Low-traffic rail transit carbon emission accounting method and system

Abstract

This invention belongs to the field of intelligent transportation carbon emission management technology, and particularly relates to a method and system for carbon emission accounting in low-capacity rail transit. The method includes: establishing a dynamic carbon emission inventory framework for the "integrated depot-station-vehicle" operation phase; constructing a dynamic accounting model based on real-time activity data; introducing vehicle load coefficient, start-stop mode, and regenerative braking energy recovery rate as key correction parameters specific to low-capacity systems; using adaptive Monte Carlo simulation to quantify uncertainties; and achieving second-level carbon emission visualization and early warning through edge computing terminals. This invention focuses on the operation phase of low-capacity systems, incorporating the intermittent energy consumption of integrated depot process equipment and the dynamic start-stop characteristics of station electromechanical equipment based on passenger flow into the real-time accounting boundary. It solves the technical problem of large errors in traditional emission factors caused by frequent start-stops and large load fluctuations of low-capacity vehicles, and can provide real-time data support for refined carbon emission reduction strategies in the low-capacity rail transit industry.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation carbon emission management technology, and particularly relates to a carbon emission accounting method and system for low-capacity rail transit. Specifically, it is a real-time dynamic carbon emission accounting method and edge computing system for the operation phase of low-capacity rail transit (tram, guide rail rubber-tired system, Skybus, suspended monorail, etc.). Background Technology

[0002] With the low-carbon development of the rail transit industry, carbon emission accounting has become a core foundation for achieving carbon reduction targets. Existing rail transit carbon emission accounting technologies are mainly geared towards high-capacity subways and national railway systems, and the accounting focus is mostly on the "whole life cycle," covering the construction, operation, and demolition phases. They lack technical solutions for refined and dynamic management of the operation phase and cannot meet the real-time monitoring needs of daily operation carbon emissions.

[0003] In existing technologies, some methods have attempted to achieve dynamic carbon emission tracking. For example, Chinese patent document CN120688753A discloses a dynamic carbon footprint tracking system and method for railway transportation based on blockchain and edge computing. It includes a data acquisition unit for collecting train operating status, environmental parameters, and traction energy consumption data; an edge computing unit that fuses the multi-source data collected by the data acquisition unit using adaptive Kalman filtering to obtain standardized data, converts the train operating status into carbon emissions based on an instantaneous power model, and introduces a dynamic emission factor to calculate cumulative carbon emissions; and a blockchain network unit that compresses the cumulative carbon emission data to generate a Merkle root hash and stores the carbon emission data on the blockchain. However, it still has significant limitations and cannot be applied to carbon emission accounting for low-capacity rail transit, such as trams, guide rail rubber-tired systems, and Skyrail. The core flaws lie in the fact that its instantaneous power model is based on the steady-state operation assumptions of high-speed trains, assuming that the train speed v(t) changes continuously, making it more suitable for long-distance, high-capacity train scenarios. Low-capacity rail transit has distinct characteristics: average station spacing is less than 800m, vehicles frequently start and stop, and the vehicle load factor is less than 30% during off-peak hours, with drastic load fluctuations. These assumptions are severely inconsistent with the actual operating conditions of low-capacity vehicles, resulting in a significant decrease in the accuracy of carbon emission accounting when this scheme is applied to low-capacity rail transit, failing to meet the needs of refined management. Furthermore, this scheme does not consider the energy consumption and carbon emissions of integrated depots, stations, and other supporting facilities in low-capacity rail transit systems, further limiting its applicability in low-capacity scenarios. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method and system for carbon emission accounting in low-capacity rail transit. This method and system are specifically designed for carbon emission accounting during the operational phase of low-capacity rail transit systems such as trams, guide rail rubber-tired systems, and Skyrails, differing from the static lifecycle accounting of traditional high-capacity subways. The method includes: a) establishing a dynamic carbon emission inventory framework for the operational phase of "integrated depot-station-vehicle"; b) constructing a dynamic accounting model based on real-time activity data; introducing vehicle load coefficients, start-stop modes, and regenerative braking energy recovery rates as key correction parameters specific to low-capacity systems; c) using adaptive Monte Carlo simulation to quantify uncertainties; and d) achieving second-level carbon emission visualization and early warning through edge computing terminals. This invention focuses on the operational phase of low-capacity systems, incorporating the intermittent energy consumption of integrated depot process equipment and the dynamic start-stop characteristics of station electromechanical equipment based on passenger flow into the real-time accounting boundary. This solves the technical problem of large errors in traditional emission factors caused by frequent start-stops and large load fluctuations of low-capacity vehicles, providing real-time data support for refined carbon reduction strategies in the low-capacity rail transit industry.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A method for carbon emission accounting of low-capacity rail transit includes: S1. Establish a dynamic carbon emission inventory framework for the "integrated depot-station-vehicle" operation phase; S2. Based on the carbon emission inventory framework established in step S1, real-time activity data of the integrated depot, stations and vehicles are obtained. A dynamic accounting model is constructed based on the obtained real-time activity data to calculate the total carbon emissions during the operation phase of the integrated depot-station-vehicle. The dynamic accounting model includes a comprehensive depot emission model, a station emission model, and a vehicle operation emission model. The vehicle operation emission model incorporates vehicle load factor, start-stop mode, and regenerative braking energy recovery rate as key correction parameters, taking into account the characteristics of low-capacity systems. S3. Adaptive Monte Carlo simulation is used to quantify the uncertainty of the parameters in the dynamic accounting model constructed in step S2. S4. Based on the carbon emission accounting results obtained from the uncertainty quantification in step S3, the carbon emission visualization and early warning are realized in seconds through the edge computing terminal.

[0006] Preferably, in step S1, the carbon emission accounting boundary during the operation phase is first determined, including three subsystems: integrated depot, station, and operating vehicles. Then, based on the determined carbon emission accounting boundary during the operation phase, a dynamic carbon emission inventory framework for the operation phase of "integrated depot-station-vehicle" is established. Preferably, step S2 specifically includes: S21. Obtain real-time activity data of integrated depots, stations, and vehicle operations; S22. Based on the acquired real-time activity data, construct a comprehensive vehicle emissions model: (1) (2) (3) in, The total carbon emissions from the parking lot are expressed in tCO2 (tons of carbon dioxide). Carbon emissions (tCO2) for continuously operating equipment; Carbon emissions from intermittent operation equipment, tCO2; , The activity levels, in kWh, are for continuously operating equipment and intermittently operating equipment, respectively. The energy carbon emission factor (tCO2 / kWh) for process equipment in continuously operating equipment; The energy carbon emission factor (tCO2 / kWh) for process equipment in intermittent operation equipment; N c N represents the number of categories of continuously operating equipment, dimensionless. i ζ represents the number of intermittent operation equipment categories, dimensionless; m Let m be the intermittent operation coefficient of the m-th type of process equipment, which is dimensionless; Let be the energy efficiency attenuation correction coefficient for the m-th type of process equipment, which is dimensionless; Let be the energy efficiency attenuation correction coefficient for the nth type of process equipment, which is dimensionless; S23. Based on the acquired raw activity data, construct a station emission model: (4) in, For the carbon emissions of the station, tCO2; P j Rated power of electromechanical equipment, kW; E is the carbon emission factor of electricity, tCO2 / kWh; T j h represents the actual running time of the passenger flow triggering strategy; β represents the actual running time of the strategy. j γ is the real-time passenger flow fluctuation coefficient (calculated from data from entrance and exit gates), dimensionless; j The regenerative braking feedback ratio measured at the station substation is dimensionless; θ j This is a dimensionless correction factor for the equipment start-stop frequency. S24. Based on the real-time activity data of the acquired vehicle operation phases, i.e., according to the low-capacity system operation diagram, the vehicle operation phases are subdivided into four operating conditions: traction, coasting, braking, and stopping, and each condition-load correction coefficient is assigned to it. , , , To construct vehicle operating emission models: (5) in, tCO2 represents the carbon emissions from vehicle operation; k represents the k-th operating section of the line (between two adjacent stations); F k Fuel or electricity consumption rate based on onboard energy consumption monitoring, in kWh / km or kg / km; L k The distance is the route length, in km; λ represents the carbon emission factor for the corresponding energy source, tCO2 / kWh; k This is a dimensionless correction coefficient for the combined gradient and curvature of the railway line. The condition type of the k-th running segment is determined in real-time by parsing the running graph: When the train is in a traction acceleration state The traction condition correction factor is adopted. ; When the train is in a coasting, constant-speed state The traction condition correction factor is adopted. ; When the train is braking and decelerating The traction condition correction factor is adopted. ; When the train is stopped at the platform The station stop condition correction factor is adopted. .

[0007] Further preferably, the activity level described in step S22 , The specific acquisition is as follows: a. Establish an equipment ledger, recording equipment number, category, and rated power. Rated energy efficiency Commissioning date; b. Collect the power curve P(t) of the equipment through the smart meter, and collect the start-stop status S(t) of the equipment through the equipment control system, where S(t)=1 indicates operation and S(t)=0 indicates shutdown. c. For continuously operating equipment, activity level A n Calculate using the following formula: (6) Where t1 and t2 are the start and end times of the accounting period, in hours; This represents the actual operating power (kW) of the nth type of continuously operating equipment at time t. This indicates the start / stop status of the nth type of continuously operating equipment at time t; For intermittently operating equipment, activity level A m The calculation formula is as follows: (7) Where K m This refers to the number of operations performed on equipment of type m within the accounting period. The power consumption of the m-th type of equipment during the k-th operation is given by , / kWh, and is obtained from the difference in readings of the smart meter at the start and end of the operation.

[0008] The intermittent operation coefficient ζ m Automatically generated through the following steps: a. Collect historical data on the power curves of integrated parking lot equipment; b. Generate an equipment start-up and shutdown schedule based on the daily work plan, as follows: (8) in To calculate the actual daily equipment operating time, in hours; To calculate the daily planned operating time, h; The energy efficiency attenuation correction coefficient Determined according to the following formula: (9) in, Rated energy efficiency of the equipment (determined by testing after commissioning or major overhaul). Actual energy efficiency (calculated through real-time monitoring of input / output power).

[0009] More preferably, the real-time passenger flow fluctuation coefficient β mentioned in step S23 j : The β j It reflects the degree of deviation between real-time passenger flow and designed passenger flow, calculated using data from entrance and exit gates: (10) in: Real-time passenger flow, people / hour, is calculated from data from the entrance and exit gates: (11) The number of people entering the station at the current moment Reduce the number of people leaving the station In addition to the number of people stranded on the platform at one time , person / h; The service passenger flow for the equipment is calculated in people / hour, obtained from the equipment design documents or BAS system configuration parameters. The θ j To correct for the additional energy consumption (peak starting current, preheating energy consumption) caused by frequent equipment start-ups and shutdowns, the following steps are used to determine the appropriate energy consumption: (12) in: , The number of start-stop cycles (start-up criteria: S(t-1)=0 and S(t)=1, lasting >30s; stop-down criteria: S(t-1)=1 and S(t)=0, lasting >60s). , Additional energy consumption per start-stop cycle (for electric motors: 3-5 times rated power × start-up time); The measured regenerative braking feedback ratio γ at the station substation j The net consumption of traction energy consumption and feedback energy is calculated by measuring the bidirectional power flow of the station substation, with a measurement accuracy of ±1%, and internal losses of the substation are excluded.

[0010] More preferably, the fuel or electricity consumption rate F based on on-board energy consumption monitoring in step S24 k The energy consumption rate per unit mileage in the k-th interval is collected and calculated in real time by the onboard energy consumption monitoring unit (OEMU): (13) in: Let be the total energy consumption of the k-th interval, in kWh, which is obtained by summing the energy consumption of the traction subsystem and the energy consumption of the auxiliary subsystem; The k-th interval mileage, in km, is determined by matching the GNSS / transponder positioning system with the operation map.

[0011] The combined slope-curvature correction coefficient λ of the line k Calculated using the following formula: (14) in: The average gradient of the kth interval (obtained from the railway GIS database), ‰; Minimum curve radius in the k-th interval, m (obtained from the railway GIS database).

[0012] Operating condition-load correction factor (Traction condition) (Coasting condition) (Braking condition) (Station stop conditions) are dynamically updated according to the following formula: (15) (16) (17) (18) in, The measured load of the vehicle-mounted weighing system is in kg. Rated passenger capacity, in kg; This refers to the number of stops within a given section. The total number of intervals, For regenerative braking efficiency, Station dwell time, in minutes; The platform passenger density is measured in people per square meter. 2 ; α1-α4 are the system calibration coefficients, and k1-k4 are the sensitivity indices.

[0013] The regenerative braking efficiency It is calculated using the following formula: (19) in, Feedback energy, measured in kWh, from the bidirectional energy meter on the DC side of the traction inverter; The measured load of the vehicle-mounted weighing system is in kg. , The braking start and stop speed, in m / s, is collected by a speed sensor.

[0014] Preferably, step S3 specifically involves: using adaptive Monte Carlo simulation to perform probability distribution uncertainty analysis on the parameters in step S2, outputting a 95% confidence interval, and dynamically adjusting the probability distribution parameters based on historical data feedback according to the dynamic adjustment rule.

[0015] More preferably, the parameters in the dynamic calculation model include the traction condition correction coefficient ε. t Coasting condition correction factor ε c Braking condition correction factor ε b Correction coefficient ε for station operation conditions s Intermittent operation coefficient ζ m Passenger flow fluctuation coefficient β j Regenerative braking feedback ratio γ j Vehicle energy consumption monitoring data F k Smart meter power data P j .

[0016] Preferably, step S4 specifically comprises: The calculation results are visualized in seconds through edge computing terminals deployed locally in the parking lot / station, and an early warning signal is issued when the carbon emission intensity exceeds the dynamic threshold.

[0017] More preferably, the dynamic threshold is: hourly carbon intensity > 110% of the benchmark value or minute-level carbon intensity > 130% of the benchmark value, or the intermittent operation coefficient ζ. m <0.2, or energy efficiency degradation correction factor , >1.15.

[0018] Preferably, the edge computing terminal in step S4 includes: a locally deployed lightweight computing model (with built-in S22-S24 models and a Monte Carlo engine) that supports offline computing; a two-way data channel with the cloud for updating model parameters; an MQTT / Modbus dual-protocol early warning push interface; and a built-in blockchain notarization module for tamper-proof recording of carbon emission data.

[0019] The present invention also includes a system for implementing the above-described method for carbon emission accounting of low-capacity rail transit, the system comprising: Multi-source heterogeneous data acquisition module: Communicates with SCADA, BAS, smart meters, vehicle energy consumption monitoring units, passenger flow detection systems, equipment control systems, and substation metering systems to acquire real-time activity data, and supports edge-side data cleaning and compression; Distributed edge computing module: Deployed in integrated parking lots, stations and depots, with built-in computing model and lightweight Monte Carlo engine, single-point computing power ≥4 TOPS, latency <2 seconds; 3D visualization module: Displays real-time carbon emission heat maps of parking lots, stations, and vehicles in WebGL format, supporting timeline backtracking and prediction modes; Intelligent early warning module: When hourly carbon intensity > 110% of the benchmark value, or minute-level carbon intensity > 130% of the benchmark value, or the intermittent operation coefficient ζ m <0.2, or energy efficiency degradation correction coefficient , At >1.15, alarm information is pushed via MQTT and emission reduction suggestions are automatically generated.

[0020] Beneficial effects of the present invention Compared with the prior art, the present invention has the following beneficial effects: (1) Significantly improved accounting accuracy: By introducing a multi-dimensional correction system consisting of working condition-load dynamic correction coefficient ε, intermittent operation coefficient ζ, passenger flow fluctuation coefficient β, regenerative braking feedback ratio γ, and equipment start-stop frequency correction coefficient θ, the carbon emission accounting error of low-capacity rail transit is reduced; among them, ε is based on the design of train dynamics equations and traction characteristic curves, distinguishing between four working conditions: traction / coasting / braking / stopping, eliminating the systematic deviation of traditional methods calculated based on rated load; ζ is based on the time series matching algorithm of historical data of equipment power curves and the next day's work plan, quantifying the intermittent operation characteristics, avoiding the overestimation error of traditional methods calculated based on 24-hour continuous operation; β is based on dual-source verification of passenger heat load model and gate data, dynamically linking passenger flow density and station environmental control energy consumption; γ is based on the bidirectional power flow metering device of substation, distinguishing between vehicle-level recycling efficiency and grid-level feedback ratio, realizing accurate measurement of "net carbon emissions"; θ is based on the peak characteristics of motor starting current, incorporating the equivalent energy consumption increment caused by frequent start-stop into the accounting.

[0021] (2) Real-time performance breakthrough: By deploying distributed edge computing nodes (depot / station / vehicle depot), local data preprocessing, lightweight model inference, and result output in seconds are achieved, with a calculation latency of <2 seconds, which is 2-3 orders of magnitude higher than the traditional cloud batch processing mode (latency of 5-10 minutes); (3) Complete accounting boundary: For the first time, the intermittent operation energy consumption of integrated depot process equipment and the dynamic start-stop characteristics of station electromechanical equipment based on passenger flow are included in the real-time accounting boundary, solving the problem that the energy consumption of depot + station accounts for 20-30% of the low-capacity system but is systematically omitted by traditional methods. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the carbon emission accounting system architecture for low-capacity rail transit in Embodiment 2 of the present invention. Detailed Implementation

[0023] Example 1 A method for carbon emission accounting of low-capacity rail transit includes: S1. Establish a dynamic carbon emission inventory framework for the "integrated depot-station-vehicle" operation phase; Specifically, the carbon emission accounting boundaries for the operation phase are first determined, including three subsystems: the integrated depot, stations, and operating vehicles. Then, based on these determined boundaries, a dynamic carbon emission inventory framework for the operation phase, encompassing the integrated depot, stations, and vehicles, is established. This carbon emission inventory framework represents the sum of carbon emissions from the integrated depot, stations, and vehicles. S2. Based on the carbon emission inventory framework established in step S1, real-time activity data of the integrated depot, stations and vehicles are obtained. A dynamic accounting model is constructed based on the obtained real-time activity data to calculate the total carbon emissions during the operation phase of the integrated depot-station-vehicle. The dynamic accounting model includes a comprehensive depot emission model, a station emission model, and a vehicle operation emission model. The vehicle operation emission model incorporates vehicle load factor, start-stop mode, and regenerative braking energy recovery rate as key correction parameters, taking into account the characteristics of low-capacity systems. Specifically: S21. Obtain real-time activity data of integrated depots, stations, and vehicle operations; S22. Based on the acquired real-time activity data, construct a comprehensive vehicle emissions model: (1) (2) (3) in, The total carbon emissions from the parking lot, tCO2; Carbon emissions (tCO2) for continuously operating equipment; Carbon emissions from intermittent operation equipment, tCO2; , The activity levels, in kWh, are for continuously operating equipment and intermittently operating equipment, respectively. The energy carbon emission factor for continuously operating equipment is tCO2 / kWh; Energy carbon emission factor for intermittent operation equipment, tCO2 / kWh; N c N represents the number of categories of continuously operating equipment, dimensionless. i ζ represents the number of intermittent operation equipment categories, dimensionless; m Let m be the intermittent operation coefficient of the m-th type of process equipment, which is dimensionless; Let be the energy efficiency attenuation correction coefficient for the m-th type of process equipment; Let be the energy efficiency attenuation correction coefficient for the nth type of process equipment, which is dimensionless; The activity level , The specific acquisition is as follows: a. Establish an equipment ledger, recording equipment number, category, and rated power. Rated energy efficiency Commissioning date; b. Collect the power curve P(t) of the equipment through the smart meter, and collect the start-stop status S(t) of the equipment through the equipment control system, where S(t)=1 indicates operation and S(t)=0 indicates shutdown. c. For continuously operating equipment, activity level A n Calculate using the following formula: (4) Where t1 and t2 are the start and end times of the accounting period, in hours; This represents the actual operating power (kW) of the nth type of continuously operating equipment at time t. This indicates the start / stop status of the nth type of continuously operating equipment at time t; For intermittently operating equipment, activity level A m The calculation formula is as follows: (5) Where K m This refers to the number of operations performed on equipment of type m within the accounting period. The power consumption of the m-th type of equipment during the k-th operation is given by , / kWh, and is obtained from the difference in readings of the smart meter at the start and end of the operation.

[0024] The intermittent operation coefficient ζ m Automatically generated through the following steps: a. Collect historical data on the power curves of integrated parking lot equipment; b. Generate an equipment start-up and shutdown schedule based on the daily work plan, as follows: (6) in, To calculate the actual daily equipment operating time, in hours; To calculate the planned daily running time, h; The energy efficiency attenuation correction coefficient Determined according to the following formula: (7) in, Rated energy efficiency of the equipment (determined by testing after commissioning or major overhaul). Actual energy efficiency (calculated through real-time monitoring of input / output power); S23. Based on the acquired real-time activity data, construct a station emission model: (8) in, For the carbon emissions of the station, tCO2; P j Rated power of electromechanical equipment, kW; E is the carbon emission factor of electricity, tCO2 / kWh; T j h represents the actual running time of the passenger flow triggering strategy; β represents the actual running time of the strategy. j γ is the real-time passenger flow fluctuation coefficient (calculated from data from entrance and exit gates), dimensionless; j The regenerative braking feedback ratio measured at the station substation is dimensionless; θ j This is a dimensionless correction factor for the equipment start-stop frequency. The T jThe calculations are performed in real time by edge computing nodes: receiving the operating status of BAS system equipment, gate passenger flow data, and sensor data, and accumulating the actual operating time of each device.

[0025] Real-time passenger flow fluctuation coefficient β j : The β j It reflects the degree of deviation between real-time passenger flow and designed passenger flow, calculated using data from entrance and exit gates: (9) in: Real-time passenger flow, people / hour, is calculated from data from the entrance and exit gates: (10) The number of people entering the station at the current moment Reduce the number of people leaving the station In addition to the number of people stranded on the platform at one time , person / h; The service passenger flow for the equipment is calculated in people / hour, obtained from the equipment design documents or BAS system configuration parameters. Equipment start-stop frequency correction factor θ j : The θ j To correct for the additional energy consumption (peak starting current, preheating energy consumption) caused by frequent equipment start-ups and shutdowns, the following steps are used to determine the appropriate energy consumption: (11) in: , The number of start-stop cycles (start-up criteria: S(t-1)=0 and S(t)=1, lasting >30s; stop-down criteria: S(t-1)=1 and S(t)=0, lasting >60s). , Additional energy consumption for a single start-stop cycle (for electric motors: 3-5 times rated power × start-up time).

[0026] The measured regenerative braking feedback ratio γ at the station substation j The net consumption of traction energy consumption and feedback energy is calculated by measuring the bidirectional power flow of the station substation, with a measurement accuracy of ±1%, and internal losses of the substation are excluded.

[0027] S24. Based on the real-time activity data of the acquired vehicle operation phases, i.e., the low-capacity system operation diagram, the vehicle operation phases are subdivided into four operating conditions: traction, coasting, braking, and stopping, and each condition-load correction coefficient is assigned to it. , , , To construct vehicle operating emissions models: (12) in, tCO2 represents the carbon emissions from vehicle operation; k represents the k-th operating section of the line (between two adjacent stations); F k Fuel or electricity consumption rate based on onboard energy consumption monitoring, in kWh / km or kg / km; L k The distance is the route length, in km; λ represents the carbon emission factor for the corresponding energy source, tCO2 / kWh; k This is a dimensionless correction coefficient for the combined gradient and curvature of the railway line. The F k The energy consumption rate per unit mileage in the k-th interval is collected and calculated in real time by the onboard energy consumption monitoring unit (OEMU): (13) in: The total energy consumption in the k-th interval is kWh, which is obtained by the sum of the energy consumption of the traction subsystem and the energy consumption of the auxiliary subsystem. The energy consumption of the traction subsystem is collected by the DC side energy meter of the traction inverter at a sampling frequency of 100Hz. The energy consumption of the auxiliary subsystem is collected by the energy meter at the output end of the auxiliary inverter at a sampling frequency of 10Hz. The k-th interval mileage, in km, is determined by matching the GNSS / transponder positioning system with the operation map.

[0028] The combined slope-curvature correction coefficient λ of the line k Calculated using the following formula: (14) in: Average slope of the kth interval, ‰ (obtained from the route GIS database); Minimum curve radius in the k-th interval, m (obtained from the railway GIS database).

[0029] The condition type of the k-th running segment is determined in real-time by parsing the running graph: When the train is in a traction acceleration state The traction condition correction factor is adopted. ; When the train is in a coasting, constant-speed state The traction condition correction factor is adopted. ; When the train is braking and decelerating The traction condition correction factor is adopted. ; When the train is stopped at the platform The station stop condition correction factor is adopted. ; The working condition-load correction factor (Traction condition) (Coasting condition) (Braking condition) (Station stop conditions) are dynamically updated according to the following formula: (15) (16) (17) (18) in, The measured load of the vehicle-mounted weighing system is in kg. Rated passenger capacity, in kg; This refers to the number of stops within a given section. The total number of intervals, To determine the regenerative braking efficiency, the feedback energy is measured by a bidirectional energy meter in the DC link of the traction inverter, and the ratio of this feedback energy to the reduction in braking kinetic energy calculated by the on-board weighing system and speed sensor is used to determine the efficiency. Station dwell time, in minutes. The platform passenger density is measured in people per square meter. 2 , - These are the system calibration coefficients. - This is the sensitivity index.

[0030] The regenerative braking efficiency It is calculated using the following formula: (19) in, Feedback energy, measured in kWh, from the bidirectional energy meter on the DC side of the traction inverter; The measured load of the vehicle-mounted weighing system is in kg. , The braking start and stop speed, in m / s, is collected by a speed sensor.

[0031] S3. Adaptive Monte Carlo simulation is used to quantify the uncertainty of the parameters in the dynamic accounting model constructed in step S2. Specifically, adaptive Monte Carlo simulation is used to perform probability distribution uncertainty analysis on the parameters in step S2, outputting a 95% confidence interval, and the probability distribution parameters are dynamically adjusted based on historical data feedback according to the dynamic adjustment rule. The parameters in the dynamic calculation model include the traction condition correction coefficient ε. t Coasting condition correction factor ε cBraking condition correction factor ε b Correction coefficient ε for station operation conditions s Intermittent operation coefficient ζ mi Passenger flow fluctuation coefficient β j Regenerative braking feedback ratio γ j Vehicle energy consumption monitoring data F k Smart meter power data P j ; The traction condition correction coefficient ε t The normal distribution is adopted, with an initial mean μ=1.20 and an initial standard deviation σ=0.096. The mean is dynamically updated based on the real-time OD matrix. The coasting condition correction coefficient ε c The data are normally distributed with an initial mean μ = 0.95 and an initial standard deviation σ = 0.057. The braking condition correction coefficient ε b The data follows a normal distribution with an initial mean μ = 0.80 and an initial standard deviation σ = 0.064, and is affected by regenerative braking efficiency. Influence; The stop condition correction coefficient ε s Using a normal distribution, the initial mean is... (Initial standard deviation σ = 0.15μ). The intermittent operation coefficient ζ i The Beta distribution is used, with initial α=2, β=3, and support interval [0.2, 1.0]. The passenger flow fluctuation coefficient β j The normal distribution is adopted, μ=1.0, σ=0.15, and it is dynamically updated based on the gate data; Regenerative braking feedback ratio γ j The Beta distribution is used, with initial α=3, β=2, and support interval [0, 0.3]. Vehicle energy consumption monitoring data F k The measurement follows a normal distribution with an error of ±2%. Smart meter power data P j It adopts a normal distribution, with an accuracy class of 0.5S and σ = 0.5% × range; The probability distribution parameters include: Distribution type parameter: Normal distribution / Beta distribution; Location parameter: mean μ (normal / lognormal); Scale parameter: Standard deviation σ (normal / lognormal); The dynamic adjustment rule is as follows: based on the deviation between the historical data of the most recent 30 days and the measured values ​​(the deviation refers to the difference between the carbon emission accounting prediction value output by Monte Carlo simulation and the actual measured carbon emission value collected by the edge computing terminal), μ and σ are updated using an exponential weighted moving average (EWMA), with an adjustment period of 7 days.

[0032] S4. Based on the carbon emission accounting results obtained from the uncertainty quantification in step S3, the carbon emission visualization and early warning are realized in seconds through the edge computing terminal.

[0033] Step S4 specifically involves: The calculation results are visualized in seconds through edge computing terminals deployed locally in the parking lot / station, and an early warning signal is issued when the carbon emission intensity exceeds the dynamic threshold.

[0034] The dynamic threshold is: hourly carbon intensity > 110% of the benchmark value or minute-level carbon intensity > 130% of the benchmark value, or the intermittent operation coefficient ζ. m <0.2, or energy efficiency degradation correction factor , >1.15.

[0035] The edge computing terminal described in step S4 includes: a locally deployed lightweight computing model (with built-in S22-S24 models and a Monte Carlo engine) that supports offline computation; a two-way data channel with the cloud for updating model parameters; an MQTT / Modbus dual-protocol early warning push interface; and a built-in blockchain storage module for tamper-proof recording of carbon emission data.

[0036] Example 2 This embodiment provides a system for implementing the above-described carbon emission accounting method for low-capacity rail transit, the system comprising: Multi-source heterogeneous data acquisition module: Communicates with SCADA, BAS, smart meters, vehicle energy consumption monitoring units, passenger flow detection systems, equipment control systems, and substation metering systems to acquire real-time activity data, and supports edge-side data cleaning and compression; Distributed edge computing module: Deployed in integrated parking lots, stations and depots, with built-in computing model and lightweight Monte Carlo engine, single-point computing power ≥4 TOPS, latency <2 seconds; 3D visualization module: Displays real-time carbon emission heat maps of parking lots, stations, and vehicles in WebGL format, supporting timeline backtracking and prediction modes; Intelligent early warning module: When hourly carbon intensity > 110% of the benchmark value, or minute-level carbon intensity > 130% of the benchmark value, or the intermittent operation coefficient ζ m <0.2, or energy efficiency degradation correction coefficient , At >1.15, alarm information is pushed via MQTT and emission reduction suggestions are automatically generated.

[0037] The low-capacity rail transit carbon emission accounting system of this invention adopts a three-layer architecture of "cloud-edge-device", such as... Figure 1 As shown, it includes: 1. End layer (multi-source heterogeneous data acquisition layer) This layer is the system's bottom-level sensing unit, deployed across the entire scenario of integrated parking lots, stations, and depots. It completes full-dimensional data collection through multi-source heterogeneous sensing devices, providing raw data input for upper-layer edge nodes. Integrated vehicle depot: Deploy SCADA system, smart meters, and power analyzers to collect power data from substations and energy consumption data from process equipment such as car wash machines, maintenance equipment, and lighting systems; At the station: The system collects operating data of mechanical and electrical equipment such as ventilation, air conditioning, escalators, and lighting through the BAS system; real-time passenger flow data is collected through the turnstile system; and regenerative braking feedback data is collected through the two-way power metering device in the substation. Vehicle side: Real-time energy consumption data of traction system and auxiliary system are collected through vehicle energy consumption monitoring unit, and passenger load data are collected through vehicle weighing system.

[0038] 2. Edge layer (edge ​​computing layer) Distributed edge computing nodes are deployed in the integrated depot, stations, and train depots. Each node has a built-in lightweight carbon emission accounting model and Monte Carlo simulation engine, with a single-point computing power of ≥4 TOPS and a latency of <2 seconds. The edge nodes have local data cleaning, compression, and accounting capabilities and support offline operation.

[0039] 3. Cloud layer (cloud management layer) The system deploys model training servers and data storage centers, communicating with edge nodes via bidirectional data channels to enable model parameter updates, historical data archiving, and global optimization analysis. This includes a railway GIS database.

Claims

1. A method for carbon emission accounting of low-capacity rail transit, characterized in that, Includes the following steps: S1. Establish a dynamic carbon emission inventory framework for the "integrated depot-station-vehicle" operation phase; S2. Based on the carbon emission inventory framework established in step S1, real-time activity data of the integrated depot, stations and vehicles are obtained. A dynamic accounting model is constructed based on the obtained real-time activity data to calculate the total carbon emissions during the operation phase of the integrated depot-station-vehicle. The dynamic accounting model includes a comprehensive depot emission model, a station emission model, and a vehicle operation emission model. The vehicle operation emission model incorporates vehicle load factor, start-stop mode, and regenerative braking energy recovery rate as key correction parameters, taking into account the characteristics of low-capacity systems. S3. Adaptive Monte Carlo simulation is used to quantify the uncertainty of the parameters in the dynamic accounting model constructed in step S2. S4. Based on the carbon emission accounting results obtained from the uncertainty quantification in step S3, the carbon emission visualization and early warning are realized in seconds through the edge computing terminal.

2. The carbon emission accounting method for low-capacity rail transit according to claim 1, characterized in that, Step S2 is as follows: S21. Obtain real-time activity data of integrated depots, stations, and vehicle operations; S22. Based on the acquired real-time activity data, construct a comprehensive vehicle emissions model: （1） （2） （3） in, The total carbon emissions from the parking lot, tCO2; Carbon emissions (tCO2) for continuously operating equipment; Carbon emissions from intermittent operation equipment, tCO2; , The activity levels, in kWh, are for continuously operating equipment and intermittently operating equipment, respectively. The energy carbon emission factor (tCO2 / kWh) for process equipment in continuously operating equipment; The energy carbon emission factor (tCO2 / kWh) for process equipment in intermittent operation equipment; N c N represents the number of categories of continuously operating equipment, dimensionless. i ζ represents the number of intermittent operation equipment categories, dimensionless; m Let m be the intermittent operation coefficient of the m-th type of process equipment, which is dimensionless; Let be the energy efficiency attenuation correction coefficient for the m-th type of process equipment, which is dimensionless; Let be the energy efficiency attenuation correction coefficient for the nth type of process equipment, which is dimensionless; S23. Based on the acquired raw activity data, construct a station emission model: （4） in, For the carbon emissions of the station, tCO2; P j Rated power of electromechanical equipment, kW; E is the carbon emission factor of electricity, tCO2 / kWh; T j h represents the actual running time of the passenger flow triggering strategy; β represents the actual running time of the strategy. j γ is the real-time passenger flow fluctuation coefficient, dimensionless; j The regenerative braking feedback ratio measured at the station substation is dimensionless; θ j This is a dimensionless correction factor for the equipment start-stop frequency. S24. Based on the real-time activity data of the acquired vehicle operation phases, i.e., according to the low-capacity system operation diagram, the vehicle operation phases are subdivided into four operating conditions: traction, coasting, braking, and stopping, and each condition-load correction coefficient is assigned to it. , , , To construct vehicle operating emission models: （5） in, tCO2 represents the carbon emissions from vehicle operation; k represents the k-th operating section of the line (between two adjacent stations); F k Fuel or electricity consumption rate based on onboard energy consumption monitoring, in kWh / km or kg / km; L k The distance is the route length, in km; λ represents the carbon emission factor for the corresponding energy source, tCO2 / kWh; k This is a dimensionless correction coefficient for the combined gradient and curvature of the railway line. When the train is in a traction acceleration state, the traction condition correction coefficient is used. ; When the train is in a coasting, constant-speed state, the traction condition correction factor is used. ; When the train is braking and decelerating, the traction condition correction factor is used. ; When the train is stopped at the platform, the station stop condition correction factor is used. .

3. The carbon emission accounting method for low-capacity rail transit according to claim 2, characterized in that, The activity level described in step S22 , The specific acquisition is as follows: a. Establish an equipment ledger, recording equipment number, category, and rated power. Rated energy efficiency Commissioning date; b. Collect the power curve P(t) of the equipment through the smart meter, and collect the start-stop status S(t) of the equipment through the equipment control system, where S(t)=1 indicates operation and S(t)=0 indicates shutdown. c. For continuously operating equipment, activity level A n Calculate using the following formula: （6） Where t1 and t2 are the start and end times of the accounting period, in hours; This represents the actual operating power (kW) of the nth type of continuously operating equipment at time t. This indicates the start / stop status of the nth type of continuously operating equipment at time t; For intermittently operating equipment, activity level A m The calculation formula is as follows: （7） Where K m This refers to the number of operations performed on equipment of type m within the accounting period. The power consumption of the m-th type of equipment during the k-th operation is given by the value in kWh, which is obtained from the difference in the readings of the smart meter at the start and end of the operation. The intermittent operation coefficient ζ m Automatically generated through the following steps: a. Collect historical data on the power curves of integrated parking lot equipment; b. Generate an equipment start-up and shutdown schedule based on the daily work plan, as follows: （8） in To calculate the actual daily equipment operating time, in hours. To calculate the daily planned operating time, h; The energy efficiency attenuation correction coefficient Determined according to the following formula: （9） in, The rated energy efficiency of the equipment, This refers to actual energy efficiency.

4. The carbon emission accounting method for low-capacity rail transit according to claim 2, characterized in that, Step S23 describes the real-time passenger flow fluctuation coefficient β j : The β j It reflects the degree of deviation between real-time passenger flow and designed passenger flow, calculated using data from entrance and exit gates: （10） in: Real-time passenger flow, people / hour, is calculated from data from the entrance and exit gates: （11） The number of people entering the station at the current moment Reduce the number of people leaving the station In addition to the number of people stranded on the platform at one time , person / h; The service passenger flow for the equipment is calculated in people / hour, obtained from the equipment design documents or BAS system configuration parameters. The θ j To correct the extra energy consumption caused by frequent device start-ups and shutdowns, the following steps are used to determine: （12） in: , For the number of start-stop cycles, , Additional energy consumption for a single start-stop cycle; The measured regenerative braking feedback ratio γ at the station substation j The net consumption of traction energy consumption and feedback energy is calculated by measuring the bidirectional power flow of the station substation, with a measurement accuracy of ±1%, and internal losses of the substation are excluded.

5. The carbon emission accounting method for low-capacity rail transit according to claim 2, characterized in that, Step S24 describes the fuel or electricity consumption rate F based on onboard energy consumption monitoring. k The energy consumption rate per unit mileage in the k-th interval is collected and calculated in real time by the onboard energy consumption monitoring unit (OEMU): （13） in: Let be the total energy consumption of the k-th interval, in kWh, which is obtained by summing the energy consumption of the traction subsystem and the energy consumption of the auxiliary subsystem; The k-th interval mileage, in km, is determined by matching the GNSS / transponder positioning system with the operation map; The combined slope-curvature correction coefficient λ of the line k Calculated using the following formula: （14） in: The average slope of the kth interval, ‰; Minimum curve radius in the k-th interval, m; Operating condition-load correction factor (Traction condition) (Coasting condition) (Braking condition) (Station stop conditions) are dynamically updated according to the following formula: （15） （15） （17） （18） in, The measured load of the vehicle-mounted weighing system is in kg. Rated passenger capacity, kg. This refers to the number of stops within a given section. The total number of intervals, For regenerative braking efficiency, Station dwell time, in minutes. The platform passenger density is measured in people per square meter. 2 α1-α4 are the system calibration coefficients, and k1-k4 are the sensitivity exponents; The regenerative braking efficiency It is calculated using the following formula: （19） in, Feedback energy, measured in kWh, from the bidirectional energy meter on the DC side of the traction inverter; The measured load of the vehicle-mounted weighing system is in kg. , The braking start and stop speed, in m / s, is collected by a speed sensor.

6. The carbon emission accounting method for low-capacity rail transit according to claim 1, characterized in that, Step S3 specifically involves: using adaptive Monte Carlo simulation to perform probability distribution uncertainty analysis on the parameters of the dynamic accounting model in step S2, outputting a 95% confidence interval, and dynamically adjusting the probability distribution parameters based on historical data feedback according to the dynamic adjustment rules.

7. The carbon emission accounting method for low-capacity rail transit according to claim 6, characterized in that, The parameters in the dynamic calculation model include the traction condition correction coefficient ε. t Coasting condition correction factor ε c Braking condition correction factor ε b Correction coefficient ε for station operation conditions s Intermittent operation coefficient ζ m Passenger flow fluctuation coefficient β j Regenerative braking feedback ratio γ j Vehicle energy consumption monitoring data F k Smart meter power data P j .

8. The carbon emission accounting method for low-capacity rail transit according to claim 1, characterized in that, Step S4 specifically involves: The calculation results are visualized in seconds via edge computing terminals deployed locally in the depot / station, and an early warning signal is issued when the carbon emission intensity exceeds a dynamic threshold; preferably, the dynamic threshold is: hourly carbon intensity > 110% of the benchmark value or minute-level carbon intensity > 130% of the benchmark value, or intermittent operation coefficient ζ. m <0.2, or energy efficiency degradation correction factor , >1.

15.

9. A system for implementing the carbon emission accounting method for low-capacity rail transit according to any one of claims 1-8, characterized in that, The system includes: Multi-source heterogeneous data acquisition module: Communicates with SCADA, BAS, smart meters, vehicle energy consumption monitoring units, passenger flow detection systems, equipment control systems, and substation metering systems to acquire real-time activity data, and supports edge-side data cleaning and compression; Distributed edge computing module: Deployed in integrated parking lots, stations and depots, with built-in computing model and lightweight Monte Carlo engine, single-point computing power ≥4 TOPS, latency <2 seconds; 3D visualization module: Displays real-time carbon emission heat maps of parking lots, stations, and vehicles in WebGL format, supporting timeline backtracking and prediction modes; Intelligent early warning module: When hourly carbon intensity > 110% of the benchmark value, or minute-level carbon intensity > 130% of the benchmark value, or the intermittent operation coefficient ζ m <0.2, or energy efficiency degradation correction coefficient , At >1.15, alarm information is pushed via MQTT and emission reduction suggestions are automatically generated.

Images