuracy threshold, obtaining the health assessment model.

[0015] By adopting the above technical solutions, the server trains the model based on historical data, enabling the health assessment model to accurately reflect the correlation between the standardized evaluation data and the health index, improving the accuracy and reliability of the health assessment.

[0016] Combined with some embodiments of the first aspect, in some embodiments, determining the potential problem points of the subway station specifically includes: according to the change trends of the structural state score, the equipment efficiency score, the environmental state score, and the population distribution score in a preset time window, determining the score anomaly items with a continuously deteriorating change trend; if the score anomaly item is the structural state score, using a structural modal analysis model to locate the structural damage points; if the score anomaly item is the equipment efficiency score, using an equipment fault tree analysis model to locate the equipment fault points; if the score anomaly item is the environmental state score, using an environmental factor propagation model to locate the environmental anomaly points; if the score anomaly item is the population distribution score, using a crowd density clustering model to locate the crowded points.

[0017] By adopting the above technical solutions, the server uses a structural modal analysis model to locate the structural damage points, an equipment fault tree analysis model to locate the equipment fault points, an environmental factor propagation model to locate the environmental anomaly points, and a crowd density clustering model to locate the crowded points. This precise potential problem location method can quickly lock in the source of the problem, providing an important basis for timely taking targeted disposal measures.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of displaying the outline of the potential hazard area in the digital twin model of the subway station based on the potential hazard point, the method further includes: calculating the hazard degree index and the diffusion influence range of each potential hazard point; performing a weighted sum of the hazard degree index and the diffusion influence range to determine the risk level of the potential hazard point; sorting the potential hazard points according to the risk level to generate a list of potential hazard points.

[0019] By adopting the above technical solution, after identifying the potential hazard points, the server further calculates the hazard degree and the influence range of each potential hazard point, determines the risk level through weighted summation, and performs priority sorting and list management accordingly. This method of potential hazard management based on risk assessment can help management personnel reasonably allocate maintenance resources, prioritize the handling of high-risk potential hazards, and improve the efficiency and pertinence of maintenance work.

[0020] In a second aspect, an embodiment of the present application provides a server, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the server to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product containing instructions, when the above computer program product runs on the server, causing the above server to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, including instructions, when the above instructions run on the server, causing the above server to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] It can be understood that the server provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, and will not be elaborated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By adopting the above technical solution, the server maps the structure vibration data, temperature and humidity data, passenger flow density data, and equipment operation status data to a preset digital twin model to obtain a digital twin model of the subway station. The server can combine the health score standard system and the health assessment model to comprehensively monitor and evaluate the real-time operation status of the subway station, and visually display various potential hazard points in the digital twin model of the subway station, which helps to timely discover and handle potential risks and improve the safety and reliability of the subway station operation.

[0025] 2. By adopting the above technical solution, the server introduces a preset vibration response model, a preset heat diffusion model, a preset crowd flow model, and a preset equipment performance model, thereby accurately mapping the original data into the digital twin model of the subway station, realizing the spatial display of data, truly reflecting the real-time changes in the operation status of the subway station, and effectively improving the intelligent level of subway station monitoring and management.

[0026] 3. By adopting the above technical solution, the server can evaluate the structure vibration data according to the structural integrity standard, timely discover potential structural safety hazards, evaluate the equipment operation status data according to the electromechanical equipment standard to ensure the efficient and reliable operation of the equipment, evaluate the temperature and humidity data according to the environmental suitability standard to ensure the comfort of the station environment, and evaluate the passenger flow density data according to the crowd density standard to effectively prevent congestion risks. This multi-dimensional health score standard system enables different types of monitoring data to be uniformly quantified and compared, providing a reliable data basis for subsequent comprehensive health assessments. Brief Description of the Drawings

[0027] Figure 1 is a flowchart of a method for evaluating the health of a subway station based on digital twin in an embodiment of the present application; Figure 2 is another flowchart of a method for evaluating the health of a subway station based on digital twin in an embodiment of the present application; Figure 3 is a schematic structural diagram of an entity device of the server in an embodiment of the present application. Detailed Embodiments

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application, the singular forms "a", "an", "the above", "the", and "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term " / and" used in the present application refers to any or all possible combinations including one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0030] For ease of understanding, the application scenarios of the embodiments of the present application are introduced below.

[0031] The daily average passenger flow of the subway network in a certain megacity exceeds 10 million person-times, and the daily average passenger flow at the hub stations reaches 500,000 person-times. With the increase in the number of operating years, subway stations are facing multiple challenges: First, the impact of large passenger flows causes the station structure to bear continuous fatigue loads, and minor cracks appear at some stations; second, the electromechanical equipment operates at a high load for a long time, and equipment failures occur frequently. For example, in the first half of 2022, the ventilation system and escalator system of a certain hub station had a total of 37 failures; third, after the epidemic, the requirements for the in-station environmental quality have increased, and environmental parameters such as temperature, humidity, and ventilation need to be more strictly controlled; fourth, under the condition of large passenger flows, it is necessary to accurately control the population distribution to avoid safety accidents such as stampedes. In March 2023, at a transfer station, due to the failure to detect structural damage in time, local ground settlement occurred on the platform, resulting in water accumulation in the track area and affecting train operation for 4 hours. In August of the same year, at another subway station, due to the failure to divert large passenger flows in time, the concourse was crowded with people, causing panic among passengers and resulting in 3 people being slightly injured. These incidents have exposed that the safety management of subway stations faces major challenges, and there is an urgent need to establish an all-round and intelligent health monitoring and evaluation system.

[0032] In related technologies, traditional monitoring and management methods can be adopted. For example, a certain subway operation company has deployed multiple independent monitoring systems in subway stations, including 32 structural vibration sensors, 24 temperature and humidity sensors, 16 passenger flow counters, and 45 equipment status monitoring devices. During the morning and evening rush hours every day, two staff members need to conduct inspections in the station and view the data of each monitoring point through a portable terminal. On a certain morning rush hour day in January 2024, the monitoring system showed that the temperature in the platform area reached 29°C, exceeding the comfort range, but the staff needed to check one by one to find that the No. 3 ventilation unit was operating abnormally. At the same time, there was a large passenger flow in the concourse, but due to the lack of a unified data analysis platform, the staff could not quickly judge the cause of the congestion. At noon, the monitoring system issued an early warning of abnormal vibration, but due to the scattered data and the inability to comprehensively analyze, the damage location could not be located in time. This scattered monitoring method makes it difficult for data to be interconnected and unable to realize the fusion analysis of multi-source information, resulting in low management efficiency and difficulty in timely discovering and handling potential risks.

[0033] By adopting the digital twin-based subway station health assessment method in the embodiments of the present application, the server maps the structure vibration data, temperature and humidity data, passenger flow density data, and equipment operation status data to a preset digital twin model to obtain a subway station digital twin model. The server combines the health score standard system and the health assessment model to comprehensively monitor and evaluate the real-time operation status of the subway station, and visually display various potential hazard points in the subway station digital twin model, which helps to timely discover and handle potential risks and improve the safety and reliability of subway station operation.

[0034] The following describes the process of the method provided in this embodiment. Please refer to Figure 1 , which is a schematic flow diagram of the digital twin-based subway station health assessment method in the embodiments of the present application.

[0035] S101. Obtain subway station data and a health score standard system. The subway station data includes structure vibration data, temperature and humidity data, passenger flow density data, and equipment operation status data. The health score standard system includes a structure integrity standard, an electromechanical equipment standard, an environmental suitability standard, and a population density standard. Among them, the subway station data refers to the set of real-time monitoring data collected by the server from various sensors in the subway station, including structure vibration data reflecting the structural safety status of the subway station, temperature and humidity data reflecting the environmental quality in the subway station, passenger flow density data reflecting the passenger flow situation in the subway station, and equipment operation status data reflecting the operation status of the equipment in the subway station. The health score standard system refers to the set of standard specifications for evaluating various performance indicators of the subway station, including a structure integrity standard for evaluating structural safety, an electromechanical equipment standard for evaluating equipment performance, an environmental suitability standard for evaluating environmental comfort, and a population density standard for evaluating the distribution of people.

[0036] Specifically, the server obtains real-time monitoring data (i.e., subway station data) from sensor nodes distributed throughout the subway station through a data acquisition interface, including obtaining structure vibration data from an acceleration sensor, temperature and humidity data from a temperature and humidity sensor, passenger flow density data from a passenger flow counter, equipment operation status data from an equipment status monitoring device, etc. At the same time, the server retrieves the pre-configured health score standard system from the database. These health score standard systems are formulated according to industry specifications and actual operation experience, and include specific parameters such as the threshold range and scoring rules of each evaluation index.

[0037] S102. Map the subway station data to a preset digital twin model to obtain a subway station digital twin model, which is used to visually display the real-time operation status of the subway station. Among them, the preset digital twin model refers to a pre-established three-dimensional virtual model framework of a subway station, which is used to carry the visual display of real-time data; the digital twin model of the subway station refers to a dynamic virtual model integrating subway station data, which is used to intuitively display the current operating state of the subway station; mapping refers to the process of establishing an association between real-time data and corresponding positions in the virtual model; visual display refers to presenting data information in a graphical way.

[0038] Specifically, first, the server preprocesses the obtained subway station data through corresponding data processing models, such as filtering and denoising the structural vibration data and interpolating the temperature and humidity data. Then, the server maps the processed subway station data to the corresponding spatial positions and components in the preset digital twin model according to the preset mapping rules. For example, mapping the structural vibration data to the structural components, the temperature and humidity data to the spatial grid points, the passenger flow density data to the channels and platform areas, and the equipment operation status data to the equipment entities. Finally, the server performs real-time rendering and updating on the digital twin model of the subway station, and displays the real-time changes of various data through different visualization methods (such as colors, animations, etc.).

[0039] S103. According to this health score standard system, normalize the subway station data to obtain standardized evaluation data, which includes structural state score, equipment efficiency score, environmental state score, and population distribution score; Among them, normalization processing refers to the process of converting data with different dimensions into a unified interval; standardized evaluation data refers to the unified scoring results after normalization processing, including structural state score, equipment efficiency score, environmental state score, and population distribution score. The structural state score is a scoring index reflecting the structural safety degree; the equipment efficiency score is a scoring index reflecting the equipment performance level; the environmental state score is a scoring index indicating the environmental comfort level; the population distribution score is a scoring index reflecting the rationality of the passenger flow distribution.

[0040] Specifically, first, the server processes different types of monitoring data according to the specific requirements in the health score standard system. Then, the server takes the structural integrity standard as a reference and obtains the structural state score based on the structural vibration data, takes the electromechanical equipment standard as a reference and obtains the equipment efficiency score based on the equipment operation status data, takes the environmental suitability standard as a reference and obtains the environmental state score based on the temperature and humidity data, and takes the population density standard as a reference and obtains the population density score based on the passenger flow density data. All scores are converted into a unified interval of 0-100 through normalization processing for subsequent comprehensive evaluation.

[0041] Optionally, generally, according to this health score standard system, the normalization process of the subway station data to obtain the standardized evaluation data can be achieved in the following ways, which are not limited here: According to this structural integrity standard, convert the structural vibration data into the structural state score, and this structural integrity standard includes the vibration amplitude threshold and the vibration frequency range; According to this electromechanical equipment standard, convert the equipment operation state data into the equipment efficiency score, and this electromechanical equipment standard includes performance parameter requirements and operation state indicators; According to this environmental suitability standard, convert the temperature and humidity data into the environmental state score, and this environmental suitability standard includes the in-station temperature range and the in-station humidity range; According to this population density standard, convert the passenger flow density data into the population distribution score, and this population density standard includes the maximum allowable personnel density in each preset area within the station.

[0042] (a) Example of calculating the structural state score: Structural integrity standard: Vibration amplitude threshold: 0.5 mm (warning value), 1.0 mm (hazard value); Vibration frequency range: 2 - 15 Hz; Example of measured data: Vibration amplitude of measuring point A: 0.3 mm, vibration frequency: 8 Hz; Vibration amplitude of measuring point B: 0.6 mm, vibration frequency: 12 Hz; Vibration amplitude of measuring point C: 0.4 mm, vibration frequency: 17 Hz; Calculation of structural state score: Measuring point A: Amplitude score = 100 * (1 - 0.3 / 0.5) = 40 points, frequency score = 100 points, comprehensive score = 95 points; Measuring point B: Amplitude score = 100 * (1 - 0.6 / 0.5) = 0 points, frequency score = 100 points, comprehensive score = 70 points; Measuring point C: Amplitude score = 100 * (1 - 0.4 / 0.5) = 20 points, frequency score = 60 points, comprehensive score = 75 points; Final structural state score = Min(measuring point scores) = 70 points; (b) Example of calculating the equipment efficiency score: Electromechanical equipment standard (taking the escalator as an example): Motor temperature threshold: 65 °C (warning value), 80 °C (hazard value); Bearing vibration threshold: 4.5 mm / s (warning value), 7.0 mm / s (hazard value); Operating current deviation: ±15% (warning value), ±25% (hazard value); Example of measured data: Escalator No. 1: Motor temperature: 58°C; Bearing vibration: 3.8 mm / s; Operating current deviation: +10%; Calculation of equipment efficiency score: Temperature score = 100 * (1 - 58 / 65) = 89 points; Vibration score = 100 * (1 - 3.8 / 4.5) = 84 points; Current score = 100 * (1 - 10 / 15) = 93 points; Final equipment efficiency score = 0.4 * temperature score + 0.4 * vibration score + 0.2 * current score = 88 points; (c) Example of environmental status score calculation: Environmental suitability criteria: Temperature range: 18 - 26°C (optimal), 16 - 28°C (acceptable); Relative humidity range: 40 - 65% (optimal), 30 - 75% (acceptable); Example of measured data: Concourse area: Temperature: 24.5°C; Relative humidity: 58%; Calculation of environmental status score: Temperature score = (26 - 24.5) / (26 - 18) * 100 = 95 points; Humidity score = (65 - 58) / (65 - 40) * 100 = 92 points; Final environmental status score = 0.6 * temperature score + 0.4 * humidity score = 93.8 points; (d) Example of crowd distribution score calculation: Crowd density criteria: Concourse: 4 persons / m² (warning value), 6 persons / m² (hazard value); Platform: 5 persons / m² (warning value), 7 persons / m² (hazard value); Corridor: 3 persons / m² (warning value), 5 persons / m² (hazard value); Example of measured data: Concourse area: 3.2 persons / m²; Platform area: 4.8 persons / m²; Transfer corridor: 2.5 persons / m²; Calculation of crowd distribution score: Concourse score = 100 * (1 - 3.2 / 4) = 80 points; Platform score = 100 * (1 - 4.8 / 5) = 76 points; Channel score = 100 * (1 - 2.5 / 3) = 83 points; Final score of population distribution = Min (scores of each area) = 76 points.

[0043] S104. Input the standardized evaluation data into a pre-trained health assessment model to obtain the health index of the subway station; Among them, the health assessment model refers to a mathematical model for comprehensively evaluating the health status of subway stations trained by machine learning methods; the standardized evaluation data refers to the input features of the health assessment model, including various normalized scoring indicators; the health index refers to a comprehensive indicator reflecting the overall operation status of the subway station, with a value range of 0 - 100; pre-trained means that the health assessment model has completed the training and verification process through historical data and has good assessment capabilities.

[0044] Specifically, first, the server preprocesses and organizes the features of the standardized evaluation data to ensure that the data format meets the model input requirements. Then, the server inputs the processed standardized evaluation data into the pre-trained health assessment model. The health assessment model adopts a deep neural network architecture and can automatically learn the complex correlation relationships between various scoring indicators. The health assessment model calculates a health index between 0 - 100 through forward calculation based on the input standardized evaluation data. This health index comprehensively reflects the overall operation status of the subway station in terms of structure, equipment, environment, and passenger flow.

[0045] Optionally, generally, inputting the standardized evaluation data into the pre-trained health assessment model to obtain the health index of the subway station can also be achieved through the following method: Following step S103, weights are configured for the structure status score, equipment efficiency score, environment status score, and population distribution score as follows: Weight of structure status score: 0.35; Weight of equipment efficiency score: 0.25; Weight of environment status score: 0.20; Weight of population distribution score: 0.20; Final health index = 70 * 0.35 + 88 * 0.25 + 93.8 * 0.20 + 76 * 0.20 = 80.26 points.

[0046] S105. When the health index is lower than the preset health threshold, determine the potential hazard points of the subway station. The potential hazard points include structural damage points, equipment failure points, environmental anomaly points, and / or crowded points; Among them, the preset health threshold refers to the pre-set health warning value used to determine whether the subway station is in an abnormal state; the potential hazard point refers to the specific location where there may be safety risks; the structural damage point refers to the location where the structural component is abnormal; the equipment failure point is the location of the equipment with abnormal operation; the environmental anomaly point refers to the location where the environmental parameters exceed the standard; and the crowded point refers to the location where the personnel concentration is too high.

[0047] Specifically, the server compares the calculated health index with the preset health threshold (usually 75 points). When the health index is lower than the preset health threshold, the server starts the potential hazard point analysis program, and conducts abnormal analysis on the structural state score, equipment efficiency score, environmental state score, and population distribution score respectively. Through the trend analysis of each score item in the time dimension, the server identifies the abnormal items with continuously decreasing scores. Then, for different types of abnormal items, the server calls the corresponding professional models respectively for the location analysis of the specific location, and finally determines the specific spatial coordinates of various potential hazard points.

[0048] Optionally, generally, the potential hazard points of the subway station can be determined through the following methods, which are not limited here: according to the change trends of the structural state score, equipment efficiency score, environmental state score, and population distribution score in the preset time window, determine the score abnormal items with continuously deteriorating trends; if the score abnormal item is the structural state score, use the structural modal analysis model to locate the structural damage point; if the score abnormal item is the equipment efficiency score, use the equipment fault tree analysis model to locate the equipment failure point; if the score abnormal item is the environmental state score, use the environmental factor propagation model to locate the environmental anomaly point; if the score abnormal item is the population distribution score, use the population density clustering model to locate the crowded point.

[0049] Example of score trend analysis (the preset time window is the last 6 hours, and the sampling interval is 30 minutes): Trend of structural state score: 92 → 90 → 89 → 87 → 85 → 82 → 80 (showing a downward trend); Trend of equipment efficiency score: 88 → 87 → 88 → 86 → 87 → 88 → 87 (fluctuating and stable); Trend of environmental state score: 95 → 94 → 92 → 88 → 83 → 77 → 72 (sharply decreasing); Trend of population distribution score: 85 → 83 → 82 → 84 → 83 → 85 → 84 (basically stable); Analysis result: The structural state score and the environmental state score show a continuous deteriorating trend, and specific potential hazard points need to be further located.

[0050] (a) Example of location by structural modal analysis model: Collect vibration response data: Measuring point 1: Frequency 3.8 Hz, amplitude 0.35 mm; Measuring point 2: Frequency 3.6 Hz, amplitude 0.42 mm; Measuring point 3: Frequency 3.2 Hz, amplitude 0.58 mm; Calculate modal parameters: First-order natural frequency: 3.8 Hz; Second-order natural frequency: 7.2 Hz; Modal damping ratio: 2.3%; Damage location: By analyzing the change of vibration mode through the modal curvature method, it is found that the stiffness near measuring point 3 is reduced by about 15%, and the coordinates of the structural damage point are determined as (X: 25.6 m, Y: 12.3 m, Z: -8.2 m); (b)Example of equipment fault tree analysis model (taking the air conditioning system as an example): Top event: Poor refrigeration effect First-level events: Compressor failure (probability 0.35); Condenser failure (probability 0.25); Refrigerant leakage (probability 0.30); Control system failure (probability 0.10); Second-level events (compressor branch): Bearing wear (probability 0.45); Motor overheating (probability 0.35); Valve damage (probability 0.20); Fault location result: The probability of compressor bearing wear is the highest, located in the equipment room, with coordinates (X: 32.5 m, Y: 15.8 m, Z: -3.2 m); (c)Example of environmental factor propagation model: Analysis of temperature anomaly propagation: Initial conditions: Measuring point A (0, 0): 29 °C; Peripheral measuring points: 24 - 26 °C; Air velocity: 0.3 m / s; Diffusion coefficient: 0.025 m² / s; Propagation equation: ∂T / ∂t = α(∂²T / ∂x² + ∂²T / ∂y²) + v∇T; Calculation results: Coordinates of the anomaly source (3.2, -2.5); Influence range radius 4.8 m; Diffusion direction: 45° southeast; (d)Example of crowd density clustering model: DBSCAN clustering analysis: Input parameters: Minimum density threshold: 50 people; Clustering radius (Eps): 3 m; Sampling time: 5 minutes; Density data: Area 1: 3.8 people / m²; Area 2: 4.2 people / m²; Area 3: 2.5 people / m²; Clustering results: Dense point 1: Middle of the platform (12.5, 8.3), covering an area of 42 m²; Dense point 2: Escalator entrance of the concourse (28.6, -5.2), covering an area of 35 m².

[0051] S106. Based on this potential hazard point, the outline of the potential hazard area is displayed in the digital twin model of this subway station.

[0052] Among them, the outline of the potential hazard area refers to the scope of the risk area marked in the digital twin model of the subway station; display means marking the potential hazard points through specific visualization methods (such as highlighting, flashing, etc.).

[0053] Specifically, first, the server calculates the corresponding outline of the potential hazard area in the digital twin model of the subway station according to the type and influence range of the potential hazard points. For structural damage points, calculate the complete outline of the damaged component; for equipment failure points, mark the installation location and maintenance operation space of the faulty equipment; for environmental anomaly points, determine the influence range according to the diffusion law of environmental parameters; for crowded points, mark the boundary of the personnel gathering area. Then, the server visually displays these potential hazard points in the digital twin model of the subway station through different visual effects (such as red translucent areas, flashing borders, etc.), and supports interactive viewing of the detailed information of each potential hazard point.

[0054] The following is a further and more specific process description of the method provided in this embodiment. Please refer to Figure 2 , which is another process schematic diagram of the subway station health assessment method based on digital twin in the embodiments of the present application.

[0055] S201. Obtain subway station data and a health assessment standard system. The subway station data includes structural vibration data, temperature and humidity data, passenger flow density data, and equipment operation status data. The health assessment standard system includes structural integrity standards, electromechanical equipment standards, environmental suitability standards, and crowd density standards; Specifically, refer to step S101, which will not be elaborated here.

[0056] S202. Establish a three-dimensional geometric model of the subway station. The three-dimensional geometric model includes the main structure of the station and the layout of electromechanical equipment; Among them, the three-dimensional geometric model refers to a virtual model of the subway station constructed based on a three-dimensional coordinate system; the main structure of the station refers to the load-bearing component system including diaphragm walls, middle plates, bottom plates, etc.; the layout of electromechanical equipment refers to the spatial distribution positions of various equipment in the station.

[0057] Specifically, first, the server imports the CAD design drawings of the subway station, including structural construction drawings, electromechanical equipment installation drawings, etc. Then, the server uses a modeling engine to build a three-dimensional geometric model of the subway station: (1) Build main structure components such as diaphragm walls, middle plates, and bottom plates, and accurately restore their geometric dimensions and spatial position relationships; (2) Add partitions for functional spaces such as concourses, platforms, and channels; (3) Arrange electromechanical equipment such as ventilation and air conditioning, water supply and drainage, power supply and lighting, and escalators into the model according to the actual installation positions. The entire modeling process needs to ensure that the geometric accuracy and spatial topological relationship of the model are consistent with the actual subway station.

[0058] S203. Mark the sensor acquisition points in the three-dimensional geometric model to obtain the preset digital twin model. The sensor acquisition points include the positions of structure vibration sensors, temperature and humidity sensors, passenger flow counters, and equipment status monitoring points. Among them, the sensor acquisition point refers to the actual installation position coordinates of various sensors; marking means marking the spatial position information of the sensors in the three-dimensional geometric model; the position of the structure vibration sensor refers to the installation point of the sensor for measuring structure vibration; the position of the temperature and humidity sensor refers to the layout point of the sensor for monitoring environmental parameters; the position of the passenger flow counter refers to the installation point of the device for counting the number of people; the position of the equipment status monitoring point refers to the monitoring point for collecting the operating parameters of the equipment; the preset digital twin model is used to represent the basic model architecture after the sensor layout is completed.

[0059] Specifically, the server marks the sensor acquisition points in the three-dimensional geometric model according to the sensor layout plan: First, mark the installation coordinates of the structure vibration sensors at the key positions of the load-bearing members (such as wall joints, mid-span positions, etc.); then mark the layout positions of the temperature and humidity sensors at a certain interval in the areas where people gather, such as the concourse and platform; then mark the installation points of the passenger flow counters at the passenger flow channels such as the entrance and exit channels and transfer channels; finally, mark the positions of the status monitoring sensors at various mechanical and electrical equipment. All sensor acquisition points need to record accurate three-dimensional coordinates and orientation information, and at the same time establish the corresponding relationship between the sensor numbers and positions to form a complete sensor layout file.

[0060] S204. Map the structure vibration data to the corresponding positions of the structure vibration sensors through a preset vibration response model to obtain structure deformation data. Among them, the preset vibration response model is a mathematical model established based on the principles of structural dynamics for analyzing the law of structure vibration propagation; mapping refers to the process of converting the measured data into target parameters through model calculation; the structure deformation data is a set of physical quantities reflecting the deformation states of structural members such as displacement and strain.

[0061] Specifically, first, the server preprocesses the collected vibration acceleration signals, including filtering and noise reduction, baseline correction, etc. Then, the server inputs the processed vibration acceleration signals into the preset vibration response model, which is constructed based on the finite element method and includes characteristic parameters such as the mass, stiffness, and damping of the structure. Through the dynamic response analysis of the preset vibration response model, the propagation effect of vibration in the structure is calculated, and the structure deformation parameters such as displacement, velocity, and strain at each monitoring point are obtained.

[0062] S205. Map the temperature and humidity data to the in-station space grid points through a preset thermal diffusion model to obtain the in-station temperature and humidity data. Among them, the preset thermal diffusion model refers to a physical model that describes the propagation law of temperature and humidity in a closed space; the spatial grid points refer to the regular discrete dot matrix formed by dividing the in-station space at a certain interval; the in-station temperature and humidity data is a data set that reflects the temperature and humidity distribution in the entire in-station space.

[0063] Specifically, first, the server divides the in-station space into several regular grid units, and the grid spacing is usually 0.5 - 1 meter. Then, the server inputs the temperature and humidity data of each measurement point into the preset thermal diffusion model. This preset thermal diffusion model is based on the computational fluid dynamics theory and takes into account physical processes such as air flow and heat conduction. Through the calculation of the preset thermal diffusion model, the interpolation mapping of the temperature and humidity data from discrete measurement points to spatial grid points is realized, and the continuous temperature and humidity field distribution covering the entire in-station space is obtained.

[0064] S206. Map the passenger flow density data to the in-station passage and platform areas through the preset crowd flow model to obtain the crowd density distribution data; Among them, the preset crowd flow model refers to a mathematical model that describes the movement characteristics and behavior laws of the crowd; the in-station passage refers to the passage space for passengers to pass through; the platform area refers to the space area for passengers to wait for the train; the crowd density distribution data refers to the spatial distribution data that reflects the number of people per unit area.

[0065] Specifically, the server first organizes the real-time obtained passenger flow density data according to the time and space dimensions. Then, the server inputs the passenger flow density data into the preset crowd flow model. This preset crowd flow model is usually constructed based on the cellular automaton or social force model and can simulate the movement law of the crowd. Through the calculation of the preset crowd flow model, the discrete passenger flow volume data is converted into a continuous crowd density distribution and standardized according to the area of different regions.

[0066] S207. Map the equipment operation status data to the corresponding equipment status monitoring point positions through the preset equipment performance model to obtain the equipment working condition data; Among them, the preset equipment performance model refers to a mathematical model that describes the operation characteristics and performance indicators of the equipment; the equipment status monitoring position refers to the spatial position for collecting equipment operation data; the equipment working condition data is the comprehensive index data that reflects the health status of the equipment.

[0067] Specifically, first, the server preprocesses and performs unit conversion on the equipment operation status data (such as rotational speed, temperature, vibration, current, etc.). Then, the server inputs the processed equipment operation status data into the preset equipment performance model. This preset equipment performance model is established based on the equipment characteristic curve and professional experience and can evaluate the operation status of the equipment. Through the calculation of the preset equipment performance model, the working condition indicators reflecting the health status of the equipment are obtained, such as the performance coefficient, failure probability, etc.

[0068] S208. Based on the structural deformation data, the in-station temperature and humidity data, the population density distribution data, and the equipment working condition data, dynamically update and render the preset digital twin model in real time to obtain the digital twin model of the subway station; Among them, dynamic update refers to the process of continuously adjusting the model state according to real-time data; real-time rendering refers to the graphic calculation process of converting model data into a visual effect.

[0069] Specifically, the server integrates the processed structural deformation data, in-station temperature and humidity field data, population density data, and equipment working condition data into the preset digital twin model. The structural deformation is displayed through the dynamic deformation effect of components; the temperature and humidity distribution is represented by an isosurface with color gradient; the population density is displayed in the form of a heat map; and the equipment working condition is identified by status icons. The server maintains an update frequency of at least 10 Hz to ensure the real-time nature of the model state. During the rendering process, a physically based rendering technology and a graphics acceleration algorithm are used to achieve a smooth interactive display effect.

[0070] S209. According to the health score standard system, normalize the subway station data to obtain standardized evaluation data, which includes structural state score, equipment efficiency score, environmental state score, and population distribution score; Specifically, refer to step S103, which will not be elaborated here.

[0071] S210. Obtain the historical structural state score, historical equipment efficiency score, historical environmental state score, historical population distribution score, and historical health index. Use the historical structural state score, historical equipment efficiency score, historical environmental state score, and historical population distribution score as inputs, and the historical health index as the output to train a preset model. When the accuracy of the preset model exceeds the preset accuracy threshold, obtain the health assessment model; The steps to construct a health assessment model based on deep learning are as follows: First, the server collects the historical structure status score, historical equipment efficiency score, historical environment status score, historical population distribution score, and historical health index from the storage system. The acquisition of the historical structure status score, historical equipment efficiency score, historical environment status score, and historical population distribution score can refer to step S103 and will not be elaborated here. The corresponding historical health index can be obtained through historical maintenance records or expert evaluation, which is not limited here. The server stores the historical structure status score, historical equipment efficiency score, historical environment status score, historical population distribution score, and historical health index in the dataset D in chronological order. Each data format is (historical structure status score, historical equipment efficiency score, historical environment status score, historical population distribution score, historical health index). Among them, the historical structure status score, historical equipment efficiency score, historical environment status score, and historical population distribution score are the input features for model training, and the historical health index is the output feature for model training.

[0072] Second, the server preprocesses the dataset D, deleting the missing data and abnormal data in the dataset D.

[0073] Then, the server constructs a recurrent neural network based on LSTM, including an input layer, two LSTM hidden layers, a fully connected layer, and an output layer. The input layer inputs the historical structure status score, historical equipment efficiency score, historical environment status score, and historical population distribution score. The number of hidden layer nodes is set to 64, the number of fully connected layer nodes is set to 32, and the output layer outputs the historical health index.

[0074] Next, the server uses the Adam optimizer with a learning rate set to 0.001 and a training batch size of 32, which can also be set according to the actual situation and is not limited here. 80% of the historical data is divided into the training set, and 20% is divided into the validation set. Train for 100 epochs and save the model with the highest accuracy on the validation set, which can also be set according to the actual situation and is not limited here. An epoch is the process of the entire training dataset passing through the neural network once. In machine learning and deep learning, an epoch is a unit used to measure the number of times the entire training set is repeatedly learned. Specifically, when the neural network completes a forward calculation and a backward propagation process, that is, all data has been processed by the network once, this completes one epoch. The server uses binary cross-entropy as the loss function and adopts Early Stopping to prevent overfitting. When the value of the loss function exceeds the preset function threshold, it is determined that the model training is completed, and a health assessment model is obtained. Early Stopping is a technique in deep learning and machine learning to prevent model overfitting. It decides when to stop training by monitoring the performance of the model on the validation set.

[0075] Finally, the server inputs the input features in the validation set into the health assessment model, and then obtains the predicted output of the health assessment model. The predicted output of the health assessment model is compared with the actual output features in the validation set, and some performance metrics such as accuracy, precision, recall, F1-score, mean squared error (MSE), etc. are used to evaluate the performance of the health assessment model. According to the performance of the health assessment model on the validation set, the parameters of the health assessment model are adjusted, including adjusting the learning rate, changing the model complexity (such as increasing or decreasing the number of layers or nodes in the neural network), modifying the regularization strength, etc. This process may require multiple iterations, each time adjusting based on the previous learning results to optimize the health assessment model.

[0076] S211. Input the standardized assessment data into the pre-trained health assessment model to obtain the health index of the subway station; Specifically, refer to step S104, which will not be elaborated here.

[0077] S212. When the health index is lower than the preset health threshold, determine the potential hazard points of the subway station, which include structural damage points, equipment failure points, environmental anomaly points, and / or crowded points; Specifically, refer to step S105, which will not be elaborated here.

[0078] S213. Based on the potential hazard points, display the outline of the potential hazard area in the digital twin model of the subway station; Specifically, refer to step S106, which will not be elaborated here.

[0079] S214. Calculate the hazard degree index and the spread influence range of each potential hazard point; Among them, the hazard degree index is a numerical index that quantitatively represents the degree of influence of the potential hazard point on the safe operation of the subway station; the spread influence range is the spatial area that the hazard effect of the potential hazard point may affect.

[0080] Specifically, the server uses different evaluation models for different types of potential hazard points to calculate their hazard degree indexes: for structural damage points, evaluate the degree of influence on structural safety based on the structural mechanics model; for equipment failure points, calculate the hazard index according to the equipment importance and the degree of failure impact; for environmental anomaly points, determine the hazard level based on the degree of environmental parameter exceeding the standard; for crowded points, evaluate the risk level based on the degree of crowd density exceeding the limit. At the same time, the server calculates the influence range of various potential hazards through corresponding propagation models: the stress propagation range of structural damage, the chain influence range of equipment failure, the diffusion range of environmental anomalies, the spread range of crowd congestion, etc.

[0081] S215. Perform weighted summation on the hazard degree index and the spread influence range to determine the risk level of the potential hazard point; Among them, weighted summation refers to a method of comprehensively calculating the hazard degree index and the diffusion influence range according to different weights; the risk level refers to the grading standard reflecting the severity of potential hazards, usually divided into three levels: low risk, medium risk, and high risk.

[0082] Specifically, first, the server determines the weight coefficients of the hazard degree index and the diffusion influence range based on expert experience and historical data. Usually, the weight of the hazard degree index is between 0.6 and 0.7, and the weight of the diffusion influence range is between 0.3 and 0.4. Then, the server standardizes the hazard degree index and the diffusion influence range and converts them to a unified scoring range (such as 0 - 100 points). Next, the server performs weighted calculation according to the set weights to obtain the comprehensive risk score. Finally, the server determines the risk level of this potential hazard point according to the preset grading standard (for example, below 70 points is low risk, 70 - 85 points is medium risk, and above 85 points is high risk).

[0083] S216. Sort the potential hazard points according to this risk level to generate a list of potential hazard points.

[0084] Among them, priority sorting refers to the process of sorting the urgency of handling potential hazard points according to the risk level; the list of potential hazard points refers to the list of all potential hazard points to be processed.

[0085] Specifically, first, the server sorts all potential hazard points in descending order according to the risk level, and the high - risk potential hazard points have the highest priority. For potential hazard points with the same risk level, further consider factors such as the importance of their locations (such as whether they are on key channels), the size of the influence range, and the difficulty of repair for refined sorting. Then, the server generates a standardized list of potential hazard points. The list of potential hazard points contains information such as the specific location, type, hazard degree index, influence range, risk level, and handling priority of each potential hazard point, and at the same time marks the recommended handling time limit: high - risk potential hazards are recommended to be handled within 24 hours, medium - risk potential hazards are recommended to be handled within 72 hours, and low - risk potential hazards are recommended to be handled within one week.

[0086] The server in the embodiment of the present invention application will be described from the perspective of hardware processing. Please refer to Figure 3 , which is a schematic structural diagram of an entity device of the server in the embodiment of the present application.

[0087] It should be noted that Figure 3 The structure of the server shown is only an example and should not bring any restrictions to the functions and usage scope of the embodiments of the present invention.

[0088] Such as Figure 3As shown, the server includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a Read-Only Memory (ROM) 302 or a program loaded from a storage section 308 into a Random Access Memory (RAM) 303, such as executing the method described in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0089] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a Liquid Crystal Display (LCD), an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read from it can be installed into the storage section 308 as needed.

[0090] Specifically, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the Central Processing Unit (CPU) 301, various functions defined in the present invention are executed.

[0091] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program, and the program can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0092] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings.

[0093] Specifically, the server in this embodiment includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, the method for evaluating the health of a subway station based on digital twins provided in the above embodiment is implemented.

[0094] On the other hand, the present invention also provides a computer-readable storage medium, which may be included in the server described in the above embodiment; or it may exist separately and not be assembled into the server. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the server, the server is enabled to implement the method for evaluating the health of a subway station based on digital twins provided in the above embodiment.

[0095] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application.

[0096] As used in the foregoing embodiments, depending on the context, the term "when" may be construed to mean "if", "after", "in response to determining", or "in response to detecting". Similarly, depending on the context, the phrase "when determining" or "if (the stated condition or event) is detected" may be construed to mean "if determined", "in response to determining", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0097] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented by a computer program instructing relevant hardware. The program can be stored in a computer-readable storage medium. When the program is executed, it may include the processes of the foregoing method embodiments. The foregoing storage media include various media that can store program codes, such as ROM, random access memory (RAM), magnetic disks, or optical discs.

Claims

1. A method for evaluating the health of a subway station based on digital twin, characterized in that, Applied to a server, the method includes: Obtain subway station data and a health score standard system. The subway station data includes structural vibration data, temperature and humidity data, passenger flow density data, and equipment operation status data. The health score standard system includes a structural integrity standard, an electromechanical equipment standard, an environmental suitability standard, and a population density standard; Map the subway station data to a preset digital twin model to obtain a subway station digital twin model, which is used to visually display the real-time operation status of the subway station; According to the health score standard system, normalize the subway station data to obtain standardized evaluation data, which includes a structural status score, an equipment efficiency score, an environmental status score, and a population distribution score; Input the standardized evaluation data into a pre-trained health evaluation model to obtain a health index of the subway station; When the health index is lower than a preset health threshold, determine the potential hazard points of the subway station, which include structural damage points, equipment failure points, environmental anomaly points, and / or crowded points; Based on the potential hazard points, display the outline of the potential hazard area in the subway station digital twin model.

2. The method according to claim 1, characterized in that Before the step of mapping the subway station data to a preset digital twin model to obtain a subway station digital twin model, which is used to visually display the real-time operation status of the subway station, the method further includes: Establish a three-dimensional geometric model of the subway station, which includes the main station structure and the layout of electromechanical equipment; Mark sensor acquisition points in the three-dimensional geometric model to obtain the preset digital twin model. The sensor acquisition points include the positions of structural vibration sensors, temperature and humidity sensors, passenger flow counters, and equipment status monitoring points.

3. The method according to claim 2, characterized in that, The step of mapping the subway station data to a preset digital twin model to obtain a subway station digital twin model, which is used to visually display the real-time operation status of the subway station, specifically includes: Map the structural vibration data to the corresponding structural vibration sensor positions through a preset vibration response model to obtain structural deformation data; Map the temperature and humidity data to the in-station space grid points through a preset thermal diffusion model to obtain in-station temperature and humidity data; Map the passenger flow density data to the in-station passageways and platform areas through a preset crowd flow model to obtain crowd density distribution data; Map the equipment operation status data to the corresponding equipment status monitoring points through a preset equipment performance model to obtain equipment operating conditions data; Based on the structural deformation data, the in-station temperature and humidity data, the crowd density distribution data, and the equipment operating conditions data, dynamically update and render the preset digital twin model to obtain the subway station digital twin model.

4. The method according to claim 1, characterized in that, The step of normalizing the subway station data according to the health score standard system to obtain standardized evaluation data specifically includes: Convert the structural vibration data into the structural state score according to the structural integrity criteria, where the structural integrity criteria include a vibration amplitude threshold and a vibration frequency range; Convert the equipment operation state data into the equipment effectiveness score according to the electromechanical equipment criteria, where the electromechanical equipment criteria include performance parameter requirements and operation state indicators; Convert the temperature and humidity data into the environmental state score according to the environmental suitability criteria, where the environmental suitability criteria include the in-station temperature range and the in-station humidity range; Convert the passenger flow density data into the population distribution score according to the population density criteria, where the population density criteria include the maximum personnel density allowed in each preset area within the station.

5. The method according to claim 1, wherein Before the step of inputting the standardized evaluation data into a pre-trained health assessment model to obtain the health index of the subway station, the method further includes: Obtain the historical structural state score, historical equipment effectiveness score, historical environmental state score, historical population distribution score, and historical health index; Use the historical structural state score, historical equipment effectiveness score, historical environmental state score, and historical population distribution score as inputs, and the historical health index as an output to train a preset model; When the accuracy of the preset model exceeds a preset accuracy threshold, obtain the health assessment model.

6. The method according to claim 1, characterized in that, The determination of the potential hazard points of the subway station specifically includes: Determine the scoring anomaly items with a continuously deteriorating change trend according to the change trends of the structural state score, equipment effectiveness score, environmental state score, and population distribution score within a preset time window; If the scoring anomaly item is the structural state score, use a structural modal analysis model to locate the structural damage points; If the scoring anomaly item is the equipment effectiveness score, use an equipment fault tree analysis model to locate the equipment fault points; If the scoring anomaly item is the environmental state score, use an environmental factor propagation model to locate the environmental anomaly points; If the scoring anomaly item is the population distribution score, use a population density clustering model to locate the crowded points of the population.

7. The method according to claim 1, characterized in that After the step of displaying the outline of the potential hazard area in the subway station digital twin model based on the potential hazard points, the method further includes: Calculate the hazard degree index and the diffusion influence range of each potential hazard point; Perform a weighted sum of the hazard degree index and the diffusion influence range to determine the risk level of the potential hazard point; Sort the potential hazard points according to the risk level to generate a list of potential hazard points.

8. A server, characterized in that, The server includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the server to execute the method according to any one of claims 1-7.

9. A computer-readable storage medium, comprising instructions, characterized in that, When the instructions run on the server, cause the server to execute the method according to any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on a server, the server is caused to execute the method according to any one of claims 1 to 7.

Images