Rail transit carbon emission management method and system

Abstract

The invention relates to a rail transit carbon emission management method and system, and the method comprises the steps: enabling a visual interface to display a current overall carbon emission value and the deviation amplitude between the current overall carbon emission value and a predicted value if the abnormal risk level of the current carbon emission is a low risk level, and enabling a trend arrow to indicate the trend of the carbon emission in a future preset time range; if the carbon emission abnormal risk grade is a medium risk grade, displaying a detailed prediction deviation curve of each component and a dynamic change process of deviation by a visual interface; and if the carbon emission abnormity risk grade is a high risk grade, displaying a detailed prediction deviation curve of each component and a dynamic change process of deviation by a visual interface, and meanwhile, giving an alarm about the high risk grade and an abnormity reason. The method has the advantages that accurate prediction, early warning and reason analysis of carbon emission are achieved, visualization is achieved according to risk levels, and efficient management is assisted.

Description

Technical Field

[0001] The present invention relates to the technical field of information management, and in particular to a rail transit carbon emission management method and system. Background Art

[0002] As environmental awareness grows and sustainable development becomes a global consensus, rail transit, as a core part of the urban transportation system, has a carbon emission problem that cannot be underestimated. Efficient carbon emission management not only helps reduce the negative effects of rail transit on the environment, but also helps the industry's green transformation and long-term development.

[0003] At present, rail transit carbon emission management mainly focuses on data-driven analysis methods. In the data processing stage, the focus is on data collection and preliminary processing. Multiple sensors are used to collect heterogeneous carbon emission-related data from multiple sources such as train operation and station equipment, and then simple data cleaning is performed.

[0004] Existing technologies have major deficiencies in the core carbon emission early warning analysis mechanism. Since it is impossible to accurately compare real-time data with predicted values, it is difficult to make an accurate judgment on the abnormal risk level of carbon emissions, and it is even more difficult to deeply identify the root cause of the abnormality. In addition, the visualization link fails to effectively link with the carbon emission early warning level, and lacks targeted design based on different early warning levels. This makes it difficult for management departments to intuitively and quickly grasp the true situation of carbon emissions when faced with massive carbon emission data, and it is impossible to plan ahead and formulate effective response strategies, which seriously hinders the advancement of rail transit carbon emission management work in a scientific and efficient direction. Summary of the invention

[0005] In order to achieve accurate prediction, early warning and cause analysis of carbon emissions, visualize them according to risk levels, and assist in efficient management, this application provides a rail transit carbon emissions management method and system.

[0006] In the first aspect, the present application provides a rail transit carbon emission management method, which adopts the following technical solution: A rail transit carbon emission management method, comprising: Acquire multi-source heterogeneous data related to carbon emissions and perform data preprocessing; Extract carbon emission time series data from the multi-source heterogeneous data related to carbon emission that have completed data preprocessing, use the preset seasonal decomposition method to remove the influence of cyclical factors, and decompose the time series into trend terms, seasonal terms, and residual terms; Extract preset deep features based on the decomposed trend term, seasonal term and residual term time series information; The extracted deep features are input into the pre-built ARIMA model to predict the numerical range of carbon emission trend items, seasonal items and residual items within a preset time range in the future as the ARIMA model prediction value; Compare the ARIMA model predictions with real-time carbon emission data and calculate the prediction deviations of each component, including carbon emission trend term, seasonal term and residual term; The prediction deviation of each component is input into the pre-built carbon emission early warning analysis model to output the abnormal risk level of carbon emission and the abnormal cause. The abnormal risk level of carbon emission includes low risk level, medium risk level and high risk level. If the risk level of this carbon emission anomaly is low, the visualization interface will display the current overall carbon emission value, the deviation from the predicted value, and the trend arrow indicating the trend of carbon emissions within the preset time range in the future; If the risk level of this carbon emission anomaly is medium, the visualization interface will display the detailed prediction deviation curve of each component and the dynamic change process of the deviation; If the abnormal carbon emission risk level is high, the visualization interface will display the detailed prediction deviation curve of each component and the dynamic change process of the deviation, and at the same time warn of the high risk level and the cause of the abnormality.

[0007] By adopting the above technical solutions, through multi-source data processing and analysis, carbon emission trends can be accurately predicted. By comparing the predicted value with the real-time data, the prediction deviation can be discovered in time. With the help of the early warning model, the abnormal risk level and cause can be clarified. Different risk levels correspond to different visual displays, which facilitates intuitive understanding of carbon emissions and helps relevant departments take measures in advance to achieve effective management of rail transit carbon emissions.

[0008] Optionally, if the risk level of this carbon emission abnormality is medium risk level or high risk level, it also includes: Obtaining environmental monitoring satellite data, wherein the environmental monitoring satellite data includes regional meteorological dynamics and geographic information; Extract feature vectors related to rail transit carbon emissions from environmental monitoring satellite data, use the feature vectors related to rail transit carbon emissions as input objects, input them into a pre-built comprehensive model that simulates carbon emission changes under different emergency scenarios, and output simulation results; Based on the output simulation results, a curve of rail transit carbon emissions changing over time within a preset time range in the future is drawn, and relevant information of the emergency scenario is marked.

[0009] By adopting the above technical solutions and obtaining environmental monitoring satellite data, we can fully understand the regional meteorological and geographical information. Extracting relevant feature vectors from them and inputting them into the comprehensive model can simulate the changes in carbon emissions under different emergency scenarios. Drawing change curves based on the simulation results and annotating information can intuitively present the future trend of carbon emissions, help managers clearly grasp the dynamics of carbon emissions under different emergency scenarios, and provide a strong basis for formulating scientific and effective response strategies.

[0010] Optionally, feature vectors related to rail transit carbon emissions are extracted from environmental monitoring satellite data, including: Obtain the abnormal risk level of carbon emissions this time; Determine the real-time satellite data feature fluctuation information based on the environmental monitoring satellite data analysis, input the carbon emission abnormal risk level and the real-time satellite data feature fluctuation information into the pre-trained feature selection standard analysis model, and output the feature selection standard; From the environmental monitoring satellite data, feature vectors related to rail transit carbon emissions are extracted according to the feature selection criteria.

[0011] By adopting the above technical solution, the risk level and satellite data fluctuation information are input into the model to derive the selection criteria, which can accurately adapt to the current abnormal carbon emission situation. The feature vector is extracted according to this standard to ensure that the extracted information is highly targeted and closely related to rail transit carbon emissions. It not only improves the efficiency of data screening, but also provides more valuable data support for subsequent simulations of carbon emission changes under different emergency scenarios, helping to make accurate decisions.

[0012] Optionally, the characteristic vector related to rail transit carbon emissions is used as an input object and input into a pre-built comprehensive model that simulates the changes in carbon emissions under different emergency scenarios, and the output simulation results include: Analyze whether to obtain the three-dimensional terrain model information scanned by the laser radar device deployed on the top of the train or at key points along the line, as well as the environmental parameters detected by the dust sensor network deployed along the rail transit line and stations; If not, the characteristic vector related to rail transit carbon emissions is used as the input object and input into the pre-built comprehensive model that simulates the changes in carbon emissions under different emergency scenarios, and the simulation results are output; If yes, extracting the laser radar feature vector from the acquired three-dimensional terrain model information, and calibrating and integrating the acquired environmental parameters and extracting the smart dust sensor network feature vector; The laser radar feature vector, the smart dust sensor network feature vector and the feature vector extracted based on the environmental monitoring satellite data are fused to form a fusion vector; Adjusting the architecture of the comprehensive model according to the feature vector included in the fusion vector and the preset comprehensive model architecture quantitative adjustment method; The fusion vector is input into the comprehensive model that simulates the changes in carbon emissions under different emergency scenarios after the structural adjustment is completed, and the simulation results are output.

[0013] By adopting the above technical solutions, the accuracy of simulation results can be significantly improved through the judgment and use of additional equipment data. When there is lidar and dust sensor data, the corresponding feature vectors are extracted and fused, and the comprehensive model architecture is adjusted to allow the model to better adapt to complex data. This enables the model to fully reflect the impact of terrain and environmental factors on carbon emissions, output simulation results that are more in line with reality, and provide a more reliable decision-making basis for rail transit carbon emission management.

[0014] Optionally, high risk levels and abnormal reasons are also warned, including: Analyze whether abnormal sound information captured by acoustic sensor arrays deployed on train tracks and key equipment is obtained; If not, maintain the original setting; If yes, extract acoustic features from the abnormal sound information, and extract core sound features from the acoustic features through a dimensionality reduction method using principal component analysis; The core sound features and the prediction deviations of each component are input into the abnormal sound diagnosis model, and the abnormality cause is output; Update the abnormal reason of the warning according to the output abnormal reason, and display the high risk level together.

[0015] By adopting the above technical solution and using the acoustic sensor array information, we can deeply explore the clues behind the abnormal sound. We can extract the core features from the abnormal sound, combine the predicted deviation, and accurately output the abnormal cause through the diagnostic model, and update the warning content in real time. This not only makes the abnormal cause of the warning more accurate, but also helps the staff to quickly locate the root cause of the problem and take effective measures in time to ensure the safe and low-carbon operation of the rail transit system.

[0016] Optionally, the method further includes updating the abnormal reason of the warning according to the output abnormal reason and displaying the high risk level in parallel, as follows: Analyze the objects and abnormal locations associated with the abnormal causes, where the objects include train tracks and key equipment; If the cause of the abnormality is associated with the train track, the three-dimensional terrain model information and environmental parameters of the location area of ​​the abnormal position are obtained, and the track fault impact range is matched from the pre-built track fault impact model according to the three-dimensional terrain model information and environmental parameters of the train track associated with the abnormality cause and the location area of ​​the abnormal position, and a fault impact range analysis diagram is obtained; If the cause of the abnormality is associated with a key device, the three-dimensional terrain model information and environmental parameters of the location area of ​​the abnormal location are obtained, and the three-dimensional terrain model information and environmental parameters of the location area of ​​the abnormal location are associated with the key device and the cause of the abnormality, and the key device fault impact range is obtained from the pre-built fault propagation model, and a fault impact range analysis diagram is obtained; Obtain the flow distribution data of people in the area where the abnormal location is located, and use dynamic visualization technology to overlay and display the flow of people in real time on the fault impact range analysis diagram.

[0017] By adopting the above technical solutions, by analyzing the abnormal cause, the associated object and location, the terrain and environmental information can be obtained in a targeted manner, and the fault impact range can be accurately determined using the model. At the same time, the superposition of human flow distribution data and dynamic display can intuitively present the relationship between personnel flow and fault areas. This helps relevant departments to quickly assess potential risks, formulate scientific evacuation and response strategies, ensure personnel safety, and reduce accident losses.

[0018] Optionally, the method further includes the following steps in parallel with superimposing and displaying the personnel flow situation in real time in the fault impact range analysis diagram: Obtaining personnel behavior status data and environmental factor data in the area where the abnormal location is located, where personnel behavior status data includes movement posture, residence time and gathering pattern, and environmental factors include temperature, humidity, air flow speed and carbon dioxide concentration; Synchronize and integrate personnel behavior status data and environmental factor data in time and space dimensions; According to the different identified movements and postures of people, a personalized carbon emission coefficient is adapted for each individual; Input the environmental factor data into the dynamic correction factor analysis model of the carbon emission coefficient, output the dynamic correction factor, and adjust the carbon emission coefficient; Multiply the personalized carbon emission coefficients of all personnel in the location area by their corresponding stay time, and then add them up to obtain the total carbon emission of personnel flow in real time in the location area; On the fault impact range analysis diagram, the personnel flow carbon emission data is displayed in a preset visualization method.

[0019] By adopting the above technical solutions, the carbon emissions of personnel flow in abnormal areas can be accurately quantified. The personnel behavior and environmental data are obtained and integrated to adapt the carbon emission coefficient for individuals, and the coefficient is dynamically corrected to make it more realistic. The total carbon emissions of real-time personnel flow are calculated and visualized, which can intuitively show the impact of personnel activities on carbon emissions. It helps to understand the details of carbon emissions in the fault area, provide a basis for formulating measures to save energy, reduce emissions and optimize the environment, and help low-carbon operations and environmental management.

[0020] In the second aspect, the present application provides a rail transit carbon emission management system, which adopts the following technical solution: A rail transit carbon emission management system comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can implement the rail transit carbon emission management method as described in the first aspect when loaded and executed by the processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of a rail transit carbon emission management method according to an embodiment of the present application.

[0022] Figure 2 This is a flowchart of another embodiment of the present application when the risk level of the abnormal carbon emission is a medium risk level or a high risk level.

[0023] Figure 3 It is a flowchart of another embodiment of the present application for extracting feature vectors related to rail transit carbon emissions from environmental monitoring satellite data. DETAILED DESCRIPTION

[0024] The present application is further described in detail below in conjunction with the accompanying drawings.

[0025] Reference Figure 1 , is a rail transit carbon emission management method disclosed in this application, comprising: Step S100, obtaining multi-source heterogeneous data related to carbon emissions and performing data preprocessing.

[0026] "Multi-source" means that the data comes from multiple different channels. For example, on the one hand, it comes from the rail transit system, such as the power consumption data recorded by various energy consumption monitoring devices on the train. These data reflect the energy use during the operation of the train and are directly related to carbon emissions. On the other hand, it also comes from external relevant departments, such as the meteorological data such as temperature, wind speed, and wind direction along the line provided by the meteorological department, because meteorological conditions will affect the energy consumption of the train operation, and thus affect carbon emissions. "Heterogeneous" means that the structure and format of these data are different. For example, energy consumption data may be structured data recorded in numerical form at certain time intervals, while the meteorological report provided by the meteorological department may be semi-structured or unstructured data in various forms such as text descriptions and charts.

[0027] Data preprocessing refers to a series of operations to process the collected raw data to make it more suitable for subsequent analysis and modeling requirements. Specifically, it covers data cleaning, that is, removing abnormalities such as erroneous values, duplicate values, and missing values ​​in the data.

[0028] Step S200, extracting carbon emission time series data from the multi-source heterogeneous data related to carbon emission that has completed data preprocessing, using a preset seasonal decomposition method to remove the influence of periodic factors, and decomposing the time series into trend terms, seasonal terms, and residual terms.

[0029] Carbon emission time series data: within the scope of this rail transit carbon emission management, it refers to carbon emission related data records arranged in a certain time sequence. The method of extracting carbon emission time series data is as follows: from the multi-source heterogeneous data after the data preprocessing in step S100, filter out the data that is directly related to carbon emissions and arranged in chronological order. For example, from the integrated train energy consumption data, station equipment energy consumption data and corresponding timestamp information, extract the total carbon emission values ​​at each time point according to the established time interval (such as every hour, every day, etc.) to form time series data.

[0030] Preset seasonal decomposition method: This is a mathematical method specifically used to analyze the characteristics of time series data. Its purpose is to separate the different components contained in the time series. In this application, the part affected by factors such as seasons and cycles should be distinguished from other components. Common seasonal decomposition methods include additive models and multiplicative models. For example, for rail transit, there may be seasonal factors. For example, in summer, due to the high temperature, refrigeration equipment such as station air conditioners are frequently used, and energy consumption increases, resulting in carbon emissions showing a certain seasonal variation pattern. This method can be used to separate the parts related to this pattern (ie, seasonal components).

[0031] Trend term: represents the overall trend of change in time series data over a long period of time. It reflects the general trend of carbon emissions over time and is a relatively stable and slowly changing component. Seasonal term: As mentioned earlier, it is a component related to the seasonal law in the time series. Residual term: It is the remaining part after removing the trend term and the seasonal term. It contains random fluctuation factors in the time series that cannot be explained by the trend and seasonal laws. For example, on a certain day, due to a sudden equipment failure, the train temporarily increased its energy consumption, causing abnormal fluctuations in carbon emissions on that day. After stripping off the influence of trend and seasonal factors, this part of the abnormal fluctuation is reflected in the residual term.

[0032] Step S300, extracting preset deep features based on the decomposed trend term, seasonal term and residual term time series information.

[0033] Deep features refer to the features with representative and key information extracted from the original data after a series of complex and in-depth processing and analysis.

[0034] The deep feature extraction of trend items can involve the following: 1. Long-term trend slope and curvature: Calculate the slope of the trend item over a longer period of time, which represents the average rate of change of carbon emissions over time. For example, fit the trend item curve through linear regression and other methods to obtain its slope. A positive slope indicates that carbon emissions are generally on an upward trend, while a negative slope indicates a decline. At the same time, calculate the curvature of the curve, which can reflect the acceleration or deceleration of the trend change. If the curvature gradually increases, it means that the rate of increase or decrease of carbon emissions is accelerating. 2. Trend turning points and cycles: Analyze whether there are obvious turning points in the trend item, that is, the point where the carbon emission trend changes from rising to falling or vice versa. Determine the time and frequency of these turning points, and whether there is a periodic pattern. For example, it is observed that there is a turning point in the slowdown of carbon emission growth every few years, which is of great significance for predicting future changes in carbon emission trends.

[0035] Deep feature extraction of seasonal items can involve the following: 1. Seasonal peaks and valleys: Identify the peaks and valleys of carbon emissions in each season in the seasonal items. 2. Seasonal fluctuation amplitude: Calculate the fluctuation amplitude of carbon emissions in each season, that is, the difference between the peak and valley values. By comparing the fluctuation amplitudes of different seasons, we can understand which seasons have more dramatic changes in carbon emissions. 3. Seasonal phase and delay: Analyze the phase of the seasonal item, that is, the starting position of the seasonal change on the timeline. In addition, consider whether there is a delay in seasonal changes in different years or time periods.

[0036] Deep feature extraction of residual items may involve the following: 1. Residual standard deviation and variance: Calculate the standard deviation and variance of the residual items, which reflect the degree of dispersion of the residuals. A larger standard deviation or variance means that the residuals fluctuate more, that is, there are more random fluctuations that cannot be explained by trend and seasonal factors. Outlier detection and frequency: Detect outliers in the residual items by setting a certain threshold. Outliers may be abnormal fluctuations in carbon emissions due to equipment failure and other reasons. Statistical frequency and distribution of outliers can help to detect potential problems in a timely manner.

[0037] Step S400, input the extracted deep features into a pre-built ARIMA model, and predict and obtain the numerical range of the carbon emission trend item, seasonal item, and residual item within a preset time range in the future as the ARIMA model prediction value.

[0038] ARIMA model: ARIMA, or autoregressive moving average model, is a commonly used time series forecasting model. In rail transit carbon emission management, this model can be used to predict future carbon emissions based on historical carbon emission-related data features (that is, the deep features extracted earlier). The future preset time range refers to the time period for carbon emissions forecasting that is set in advance according to actual needs, such as the next month, quarter, or year.

[0039] Step S500, comparing the ARIMA model prediction value with the real-time carbon emission related data, and calculating the prediction deviation of each component, each component including the carbon emission trend term, the seasonal term and the residual term.

[0040] The ARIMA model prediction value is the numerical range of the carbon emission trend item, seasonal item and residual item for the future preset time range predicted by using the ARIMA model in step S400 based on the carbon emission depth characteristics extracted in the early stage. For example, it is predicted that in the next quarter, the value of the trend item may be in a specific range, the seasonal item will show a fluctuation value range of the corresponding cycle, and the residual item also has a corresponding expected fluctuation range. The results of these predictions constitute the ARIMA model prediction value, which is an estimated reflection of the future carbon emissions.

[0041] The forecast deviation is the difference between the ARIMA model prediction and the real-time carbon emission related data. It is calculated in a certain way and can intuitively reflect the accuracy of the model prediction and the deviation between the actual carbon emission and the expected value. For each component (trend term, seasonal term, residual term), its forecast deviation can be calculated separately to understand the forecast accuracy of different aspects in detail.

[0042] Step S600, input the prediction deviation of each component into the pre-built carbon emission early warning analysis model, and output the abnormal risk level of carbon emission and the abnormal reason. The abnormal risk level of carbon emission includes low risk level, medium risk level and high risk level.

[0043] The carbon emission early warning analysis model is a model specially built for the carbon emissions of rail transit. It is based on relevant technologies and algorithms such as machine learning and data analysis. It comprehensively analyzes and processes the prediction deviation data of each input component (carbon emission trend item, seasonal item and residual item), so as to determine the abnormal risk level of the current carbon emissions and give the corresponding abnormal reasons.

[0044] Model analysis and judgment mechanism: The carbon emission early warning analysis model will comprehensively consider the input prediction deviation data of each component according to the pre-set algorithms and rules. It will refer to the correspondence between different deviation combinations and risk levels and abnormal causes learned during previous training. For example, if the trend item deviation continues to increase and there are many abnormal values ​​in the residual item at the same time, the model may determine that this situation is consistent with the characteristic pattern of high risk levels in the past after comprehensive analysis, and then evaluate it in the direction of judging it as a high risk level.

[0045] Step S700: If the current carbon emission abnormality risk level is a low risk level, the visualization interface displays the current overall carbon emission value, the deviation from the predicted value, and a trend arrow indicating the direction of carbon emissions within a preset time range in the future.

[0046] Visualization interface: An interactive interface that displays data in the form of intuitive graphics, charts, etc., which can transform complex carbon emission data into easy-to-understand visual information, allowing staff to quickly grasp the overall situation.

[0047] Display of current overall carbon emission values: Display the overall carbon emission values ​​of the current rail transit system in the form of numbers or bar graphs in a prominent position on the visualization interface. Display of deviation range: Calculate the deviation range between the actual carbon emission value and the predicted value, and present it in the form of percentage on the visualization interface.

[0048] Trend arrow display: Draw trend arrows on the visualization interface based on the carbon emission trend within the preset time range predicted by the ARIMA model. If the carbon emission is predicted to be on an upward trend, draw an upward arrow; if it is on a downward trend, draw a downward arrow; if it is basically stable, draw a horizontal arrow. The approximate range of the predicted change can be marked next to the arrow, such as "Carbon emissions are expected to rise / fall by [X]% in the next week" to help users understand the future trend of carbon emissions.

[0049] Step S800: If the risk level of the abnormal carbon emissions this time is a medium risk level, the visualization interface displays the detailed prediction deviation curve of each component and the dynamic change process of the deviation.

[0050] Detailed forecast deviation curves for each component: For the carbon emission trend term, seasonal term, and residual term, the deviation between the actual value and the ARIMA model forecast value is shown in the form of curves. These curves can clearly reflect the deviation size and change trend of each component at different time points.

[0051] Dynamic changes in deviations: Dynamically demonstrate the continuous changes in the prediction deviations of each component over time, allowing users to intuitively see how the deviations develop and evolve.

[0052] Step S900: If the risk level of the carbon emission abnormality is a high risk level, the visualization interface displays the detailed prediction deviation curve of each component and the dynamic change process of the deviation, and warns of the high risk level and the cause of the abnormality.

[0053] The warning involves the following: Highlight high-risk levels: Use eye-catching colors (such as red) to mark the words "high risk level", or place it in a prominent position on the visual interface, such as the top center of the interface, and present it in a larger font to ensure that staff can see at a glance that current carbon emissions are in a high-risk state.

[0054] Display the cause of the abnormality: In the area immediately below or next to the high-risk level logo, clearly and concisely list the cause of the abnormality that is determined to be a high-risk level. For example, write in text form "Cause of abnormality: A key equipment failure has caused a significant increase in energy consumption, which in turn has caused carbon emissions to exceed expectations, and the failure has lasted for [X] days", so that staff can quickly know the root cause of the high risk, providing a key basis for quickly formulating and implementing corresponding solutions.

[0055] Reference Figure 2 If the risk level of this abnormal carbon emission is medium risk or high risk, the following steps are also included: Step SA00, obtaining environmental monitoring satellite data, wherein the environmental monitoring satellite data includes regional meteorological dynamics and geographic information.

[0056] Environmental monitoring satellite data: Data collected by dedicated environmental monitoring satellites and transmitted back to the ground.

[0057] Regional meteorological dynamics: refers to the changes in meteorological elements in a specific area over time, including but not limited to temperature, air pressure, humidity, wind speed, wind direction, precipitation, etc.

[0058] Geographic information: covers various information related to the geographical features of the earth's surface, such as topography (mountains, plains, rivers, etc.), altitude, land use type (cities, forests, farmland, etc.).

[0059] Data screening and extraction involve the following: 1. Determine the regional scope: Combine the actual direction and coverage area of ​​the rail transit line to define the corresponding geographical scope in the satellite data. 2. Time screening: According to the time requirements of rail transit carbon emission analysis, filter out satellite data within a specific time period.

[0060] Step SB00, extract the feature vector related to rail transit carbon emissions from the environmental monitoring satellite data, take the feature vector related to rail transit carbon emissions as the input object, input it into the pre-built comprehensive model that simulates the changes in carbon emissions under different emergency scenarios, and output the simulation results.

[0061] Feature vector: In this context, it refers to a set of data extracted from environmental monitoring satellite data that can reflect characteristics closely related to rail transit carbon emissions.

[0062] Comprehensive model for simulating carbon emission changes under different emergency scenarios: This is a specially constructed model that aims to simulate how rail transit carbon emissions will change under various emergency scenarios such as extreme weather, natural disasters, and sudden equipment failures by inputting relevant characteristic data. The model comprehensively considers multiple factors and their interrelationships, and is constructed based on mathematical algorithms, physical principles, and past actual case data. It can output the estimated changes in carbon emissions under the corresponding scenarios, providing a reference for formulating response strategies in advance.

[0063] Step SC00, based on the output simulation results, draw a curve of rail transit carbon emissions changing over time within a future preset time range, and mark the emergency scenario related information.

[0064] Future preset time range: It is a time period set in advance according to actual needs and analysis purposes, such as the next three days, a week, or a month.

[0065] The specific operations are as follows: 1. Construct a coordinate system: In a visual drawing interface (such as professional drawing software or a customized data analysis platform interface), first determine the coordinate system. Usually time is used as the horizontal axis and divided into equal intervals. The size of the interval depends on the length of the preset time range and the requirements for display accuracy. For example, if the preset time range is the next week, it may be divided into days; if the preset time range is the next month, it can be divided into weeks. Set the vertical axis to the carbon emission value, and according to the approximate range of the carbon emission values ​​in the simulation results, reasonably determine the scale range of the vertical axis to ensure that the changes in carbon emissions can be fully and clearly displayed. 2. Draw the change curve: According to the carbon emission values ​​corresponding to different time points in the simulation results, find the corresponding coordinate points in the coordinate system in turn, and then connect these points in turn with smooth curves. If multiple emergency scenarios are simulated, it is necessary to draw corresponding curves for each scenario, and they can be distinguished by different colors and line styles (such as solid lines, dotted lines, etc.), which is convenient for viewing and comparing the changes in carbon emissions under different scenarios. 3. The relevant information of the marked emergency scenarios is as follows: 3.1. Scenario name marking: In a prominent position above or next to each curve, mark the name of the emergency scenario represented by the curve, such as "rainstorm emergency scenario" and "earthquake emergency scenario", so that the viewer can clearly see at a glance which specific situation this curve corresponds to. 3.2. Key influencing factor marking: Briefly mark and explain the key factors that have an important impact on carbon emissions in the emergency scenario. For example, mark "water accumulation on the track leads to increased friction and energy consumption" next to the "rainstorm emergency scenario" curve; mark "partial line interruption, train detours increase mileage, energy consumption and carbon emissions increase" next to the "earthquake emergency scenario" curve, etc. This can help viewers to more deeply understand the reasons behind the changes in carbon emissions reflected by the curve. 3.3. Special time point marking: For critical time points in emergency scenarios where carbon emissions may change, such as turning points where carbon emissions suddenly increase or decrease, mark the corresponding positions on the curve and briefly explain the reasons. For example, mark "carbon emissions begin to decrease here because equipment repairs are completed" at a certain time point on a curve, so that viewers can grasp the details of carbon emission changes in more detail.

[0066] Reference Figure 3 ,From the environmental monitoring satellite data, the feature vectors related to rail transportation carbon emissions are extracted,including: Step SB10, obtaining the abnormal risk level of carbon emissions this time.

[0067] Step SB20, determine the real-time satellite data characteristic fluctuation information based on the environmental monitoring satellite data analysis, input the carbon emission abnormal risk level and the real-time satellite data characteristic fluctuation information into the pre-trained feature selection standard analysis model, and output the feature selection standard.

[0068] Real-time satellite data characteristic fluctuation information: a collection of information analyzed from environmental monitoring satellite data that reflects the fluctuations and changes in various environmental-related data (such as meteorological elements, geographical features, etc.) at the current moment or in the near future.

[0069] The analysis and determination of the characteristic fluctuation information of real-time satellite data specifically involves the following contents: 1. Analysis of meteorological data fluctuations: For the temperature data obtained from environmental monitoring satellite data, the range (difference between the maximum and minimum values), standard deviation and other statistics within a certain time period (such as every hour, every day, etc.) are calculated to measure the degree of temperature fluctuation. For example, if the temperature range in a certain area reaches 10°C in one day, it means that the temperature fluctuation is large, which may have a significant impact on the temperature control energy consumption and carbon emissions of rail transit. 2. For wind speed and wind direction data, analyze the frequency and amplitude of changes in different time periods. For example, by observing whether the wind direction changes frequently within a few hours, whether the increase or decrease of wind speed is drastic, etc., judge its dynamic impact on the running resistance and energy consumption of the train, and organize these analysis results to form characteristic fluctuation information in meteorological data. 3. Analysis of geographic data fluctuations: If it involves terrain and landform related data in geographic information (such as altitude, slope, etc.), check whether these geographical features have changed in the area along the line, such as whether there is a slight rise and fall in local altitude due to geological activities, or whether new construction has changed the local terrain slope. By comparing geographic data from different periods, the fluctuation of geographic features can be determined. Although geographic data usually changes more slowly than meteorological data, some special situations will also affect rail transit energy consumption and carbon emissions, and they also need to be organized into corresponding fluctuation information. 4. Integrate fluctuation information: Summarize and integrate the fluctuation information obtained from various aspects such as meteorological data and geographic data to form a complete set of real-time satellite data feature fluctuation information for subsequent use as model input. For example, the integrated information may appear in the form of "daily temperature range of 10°C, wind speed changes 3 times per hour, and the altitude of a certain section of road has increased by 2 meters compared to last month..."

[0070] Inputting data to the feature selection standard analysis model and obtaining output involves the following: 1. Data input: The carbon emission abnormal risk level obtained in step SB10 (such as "low risk level", "medium risk level" or "high risk level" and other forms of expression) and the real-time satellite data characteristic fluctuation information just analyzed and determined are sorted according to the input format required by the feature selection standard analysis model. For example, the risk level is in the form of a string, and the fluctuation information is in a specific data structure (such as a dictionary, list, etc., according to the model settings) and is respectively transmitted to the corresponding input interface of the model to ensure that the model can accurately identify and receive these data. 2. Model operation and output: The feature selection standard analysis model will conduct a comprehensive analysis of the input carbon emission abnormal risk level and real-time satellite data characteristic fluctuation information based on the pre-learned knowledge and algorithm rules. For example, if the input is a medium risk level with large temperature fluctuations and a certain section of the road with terrain changes, the model will refer to the experience of similar situations in the past and weigh the importance of different environmental factors on carbon emissions.

[0071] Step SB30, extracting feature vectors related to rail transit carbon emissions from environmental monitoring satellite data according to feature selection criteria.

[0072] Extracting and integrating data to form feature vectors involves the following processes: 1. Data extraction: According to the analyzed feature selection criteria, the various types of located data are extracted one by one. 2. Feature vector integration: The extracted data are arranged and integrated in a certain order to form a complete feature vector.

[0073] The characteristic vectors related to rail transit carbon emissions are used as input objects and input into a pre-built comprehensive model that simulates carbon emission changes under different emergency scenarios. The simulation results include: Step SBA0: Analyze whether the three-dimensional terrain model information scanned by the laser radar device deployed on the top of the train or at key points along the line and the environmental parameters detected by the dust sensor network deployed along the rail transit line and stations are obtained. If not, execute step SBB0; if yes, execute step SBC0.

[0074] LiDAR device: This is an active sensor device that uses laser beams to detect the position, speed, shape and other information of a target.

[0075] Three-dimensional terrain model information: information contained in the three-dimensional model constructed by data generated by scanning with a laser radar device, which presents the spatial terrain conditions along the rail transit line, including the coordinate position of each point, the height value, and the undulation of the terrain.

[0076] Dust sensor network: A network system formed by multiple dust sensors deployed in various locations of rail transit lines and stations according to a certain layout and connection method. These dust sensors can detect various environmental parameters in the surrounding environment in real time, such as the concentration of particulate matter in the air, humidity, temperature, etc., and aggregate the data detected at each point through network transmission, providing a basis for a comprehensive understanding of the surrounding environment of rail transit, and then analyzing its impact on carbon emissions.

[0077] Environmental parameters: refers to the various specific indicator values ​​detected by the dust sensor network that reflect the physical properties of the environment along the rail transit line and around the station, such as the air particulate concentration, humidity, temperature, etc. mentioned above. Different environmental parameters will affect the operating status and energy consumption of train equipment to varying degrees, and thus are related to carbon emissions.

[0078] Step SBB0, taking the characteristic vector related to rail transit carbon emissions as an input object, inputting it into a pre-built comprehensive model that simulates the changes in carbon emissions under different emergency scenarios, and outputting the simulation results.

[0079] Feature vectors related to rail transit carbon emissions: This is a data set that has been screened and extracted through a series of previous steps and can reflect various key factors that affect rail transit carbon emissions. It covers characteristic information closely related to carbon emissions in aspects such as meteorology and geography, and is combined in a specific order and format as the basic input content for subsequent model analysis, such as wind speed, wind direction, temperature, altitude and other related data elements.

[0080] A comprehensive model that simulates changes in carbon emissions under different emergency scenarios: This is a mathematical model specifically built for the field of rail transit. It is based on a large amount of historical data, physical principles, and relevant professional knowledge. It comprehensively considers a variety of possible emergency scenarios (such as extreme weather, equipment failures, natural disasters, etc.) and the interaction of various factors on carbon emissions under these scenarios. By inputting corresponding characteristic data, it can simulate the changes in rail transit carbon emissions over time under different emergency conditions, providing a reference for decision-making.

[0081] Step SBC0, extracting the laser radar feature vector from the acquired three-dimensional terrain model information, calibrating and integrating the acquired environmental parameters, and extracting the smart dust sensor network feature vector.

[0082] Extracting LiDAR feature vectors from three-dimensional terrain model information involves the following: 1. Identifying key terrain elements: Conducting a comprehensive and in-depth analysis of the three-dimensional terrain model information to accurately identify terrain elements that are closely related to rail transit carbon emissions. 2. Quantifying and constructing feature vectors: Quantifying the identified key terrain elements, expressing the slope in a specific angle value or percentage form, the degree of terrain undulation can be quantified by the change in height difference within a unit distance, and the altitude difference directly records the height difference of the corresponding section. Organize these quantified terrain element data in a certain logical order to construct the LiDAR feature vector.

[0083] The following process is involved in calibrating and integrating environmental parameters and extracting the feature vector of the smart dust sensor network: Environmental parameter calibration: 1. Temperature calibration: Due to the differences in environmental microclimates at different dust sensors and the accuracy deviation of the sensors themselves, the temperature data detected by each sensor may have errors. To this end, a standard thermometer or a reference source with a known temperature is used to perform comparative measurements near the installation location of the sensor, and the temperature measurement values ​​of each sensor are corrected according to the comparison results so that it can accurately reflect the actual ambient temperature. 2. Humidity calibration: Similarly, for humidity sensors, with the help of professional high-precision humidity calibration equipment, or using a standard environmental chamber with humidity controlled at a stable value, the humidity values ​​detected by each sensor are calibrated to eliminate measurement deviations caused by sensor aging, installation conditions and other factors, ensuring that all humidity data are under a unified and accurate standard. 3. Calibration of other parameters: For other environmental parameters such as air pressure, particulate matter concentration, and harmful gas concentration, corresponding professional calibration methods are also used, such as comparing and calibrating air pressure values ​​with standard barometers, and using high-precision particulate matter monitoring instruments to calibrate particulate matter concentrations, etc., to ensure the accuracy and reliability of each environmental parameter data, laying the foundation for subsequent integrated analysis.

[0084] Integration and extraction of feature vectors: From the calibrated environmental parameters, select key indicators that are highly correlated with rail transit carbon emissions. For example, temperature and humidity will affect the energy consumption of the temperature control system in the train compartment, excessive particle concentration may increase the energy consumption of the ventilation system, and changes in air pressure will affect the aerodynamic performance of the train and thus affect energy consumption. These key indicators are selected as the core elements for constructing feature vectors. Arrange the selected key environmental parameter data in a reasonable order to form the smart dust sensor network feature vector.

[0085] Step SBD0, fusing the laser radar feature vector, the smart dust sensor network feature vector and the feature vector extracted based on the environmental monitoring satellite data to form a fusion vector.

[0086] Fusion vector: A new vector formed by organically integrating the characteristic vectors from the above three different sources. It integrates key information related to rail transit carbon emissions from multiple aspects such as terrain, environmental parameters and macro-environment.

[0087] The specific fusion operation involves the following process: Fusion operation: 1. Fusion in order of elements: A common fusion method is to merge the corresponding elements together in order of the elements of each feature vector to form a fusion vector. For example, assuming that the laser radar feature vector is [slope value 1, altitude difference value 1, ...], the smart dust sensor network feature vector is [temperature value 1, humidity value 1, ...], and the feature vector extracted based on environmental monitoring satellite data is [wind speed value 1, wind direction value 1, ...], then the fusion vector can be constructed in order of [slope value 1, altitude difference value 1, temperature value 1, humidity value 1, wind speed value 1, wind direction value 1, ...], so that the fused vector fully contains key feature information from different aspects. 2. Consider weighted fusion (if applicable): In some cases, depending on the importance of different features to the carbon emissions of rail transit, different weights may be assigned to the elements in each feature vector before fusion. For example, after analysis, it was found that the weight of terrain factors (represented by the laser radar feature vector) on carbon emissions in a specific scenario is 0.4, the weight of environmental parameters (smart dust sensor network feature vector) is 0.3, and the weight of the macro environment (feature vector extracted based on environmental monitoring satellite data) is 0.3. Then, when fusion, for each corresponding element, first multiply the weight corresponding to the vector to which it belongs, and then merge, so as to reflect the relative importance of each factor in the overall carbon emission impact, so that the fusion vector can more scientifically and reasonably reflect the actual situation.

[0088] Step SBE0, adjusting the architecture of the comprehensive model according to the feature vector included in the fusion vector and the preset comprehensive model architecture quantization adjustment method.

[0089] Preset quantitative adjustment method for the comprehensive model architecture: This is a set of pre-set rules and calculation methods when constructing a comprehensive model that simulates changes in carbon emissions under different emergency scenarios. It is used to make reasonable adjustments to the model architecture based on the input data characteristics (here are the characteristics contained in the fusion vector), such as changing the number of model layers, the number of neurons (for neural network models), or adjusting the weights of each parameter in the equation, etc. The purpose is to enable the model to better adapt to new input data and improve the accuracy of the model's simulation of carbon emission changes under different emergency scenarios.

[0090] The architecture of the comprehensive model refers to the internal structure of the comprehensive model that simulates the changes in carbon emissions under different emergency scenarios, including the type of mathematical model used (such as linear regression model, neural network model, etc.), the connection relationship between the components, parameter settings, etc. It determines how the model processes input data and outputs corresponding simulation results. A reasonable architecture is the key to ensuring model performance and simulation effects.

[0091] The specific operation involves the following process: Analyze the fusion vector features: 1. Feature importance assessment: Carefully study the various feature vectors and their specific elements contained in the fusion vector, and evaluate the importance of different features for simulating changes in rail transit carbon emissions. For example, analyze the weight of the terrain slope feature (from the lidar feature vector) in the fusion vector on the changes in train energy consumption and carbon emissions under a specific rail transit line and operating environment, and compare it with environmental parameter features such as temperature and humidity (from the smart dust sensor network feature vector) and macro-environmental features such as wind speed and wind direction (from the feature vector extracted based on environmental monitoring satellite data) to determine which feature is more critical in the current scenario, or which features have a strong interaction relationship. 2. Feature correlation analysis: Calculate the correlation coefficient between each feature in the fusion vector to determine the degree of association between different features. For example, check whether there is a significant correlation between the altitude feature and the temperature feature, because changes in altitude may affect the temperature, which in turn jointly affects the temperature control energy consumption and carbon emissions of the train. Through correlation analysis, we can have a deeper understanding of the synergy or mutual restriction relationship between each feature in affecting carbon emissions, providing a basis for subsequent model architecture adjustments.

[0092] Adjust the model architecture according to the quantitative adjustment method: 1. Adjust the model complexity (taking the neural network model as an example): 1.1. Adjust the number of layers: If the fusion vector contains more and complex features, according to the preset comprehensive model architecture quantitative adjustment method, it may be necessary to increase the number of layers of the model to better fit the nonlinear relationship between these features and carbon emissions. For example, it was originally a three-layer neural network model. After analysis, it was found that the features in the fusion vector require a deeper network structure to explore its internal connection. Then, according to the adjustment method, the number of model layers is increased to four or more layers to enhance the model's expressive ability so that it can accurately capture the impact of different feature combinations on carbon emission changes. 1.2. Adjust the number of neurons: At the same time, the number of neurons in each layer is also adjusted accordingly. For those layers corresponding to those parts that are closely related to carbon emissions and have rich feature changes, the number of neurons is appropriately increased to more carefully process and analyze these key feature information. For example, between the input layer and the hidden layer, if the input fusion vector contains more environmental parameter-related features and the changes are complex, the number of neurons in the hidden layer is increased so that the model can fully learn the impact of these environmental factors on carbon emissions and improve the accuracy of the simulation. 1.3. Parameter weight adjustment: For the existing parameters in the model (such as coefficients in the linear regression model, connection weights in the neural network model, etc.), the weights are redistributed according to the preset quantitative adjustment method based on the importance and correlation analysis results of the fusion vector features.

[0093] Step SBF0, input the fusion vector into the comprehensive model that simulates the changes in carbon emissions under different emergency scenarios after completing the structural adjustment, and output the simulation results.

[0094] At the same time, high risk levels and abnormal reasons are warned, including: Step S910: Analyze whether abnormal sound information captured by the acoustic sensor array deployed on the train running track and key equipment is obtained. If not, execute step S920; if yes, execute step S930.

[0095] Among them, the acoustic sensor array: a system composed of multiple acoustic sensors in a specific arrangement and combination, deployed on the train tracks and around key equipment, used to collect sound signals in an all-round and high-precision manner.

[0096] In rail transit scenarios, train tracks and key equipment will generate sounds of specific frequencies and characteristics when they are operating normally. When cracks appear on the tracks, loose parts appear, or key equipment (such as train traction motors, braking systems, etc.) fail, the sounds they make will change. Acoustic sensor arrays can capture these sound changes and provide raw data for judging the operating status of equipment and track conditions.

[0097] Abnormal sound information: refers to sound data that is significantly different from the sound emitted by the train tracks and key equipment during normal operation in terms of acoustic characteristics such as frequency, amplitude, and timbre. These abnormal sounds are often early signs of equipment failure or track problems. Function: By analyzing abnormal sound information, potential sources of failure can be quickly located, and timely measures can be taken to repair them to avoid further deterioration of the failure, thereby ensuring the safe operation of rail transit. It also helps to analyze the causes of abnormal carbon emissions, because equipment failure or track abnormalities may lead to increased energy consumption, which in turn affects carbon emissions.

[0098] Step S920, maintaining the original settings.

[0099] Step S930: extracting acoustic features from the abnormal sound information, and extracting core sound features from the acoustic features by using a dimensionality reduction method of principal component analysis.

[0100] Principle and calculation of principal component analysis: 1. Principle: PCA aims to transform the original high-dimensional feature data into a set of new, mutually orthogonal low-dimensional feature vectors, namely principal components, through linear transformation. 2. Calculation steps: Data standardization: Standardize each column of data in the feature matrix so that its mean is 0 and its variance is 1. This is to eliminate the differences in dimensions and numerical ranges between different features and ensure that each feature has equal importance in subsequent calculations. For example, for the feature of short-time energy, its numerical range may be large, while some frequency features have a small numerical range. Through standardization, they can be treated fairly in principal component analysis. Calculate the covariance matrix: Calculate the covariance matrix for the standardized data. The covariance matrix reflects the degree of linear correlation between different features. For example, if there is a strong positive correlation between two acoustic features (such as short-time energy and a specific frequency component), the corresponding element value in the covariance matrix will be larger. Eigenvalue and eigenvector calculation: Solve the eigenvalue and eigenvector of the covariance matrix. The eigenvalue indicates the variance contained in each principal component, and the eigenvector determines the direction of the principal component. Sort the eigenvalues ​​from large to small, select the eigenvectors corresponding to the first few larger eigenvalues, and the new coordinate system formed by these eigenvectors is the principal component space after dimensionality reduction. For example, by calculating 10 eigenvalues, select the eigenvectors corresponding to the first 3 largest eigenvalues, and project the original 10-dimensional acoustic features into this 3-dimensional principal component space to achieve dimensionality reduction.

[0101] Core sound feature extraction: The original acoustic feature data is projected onto the selected principal component space to obtain the core sound features after dimensionality reduction. These core sound features not only retain most of the important information in the original acoustic features, but also remove redundant and weakly relevant information, making subsequent analysis more efficient and accurate.

[0102] In step S940, the core sound features and the prediction deviations of the components are input into the abnormal sound diagnosis model, and the cause of the abnormality is output.

[0103] The abnormal sound diagnosis model is built based on a large amount of historical data, equipment operation principles, and acoustic analysis knowledge. Common model types include neural network models (such as multi-layer perceptrons, convolutional neural networks, etc.), decision tree models, support vector machine models, etc. Select the most suitable model based on the characteristics of the rail transit field and the existing data.

[0104] After receiving the input data, the abnormal sound diagnosis model performs complex operations based on its internal algorithms and structures. Taking the neural network model as an example, the data will first pass through the input layer, and then undergo multiple nonlinear transformations and feature extractions in the hidden layer. The neurons in the hidden layer will perform weighted summation on the input data according to the pre-trained weights and biases, and perform nonlinear transformations through activation functions (such as ReLU, Sigmoid, etc.), gradually mining the potential patterns and features in the data. Different models operate in different ways, but they all aim to find the correlation between the input data and the known cause of the abnormality.

[0105] After the model's operation and analysis, the final output is the diagnosis result of the cause of the abnormal sound. This result may be presented in various forms, such as a text description, clearly stating that the cause of the abnormality is "wear of the train traction motor bearing" and "loose track fasteners", etc.; it may also be output in the form of classification labels, corresponding to different abnormal types, such as "equipment failure type 1" and "track problem type 3". At the same time, the output of the model may also be accompanied by relevant confidence information, indicating the degree of certainty of the model's diagnosis results. For example, the output result is "wear of the train traction motor bearing, confidence level 90%", which provides a clear direction and reference basis for subsequent troubleshooting and carbon emission abnormality handling.

[0106] Step S950: update the abnormal reason of the warning according to the output abnormal reason, and display the high risk level together.

[0107] Choose a suitable display platform to present high-risk levels and updated abnormal reasons. Common display platforms include the large-screen monitoring system of the rail transit control center, the handheld terminal devices of operation and maintenance personnel (such as tablets, smart phones), and the dedicated carbon emission management software interface, etc. Configure accordingly according to the characteristics of different platforms to ensure that the information can be displayed clearly and intuitively.

[0108] Use intuitive visualization to display high-risk levels and abnormal causes. For high-risk levels, use eye-catching colors (such as red) and larger fonts to highlight them, such as displaying the words "high risk level" on the screen with a red background. For abnormal causes, use concise and clear text to present them, avoiding overly complex terms and sentences. For example, below the high-risk level, display "Abnormal cause: excessive wear of brake pads in the train braking system."

[0109] A rail transit carbon emission management method also includes the steps of updating the abnormal cause of the warning according to the output abnormal cause and displaying the high risk level together, as follows: Step Sa00, analyzing the objects and abnormal locations associated with the abnormal causes, wherein the objects include train tracks and key equipment.

[0110] Extract keywords related to objects and locations from the description of the cause of the abnormality. Taking "severe wear of the brushes of the train traction motor" as an example, the keywords are "train traction motor" (associated object) and "brush" (to further clarify the specific components of the associated object). For "turn-to-turn short circuit in the choke transformer of the track circuit", the keywords are "track circuit" (associated object) and "choke transformer" (to clarify the specific equipment of the associated object). By extracting these keywords, a basis is provided for the subsequent determination of the objects and locations associated with the abnormality.

[0111] The judgment of the associated object is as follows: 1. Determine the object category: According to the extracted keywords, determine whether the object associated with the abnormal cause belongs to the train track or the key equipment. For example, if the abnormal cause involves "loose track fasteners", it is obvious that the associated object is the train track; while for "train air conditioning compressor failure", the associated object is the key equipment of the train. Establishing clear object category judgment standards will help to quickly classify and handle different types of abnormal situations. For example, formulate rules: all objects related to descriptions such as track structure, track bed, and track ancillary facilities are determined to be objects related to train tracks; those related to descriptions such as train onboard equipment, power system, and control system are determined to be objects related to key equipment. 2. Refine object information: For objects determined to be related to train tracks, further refine the information. If the keyword is "track fasteners", it is necessary to clarify which section of the track is the fastener, whether it is a straight section, a curved section, or a fastener in the turnout area. It can be accurately located through information such as track numbers and mileage marks. For example, in the rail transit route map, the track number and mileage information are clearly marked, so that when an abnormality is found, the specific track section can be quickly located. For key equipment-related objects, it is necessary to determine the specific location of the equipment on the train. For example, the train traction motor is generally installed on the bogie, so it is necessary to clarify which bogie on which carriage the traction motor is on. This information can be accurately obtained by consulting the train equipment layout diagram and related technical documents.

[0112] The abnormal location can be determined based on the location of the object and combined with auxiliary information positioning.

[0113] If the associated object is a train track, the specific track location area where the abnormal position is located is determined based on the track number, mileage mark, and detailed records of track facilities. For example, the known cause of the abnormality is "cracked track weld at K10+200". Through the mileage measurement system of the rail transit line and the track facility ledger, the location of the weld in the actual track line can be accurately found. For key equipment, the train equipment layout diagram and equipment numbering system are used to determine the specific installation location of the equipment. For example, the train's braking system is distributed in various carriages. Through the carriage number and braking equipment number, the carriage and specific installation location of the faulty braking component can be quickly located.

[0114] Step Sb00: If the cause of the abnormality is associated with the train track, the three-dimensional terrain model information and environmental parameters of the area where the abnormal position is located are obtained, and the track fault impact range is matched from the pre-constructed track fault impact model according to the train track associated with the abnormality cause and the three-dimensional terrain model information and environmental parameters of the area where the abnormal position is located, and a fault impact range analysis diagram is obtained.

[0115] Acquisition of 3D terrain model information: Find the 3D terrain model information of the area where the abnormal position is located from the Geographic Information System (GIS) or a special track terrain database. This information may be stored in a special file format, such as a digital elevation model (DEM) file, which contains detailed terrain data of the area where the track is located. For example, ArcGIS software can be used to open the corresponding DEM file, and by locating the coordinates of the abnormal position, the terrain information of the area can be extracted, including the altitude, slope and other information at different locations.

[0116] Environmental parameter acquisition: Obtain environmental parameters of the area where the abnormal location is located from the environmental monitoring system. The environmental monitoring system may include multiple sensors distributed along the track, which will collect environmental data in real time or periodically and store the data in the database. The temperature, humidity, wind speed and other environmental parameters of the area where the abnormal location is located can be extracted from the database through data query statements.

[0117] The anomaly causes, three-dimensional terrain model information and environmental parameters of the area where the anomaly is located are collated to meet the input requirements of the track fault impact model.

[0118] The sorted data is input into the pre-built track fault impact model. The model will calculate the impact range of the fault based on its internal algorithms and rules, taking into account the impact of terrain on fault propagation (e.g., on a steep slope, a track fault may cause the train to slide down, which may have a larger impact range), the impact of environmental factors on the track structure (e.g., high temperature may cause track expansion, which may expand the scope of the fault), and the characteristics of the abnormal cause itself (e.g., track cracks may expand at different speeds and directions in different terrains and environments).

[0119] Regarding the generation of fault impact range analysis diagrams, the type of diagram selection: According to the needs of the analysis, select an appropriate diagram to represent the fault impact range. It can be a two-dimensional impact range diagram, which shows the impact range of the fault on the track plane, with the track line as the reference, marking the areas that may be affected; it can also be a three-dimensional impact range diagram, which more intuitively shows the impact range on the terrain, such as showing the impact range at different altitudes and plane positions.

[0120] Use of drawing tools: Use professional drawing software (such as AutoCAD, MATLAB, etc.) to generate fault impact range analysis diagrams. In the diagram, use different colors or patterns to distinguish the impact ranges of different degrees, such as red areas for severely affected areas, yellow areas for moderately affected areas, and green areas for slightly affected areas. Mark key information, mark the abnormal location, terrain contours, environmental factors (such as wind direction arrows to indicate the direction of wind impact), and possible expansion directions of the fault on the diagram to help analysts intuitively understand the impact range of the fault.

[0121] Step Sc00, if the cause of the abnormality is associated with key equipment, obtain the three-dimensional terrain model information and environmental parameters of the area where the abnormal location is located, and according to the three-dimensional terrain model information and environmental parameters of the area where the abnormal location is located and associated with the key equipment due to the abnormality, obtain the impact range of the key equipment fault from the pre-built fault propagation model, and obtain the fault impact range analysis diagram.

[0122] Integrate the acquired 3D terrain model information, environmental parameters, and key equipment information associated with abnormal causes. Convert the 3D terrain model data into a format suitable for model input, such as organizing terrain elevation, slope and other data into a matrix form; organize environmental parameters into arrays by category, such as [temperature value, humidity value, air pressure value, wind speed value, wind direction value]; at the same time, associate the type, model, failure mode and other information of key equipment with the above data in text or specific coding form to form a complete input data set.

[0123] Input the fault propagation model: The integrated data set is accurately input into the model according to the input interface and format required by the pre-built fault propagation model. For example, if the model is a machine learning model developed based on Python, the model's prediction function is called and the organized input data set is passed in as a parameter to trigger the model's calculation and analysis process.

[0124] Model calculation: After receiving the input data, the fault propagation model first analyzes the failure mode of the key equipment and determines the possible propagation path of the fault within the equipment based on the physical structure and working principle of the equipment. For example, if the key equipment is the brake system of the train, when a brake cylinder failure occurs, the model will analyze how the fault affects the action of the brake shoe and the transmission of the braking force based on the pipeline connection and mechanical transmission relationship of the brake system.

[0125] Next, the model considers the impact of three-dimensional terrain model information and environmental parameters on fault propagation. In areas with complex terrain, equipment failure may cause changes in the train's operating status, which in turn affects other related equipment. For example, on a climbing section, a traction motor failure may cause the train to slow down, causing the braking system to operate frequently and expanding the scope of the fault. In terms of environmental parameters, high temperatures may accelerate equipment damage, and high humidity may cause electrical equipment to short-circuit. The model will combine these factors to perform complex calculations and simulations.

[0126] Result acquisition: After the model calculation, the result data of the impact range of key equipment failure is obtained. These data may be presented in different forms, such as a list of affected equipment, the geographical coordinate range of the fault-affected area, the hierarchical structure of the affected system, etc. For example, the result data shows that the failure of the train's brake system will affect the power supply system and signal system of the same carriage, as well as some control circuits of the adjacent carriages, and the specific location coordinates of these affected equipment and systems are given.

[0127] Step Sd00, obtaining the flow distribution data of people in the area where the abnormal location is located, and using dynamic visualization technology to overlay and display the flow of people in real time on the fault impact range analysis diagram.

[0128] The acquisition of pedestrian flow distribution data can be based on sensor data collection or video surveillance data extraction.

[0129] Data collection based on sensors is as follows: 1. Infrared sensors: Infrared sensors are deployed at key locations such as entrances and exits, passages, and platforms of rail transit stations. These sensors identify the entry, exit, and movement of people by detecting infrared radiation from the human body. For example, a row of infrared sensors is set at the edge of the platform. When a passenger passes by, the sensor will generate an electrical signal change according to the change in infrared intensity. The system counts and analyzes these signals to count the number of people passing through the area. At the same time, combined with the location information of the sensor, it can preliminarily determine the approximate distribution of people on the platform. 2. Pressure sensors: Pressure sensors are installed in areas where people frequently pass by, such as stairs and escalators. When passengers step on these devices, the pressure sensors will sense the pressure changes and convert them into electrical signals to transmit to the data acquisition system. By analyzing and processing the pressure signals, not only can the number of people passing through be counted, but also the walking speed and density of people can be inferred based on the changes in pressure distribution.

[0130] Data extraction based on video surveillance is as follows: 1. Video analysis algorithm: Use advanced video analysis algorithms to process the surveillance camera videos in the rail transit area in real time. These algorithms can identify the outline of the human body in the video, track and analyze it through image recognition technology. For example, by using the background difference method, the current video frame is compared with the pre-set background image to extract the moving human target. Then, using the target tracking algorithm, each identified human target is continuously tracked to record its movement trajectory and stop position. By analyzing a large number of video frames, the flow of people in different time periods and locations is statistically calculated. 2. Deep learning model: Use target detection and tracking models based on deep learning, such as the YOLO (YouOnlyLookOnce) series of models, Deepsort, etc. These models are trained with a large amount of human image data and can accurately identify the human body in the video and track it efficiently. For example, the video shot by the surveillance camera is input into the trained YOLO model in real time. The model will quickly detect the human target in the video and assign a unique identifier to each target. Then, the Deepsort model will continuously track the target based on its motion and appearance features, thereby obtaining the flow trajectory and real-time location information of the personnel. By integrating and analyzing the video data of multiple surveillance cameras, it is possible to obtain comprehensive data on the distribution of human traffic in the area where the abnormal location is located.

[0131] The principle of layer overlay for overlay display: In the visualization tool, the personnel flow data is overlaid as a new layer on the base map of the fault impact range analysis map. Each layer has its own independent attributes and data, but when visualized, they are overlaid and displayed in a certain order. For example, in ArcGIS, the fault impact range analysis map is used as the base layer and set to the bottom layer display; the personnel flow data is created in the form of a point layer or a line layer and set to the upper layer display. By adjusting the transparency and display order of the layer, it is ensured that the personnel flow situation can be clearly displayed on the fault impact range analysis map without obstructing the key information of the base map.

[0132] The display effect optimization can be as follows: Color and symbol design: In order to make the flow of people more intuitive and easy to understand, the color and symbol design of the superimposed flow data is carried out. For example, different colors are used to represent different density of human traffic. Red represents highly densely populated areas, yellow represents medium-density areas, and green represents sparsely populated areas. For the direction of human traffic, arrow symbols can be used to represent it. The direction of the arrow represents the direction of human traffic, and the length or thickness of the arrow can represent the speed of traffic. Through reasonable color and symbol design, users can quickly understand the distribution and flow trend of people in the fault area.

[0133] Adding interactive functions: Adding interactive functions to dynamic visualization to improve user experience and data analysis efficiency. For example, in the Web visualization interface, add a mouse hover prompt function. When the user hovers the mouse over a certain personnel flow data point or area, the detailed personnel flow information of the location is displayed, such as the specific number of people, time, etc. At the same time, you can add zoom and pan functions, so that users can zoom in or out of the fault impact range analysis diagram as needed to view the personnel flow in different areas; through panning operations, you can view the entire fault area. With the addition of these interactive functions, users can more flexibly analyze and understand the relationship between personnel flow data and the fault impact range.

[0134] A rail transit carbon emission management method also includes the following steps in parallel with superimposing and displaying the personnel flow situation in real time in the fault impact range analysis diagram: Step Se00, obtaining personnel behavior status data and environmental factor data in the location area of ​​the abnormal location, wherein the personnel behavior status data includes movement posture, residence time and aggregation pattern, and the environmental factors include temperature, humidity, air flow velocity and carbon dioxide concentration.

[0135] Action and posture recognition: Relying on advanced computer vision technology, real-time analysis of videos captured by surveillance cameras in abnormal location areas is performed. Using the convolutional neural network (CNN) algorithm based on deep learning, it can accurately extract human motion features by learning from a large number of human motion samples. For example, for walking movements, the algorithm captures the motion trajectory of human joints, the amplitude of limb swings, and other features. When analyzing new video streams, it can quickly determine whether a person is walking and further calculate the walking speed. Taking the surveillance video analysis of a subway station platform as an example, by tracking the coordinate changes of the human hip, knee, and ankle joints in the video, combined with the video frame rate, the displacement of the person per second is calculated, thereby obtaining an accurate walking speed value.

[0136] Stay time monitoring: With the help of video tracking technology, the system continuously tracks each person target. Set a time threshold, such as 30 seconds. When the length of time a person stays at a certain location exceeds the threshold, the system automatically records their stay time. For example, at the door of a store in the station hall, the video surveillance system detected that a passenger stayed for 5 minutes, and this data will be accurately recorded. To ensure the accuracy of the stay time calculation, the system will judge the slight movement of the person. If the moving distance is less than a certain threshold (such as 0.5 meters), it will still be considered as a stay state to avoid incorrect calculation of the stay time due to slight adjustments made by the person.

[0137] Gathering pattern analysis: Set a specific area range. When a certain number of people (such as more than 5 people) appear in the area at the same time, it is determined to be a gathering of people. By analyzing the time when people enter and leave the area, as well as the relative position relationship between people, the closeness and duration of the gathering can be further determined. For example, in a narrow section of the transfer channel, the system detected that 8 people gathered in a short period of time, and the distance between people was small, and the duration was 2 minutes. These data will be recorded in detail for subsequent analysis of the impact of gathering of people on the environment and carbon emissions.

[0138] Step Sf00, synchronously integrating personnel behavior status data and environmental factor data according to time and space dimensions.

[0139] After obtaining personnel behavior and environmental data, the two need to be integrated in time and space dimensions to lay a solid foundation for subsequent carbon emissions analysis.

[0140] Time dimension integration, as follows: 1. Timestamp calibration: Check the timestamp format of personnel behavior and environmental data, and unify them into a standard format, such as "YYYY-MM-DDHH:MM:SS". For the timestamps of data from different sources, round them to the nearest second to eliminate millisecond differences and ensure that the data time points are consistent. 2. Data completion: Due to the different collection frequencies of the two types of data, data missing may occur. For missing values, interpolation is used to complete them. For example, for ambient temperature data, linear interpolation is performed based on the temperature at adjacent moments; for statistical data such as the number of people gathered, the data is completed by weighted average calculation based on the previous and next time points. 3. Sequence construction: Sort the two types of data after timestamp unification and data completion by time, and construct a time series data set. Using time as the index, the personnel behavior (such as the number of people gathered, the length of stay) corresponding to each moment is combined with environmental data (such as temperature, humidity, and carbon dioxide concentration) to facilitate the analysis of data associations at different moments.

[0141] The spatial dimension integration is as follows: 1. Location matching: clarify the geographical location of personnel behavior and environmental data. Personnel behavior data is located through monitoring equipment and Bluetooth positioning; environmental data is located according to the sensor installation point. Accurately match two types of data in the same area, such as personnel monitoring and temperature and humidity sensor data in area A of the platform. 2. Range division: For data that cannot be accurately located, divide the spatial range. For example, in a large station hall, sub-areas are divided according to the area, and the personnel behavior and environmental data are matched to the corresponding sub-areas in combination with the monitoring screen and sensor coverage. 3. Fusion structure: Create a fused spatial data structure, using the GIS feature class concept to organize personnel behavior and environmental data according to spatial location. Set corresponding personnel behavior and environmental attribute fields for each spatial location (point or polygon) to achieve synchronous integration of spatial dimensions.

[0142] Step Sg00, adapting a personalized carbon emission coefficient for each individual according to the identified different human motion postures.

[0143] The analysis of the correlation between motion posture and carbon emissions is as follows: 1. Refine the motion classification: Based on the personnel behavior data, subdivide the motion posture, such as standing is subdivided into static standing, standing with luggage, etc.; walking is divided into slow, normal, fast walking and walking with luggage carts according to speed and whether the cart is pushed. Use image recognition and motion analysis algorithms, such as the OpenPose algorithm, to accurately identify postures based on the movement of human joints. 2. Explore the correlation between carbon emissions: Determine the energy consumption and carbon emissions of each posture through experimental measurement, theoretical derivation and reference research results. For example, the laboratory simulates different walking speeds and measures carbon dioxide emissions. At the same time, consider the impact of airflow, temperature and humidity in rail transit on energy consumption, and establish a relationship model between environmental factors and carbon emission correction coefficients.

[0144] Adapt personalized carbon emission coefficients, as follows: 1. Individual action recognition and tracking: Use multi-camera fusion and human body recognition algorithms to monitor individual actions in abnormal areas in all directions. Give each individual a unique identifier, build an action information database, and record their posture, position, and duration at different times. 2. Use adaptation algorithms: Design algorithms based on the results of action classification and carbon emission association. The algorithm matches the corresponding coefficients from the carbon emission coefficient database according to the individual's real-time actions, and adjusts the coefficients considering the individual's physical characteristics.

[0145] Step Sh00, input the environmental factor data into the dynamic correction factor analysis model of the carbon emission coefficient, output the dynamic correction factor, and adjust the carbon emission coefficient.

[0146] The operation of the dynamic correction factor analysis model for carbon emission coefficient involves the following: 1. Model operation: input the sorted environmental data into the dynamic correction factor analysis model for carbon emission coefficient. The model is built based on a large amount of historical data and research results on the relationship between environment and carbon emissions, and is operated internally through complex algorithms and logic. For example, the model may use a machine learning algorithm to analyze the association between the input environmental data and the adjustment of the carbon emission coefficient based on their characteristics. 2. Output correction factor: After model operation, one or more dynamic correction factors are output. These factors reflect the adjustment range of the basic carbon emission coefficient by the current environmental factors. For example, in a high temperature and low airflow velocity environment, the correction factor output by the model may increase the basic carbon emission coefficient by 10%, indicating that the carbon emissions of human activities in this environment are relatively increased.

[0147] The adjustment of the carbon emission coefficient involves the following process: 1. Coefficient update: Apply the dynamic correction factor output by the model to the personalized carbon emission coefficient adapted for each individual in step Sg00. Adjustment is made through multiplication or other established operation rules. For example, the original carbon emission coefficient of the individual is 0.5, the correction factor is 1.1, and the adjusted carbon emission coefficient becomes 0.5×1.1=0.55. 2. Real-time adjustment mechanism: As environmental factors change in real time, the above steps are repeated continuously to ensure that the carbon emission coefficient can be updated in time according to the latest environmental conditions, providing support for the subsequent accurate calculation of the total carbon emissions of personnel flow.

[0148] Step Si00, multiply the personalized carbon emission coefficients of all personnel in the location area by their corresponding residence time, and then add them up to obtain the real-time total carbon emission of personnel flow in the location area.

[0149] Data preparation: 1. Confirm individual information and coefficients: Sort out all monitored individual information in the abnormal location area, including the unique identifier assigned to each individual and the personalized and dynamically adjusted carbon emission coefficient determined in steps Sg00 and Sh00. For example, in a specific area of ​​a subway station, the system records the information of 100 individuals, each of which has a corresponding unique ID, such as P001, P002, etc., and its latest carbon emission coefficient, such as the carbon emission coefficient of individual P001 is 0.6, and the carbon emission coefficient of P002 is 0.55.

[0150] Extract residence time data: Accurately extract the residence time of each individual in the location area from the personnel behavior status data. These residence time data record the duration of the individual's activities in the area, which is indispensable for calculating the total carbon emissions. For example, individual P001 stayed in the area for 10 minutes, and P002 stayed for 8 minutes. Ensure the accuracy and completeness of the residence time data to avoid calculation deviations due to missing or incorrect data.

[0151] The calculation process is as follows: 1. Calculation of individual carbon emissions: For each individual, multiply its personalized carbon emission coefficient by the corresponding residence time to obtain the individual's carbon emissions in the location area. Use the formula: individual carbon emissions = carbon emission coefficient × residence time. For example, the carbon emission coefficient of individual P001 is 0.6, and the residence time is 10 minutes, so its carbon emissions are 0.6×10=6 (unit: kilograms of carbon dioxide equivalent, assuming that the carbon emission coefficient unit is kilograms of carbon dioxide equivalent / minute, the same below); the carbon emission coefficient of individual P002 is 0.55, and the residence time is 8 minutes, so its carbon emissions are 0.55×8=4.4.

[0152] 2. Total cumulative sum: The carbon emissions of all individuals in the location area are cumulatively summed to obtain the total carbon emissions of the real-time personnel flow in the location area. Use the formula: Total carbon emissions of personnel flow = Σ (individual carbon emission coefficient × individual stay time). For example, add the carbon emissions of the above 100 individuals one by one. Assuming that after calculation, the total carbon emissions of the first 99 individuals are 300, plus the carbon emissions of individual P100 5, the final total carbon emissions of the real-time personnel flow in the location area is 300+5=305 kg of carbon dioxide equivalent.

[0153] Step Sj00: Displaying personnel flow carbon emission data in a preset visualization manner on the fault impact range analysis diagram.

[0154] Select appropriate visualization tools based on the format and characteristics of the fault impact range analysis diagram. If the analysis diagram is generated based on geographic information system (GIS) software, you can continue to use the visualization functions of the software, such as ArcGIS symbolization, annotation and other tools; if the analysis diagram is displayed in the form of a web page, you can use JavaScript visualization libraries, such as D3.js, Echarts, etc.

[0155] Determine the visualization technology to be used, such as using heat maps to show the carbon emission intensity of different regions, using bar charts to show the changes in total carbon emissions in different time periods, and using scatter plots to show the relationship between per capita carbon emissions and population density. According to the characteristics of the data and the purpose of the display, choose the visualization technology that can best convey the information.

[0156] Further consideration is given to the fact that when the total carbon emissions from personnel flow in a certain area rise sharply in a short period of time and exceed the preset high carbon emission threshold, personnel dispatch is triggered. Carbon emission control also needs to be considered during personnel dispatch. The specific considerations are as follows: Step Sk00, analyzing whether the total carbon emissions from personnel movement in the location area within a preset time exceeds the preset total carbon emissions.

[0157] Step SL00, if no, maintain the original settings.

[0158] If the total carbon emissions from the flow of people in the location area does not exceed the preset total carbon emissions within the preset time, it indicates that the current carbon emissions situation is within a relatively controllable range. At this time, the original personnel scheduling and resource allocation settings are maintained, and regular monitoring of carbon emissions continues.

[0159] Step Sm00, if yes, then with the help of high-precision personnel flow sensors and positioning systems, determine the number of people in different areas, personnel flow trends, and the urgency of personnel's safety guarantees (such as lighting needs, ventilation needs), and obtain the existing carbon emission data of each area and the energy consumption data of various emergency resources (lighting equipment, ventilation equipment, etc.) through carbon emission monitoring equipment and energy consumption monitoring devices.

[0160] Step Sn00, construct a matrix with regions as rows and emergency resources as columns. Each element in the matrix represents the benefit value generated by allocating a certain type of resource to a specific region.

[0161] Each row of this matrix represents a specific area, and each column represents an emergency resource, such as lighting equipment, ventilation equipment, etc.

[0162] Quantification of benefit value: Each element in the matrix represents the benefit value generated by allocating a certain type of resource to a specific area. When quantifying the benefit value, three key factors are considered: the degree of satisfaction of personnel needs, the increase in carbon emissions, and energy consumption.

[0163] Among them, the quantitative benefit values ​​are as follows: for the degree of satisfaction of personnel needs, it can be quantified based on factors such as the proportion of the number of personnel in the area to the total number of personnel, the urgency of the needs, etc.; for the increase in carbon emissions, directly refer to the carbon emission data when the equipment is running; energy consumption can also be quantified based on the equipment energy consumption monitoring data. By setting different weights, these three factors are comprehensively calculated to obtain the benefit value of each element. For example, the weight of the degree of satisfaction of personnel needs is set to 0.5, the weight of the increase in carbon emissions is set to 0.3, and the weight of energy consumption is set to 0.2, which is calculated by the formula (benefit value = 0.5 × score of satisfaction of personnel needs + 0.3 × score of increase in carbon emissions + 0.2 × score of energy consumption).

[0164] Step So00, input the constructed resource allocation matrix into the Hungarian algorithm model. The Hungarian algorithm determines the optimal resource allocation solution by continuously searching for independent zero elements in the matrix (ie, a combination in which each row and column has only one zero element).

[0165] During this process, the algorithm will perform a series of transformation operations on the matrix, such as row transformation, column transformation, etc., to increase the number of zero elements and find the optimal match.

[0166] After the algorithm runs, an optimal resource allocation plan will be obtained. This plan clearly defines which areas to allocate various emergency resources to while meeting the safety needs of personnel and controlling carbon emissions to the greatest extent. For example, the results may show that allocating a number of low-energy, low-carbon emission ventilation equipment to waiting areas where people are relatively concentrated and carbon emissions are high can not only meet the ventilation needs of personnel, but also effectively control carbon emissions; allocating high-brightness, medium-energy consumption lighting equipment to key evacuation channels can ensure the safety of personnel evacuation while minimizing unnecessary carbon emissions.

[0167] Step Sp00, sending the dispatching task information to the relevant personnel according to the result obtained by the Hungarian algorithm.

[0168] This information records in detail the types and quantities of emergency resources that need to be deployed, the destination area of ​​deployment, and the time requirements for deployment, etc. Then, this deployment task information is sent to relevant personnel, such as equipment maintenance personnel, dispatchers, etc.

[0169] After receiving the dispatch task information, the relevant personnel will immediately organize actions. They will quickly take out the corresponding emergency resources from the equipment storage area and transport them to the designated area according to the specified route and time requirements. After arriving at the destination, they will accurately install and debug the equipment according to the equipment installation requirements and debugging standards to ensure that the equipment can operate normally, thereby achieving effective control of personnel safety and carbon emissions.

[0170] Based on the same inventive concept, an embodiment of the present invention provides a rail transit carbon emission management system, including a memory and a processor, wherein the memory stores a program that can be executed on the processor to implement the following Figures 1 to 3 Procedure for either method.

[0171] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A rail transit carbon emission management method, characterized in that: include: Acquire multi-source heterogeneous data related to carbon emissions and perform data preprocessing; Extract carbon emission time series data from the multi-source heterogeneous data related to carbon emission that have completed data preprocessing, use the preset seasonal decomposition method to remove the influence of cyclical factors, and decompose the time series into trend terms, seasonal terms, and residual terms; Extract preset deep features based on the decomposed trend term, seasonal term and residual term time series information; The extracted deep features are input into the pre-built ARIMA model to predict the numerical range of carbon emission trend items, seasonal items and residual items within a preset time range in the future as the ARIMA model prediction value; Compare the ARIMA model predictions with real-time carbon emission data and calculate the prediction deviations of each component, including carbon emission trend term, seasonal term and residual term; The prediction deviation of each component is input into the pre-built carbon emission early warning analysis model to output the abnormal risk level of carbon emission and the abnormal cause. The abnormal risk level of carbon emission includes low risk level, medium risk level and high risk level. If the risk level of this carbon emission anomaly is low, the visualization interface will display the current overall carbon emission value, the deviation from the predicted value, and the trend arrow indicating the trend of carbon emissions within the preset time range in the future; If the risk level of this carbon emission anomaly is medium, the visualization interface will display the detailed prediction deviation curve of each component and the dynamic change process of the deviation; If the abnormal carbon emission risk level is high, the visualization interface will display the detailed prediction deviation curve of each component and the dynamic change process of the deviation, and at the same time warn of the high risk level and the cause of the abnormality.

2. A rail transit carbon emission management method according to claim 1, characterized in that: If the risk level of this abnormal carbon emission is medium or high, it also includes: Obtaining environmental monitoring satellite data, wherein the environmental monitoring satellite data includes regional meteorological dynamics and geographic information; Extract feature vectors related to rail transit carbon emissions from environmental monitoring satellite data, use the feature vectors related to rail transit carbon emissions as input objects, input them into a pre-built comprehensive model that simulates carbon emission changes under different emergency scenarios, and output simulation results; Based on the output simulation results, a curve of rail transit carbon emissions changing over time within a preset time range in the future is drawn, and relevant information of the emergency scenario is marked.

3. A rail transit carbon emission management method according to claim 2, characterized in that: From the environmental monitoring satellite data, the feature vectors related to rail transit carbon emissions are extracted, including: Obtain the abnormal risk level of carbon emissions this time; Determine the real-time satellite data feature fluctuation information based on the environmental monitoring satellite data analysis, input the carbon emission abnormal risk level and the real-time satellite data feature fluctuation information into the pre-trained feature selection standard analysis model, and output the feature selection standard; From the environmental monitoring satellite data, feature vectors related to rail transit carbon emissions are extracted according to the feature selection criteria.

4. A rail transit carbon emission management method according to claim 2, characterized in that: The characteristic vectors related to rail transit carbon emissions are used as input objects and input into a pre-built comprehensive model that simulates carbon emission changes under different emergency scenarios. The simulation results include: Analyze whether to obtain the three-dimensional terrain model information scanned by the laser radar device deployed on the top of the train or at key points along the line, as well as the environmental parameters detected by the dust sensor network deployed along the rail transit line and stations; If not, the characteristic vector related to rail transit carbon emissions is used as the input object and input into the pre-built comprehensive model that simulates the changes in carbon emissions under different emergency scenarios, and the simulation results are output; If yes, extracting the laser radar feature vector from the acquired three-dimensional terrain model information, and calibrating and integrating the acquired environmental parameters and extracting the smart dust sensor network feature vector; The laser radar feature vector, the smart dust sensor network feature vector and the feature vector extracted based on the environmental monitoring satellite data are fused to form a fusion vector; Adjusting the architecture of the comprehensive model according to the feature vector included in the fusion vector and the preset comprehensive model architecture quantitative adjustment method; The fusion vector is input into the comprehensive model that simulates the changes in carbon emissions under different emergency scenarios after the structural adjustment is completed, and the simulation results are output.

5. A rail transit carbon emission management method according to claim 4, characterized in that: At the same time, high risk levels and abnormal reasons are warned, including: Analyze whether abnormal sound information captured by acoustic sensor arrays deployed on train tracks and key equipment is obtained; If not, maintain the original setting; If yes, extract acoustic features from the abnormal sound information, and extract core sound features from the acoustic features through a dimensionality reduction method using principal component analysis; The core sound features and the prediction deviations of each component are input into the abnormal sound diagnosis model, and the abnormality cause is output; Update the abnormal reason of the warning according to the output abnormal reason, and display the high risk level together.

6. A rail transit carbon emission management method according to claim 5, characterized in that: It also includes the following steps in parallel with updating the abnormal reason of the alert according to the output abnormal reason and displaying the high risk level together, as follows: Analyze the objects and abnormal locations associated with the abnormal causes, where the objects include train tracks and key equipment; If the cause of the abnormality is associated with the train track, the three-dimensional terrain model information and environmental parameters of the location area of ​​the abnormal position are obtained, and the track fault impact range is matched from the pre-built track fault impact model according to the three-dimensional terrain model information and environmental parameters of the train track associated with the abnormality cause and the location area of ​​the abnormal position, and a fault impact range analysis diagram is obtained; If the cause of the abnormality is associated with a key device, the three-dimensional terrain model information and environmental parameters of the location area of ​​the abnormal location are obtained, and the three-dimensional terrain model information and environmental parameters of the location area of ​​the abnormal location are associated with the key device and the cause of the abnormality, and the key device fault impact range is obtained from the pre-built fault propagation model, and a fault impact range analysis diagram is obtained; Obtain the flow distribution data of people in the area where the abnormal location is located, and use dynamic visualization technology to overlay and display the flow of people in real time on the fault impact range analysis diagram.

7. A rail transit carbon emission management method according to claim 6, characterized in that: The steps in parallel with real-time overlay display of personnel flow in the fault impact range analysis diagram are as follows: Obtaining personnel behavior status data and environmental factor data in the area where the abnormal location is located, where personnel behavior status data includes movement posture, residence time and gathering pattern, and environmental factors include temperature, humidity, air flow speed and carbon dioxide concentration; Synchronize and integrate personnel behavior status data and environmental factor data according to time and space dimensions; According to the different identified movements and postures of people, a personalized carbon emission coefficient is adapted for each individual; Input the environmental factor data into the dynamic correction factor analysis model of the carbon emission coefficient, output the dynamic correction factor, and adjust the carbon emission coefficient; Multiply the personalized carbon emission coefficients of all personnel in the location area by their corresponding stay time, and then add them up to obtain the total carbon emission of personnel flow in real time in the location area; On the fault impact range analysis diagram, the personnel flow carbon emission data is displayed in a preset visualization method.

8. A rail transit carbon emission management system, characterized in that: It comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a rail transit carbon emission management method as described in any one of claims 1 to 7.

Images