Smart rail transit network collaborative carbon reduction operation optimization method, system and device

Abstract

The application discloses a kind of wisdom rail line network synergic carbon reduction operation optimization method, system and equipment, it is related to urban public transport energy-saving scheduling and low-carbon operation technical field, it is based on the dynamics characteristics of wisdom rail vehicle and power grid carbon emission factor to construct the quantification model of fine section level energy consumption and carbon emission, utilize deep learning network to fuse multi-source data to carry out cross-line short-time passenger flow prediction, and real-time passenger flow demand, energy consumption carbon emission index and intersection signal phase time window constraint are unified into multi-objective collaborative optimization framework, by the mixed integer linear programming solution under rolling horizon, the joint optimization and dynamic closed-loop control of dispatch interval, section operating speed and road right strategy are realized, so that on the premise of guaranteeing passenger flow transport demand and service quality, the total energy consumption and carbon emission of wisdom rail line network are significantly reduced, and the energy-saving and carbon-reducing operation goal of line network level is achieved.

Description

Technical Field

[0001] This invention relates to the field of energy-saving scheduling and low-carbon operation technology for urban public transportation, specifically to an operation optimization method, system and equipment for collaborative carbon reduction of intelligent rail transit networks, which can be used for energy consumption calculation, carbon emission quantification, passenger flow prediction, timetable optimization and real-time scheduling control of urban-level intelligent rail transit networks. Background Technology

[0002] The statements in this section are provided only as background information in connection with this disclosure and may not constitute prior art.

[0003] With the rapid development of new energy vehicles and smart urban transportation technologies, the Autonomous Rail Rapid Transit (ART) system, a medium-to-low capacity rail transit system that combines the advantages of high capacity rail transit and high flexibility of surface public transport, has been deployed in multiple cities. ART trains employ technologies such as virtual track following, rubber-tired guidance, and multi-axle steering, possessing quasi-rail transit service capabilities and serving as an important vehicle for achieving a green and low-carbon transformation of urban public transportation.

[0004] In the existing intelligent rail transit operation and management system, although a dispatch center and intersection signal priority mechanism have been initially established, its core dispatch logic still largely follows the traditional ground public transport or urban rail transit model, mainly focusing on punctuality and capacity matching. However, there are still many technical bottlenecks in "network-level collaborative carbon reduction" and "refined energy consumption control," specifically manifested as follows:

[0005] There is a lack of energy consumption prediction and optimization capabilities based on vehicle dynamics. Existing dispatching systems primarily manage energy consumption at the "post-event statistics" stage, failing to perform "pre-event prediction" based on factors such as track gradient, vehicle load, and dynamic characteristics (e.g., rolling resistance, air resistance). This results in dispatching centers being unable to anticipate energy consumption differences under different dispatching schemes when creating operation schedules, making it difficult to support operation schedule optimization based on minimizing energy consumption.

[0006] Passenger flow forecasting is disconnected from capacity scheduling, lacking a dynamic response mechanism. Current passenger flow forecasting models often operate independently, without deep coupling with timetable creation. Static timetables cannot proactively adapt to spatiotemporal fluctuations in passenger flow, easily leading to insufficient capacity during peak hours causing congestion, and excessive capacity during off-peak hours resulting in wasted empty runs. Furthermore, when actual passenger flow or traffic conditions change abruptly, the lack of a rolling optimization mechanism based on real-time data prevents the system from adjusting departure intervals or operating speeds in a timely manner, resulting in low operational efficiency.

[0007] Insufficient coordination between the network and right-of-way signals. In a multi-line network, each line is typically scheduled independently, lacking a unified optimization framework. Furthermore, intersection signal priority strategies are usually based on local logic triggered by vehicle arrival, failing to coordinate with the energy consumption targets of the scheduling layer and the overall timetable. This often leads to vehicles making unnecessary rapid accelerations to catch green lights, or, even after gaining priority, having to stop for extended periods at the next station due to time constraints, resulting in "ineffective acceleration and deceleration" and significantly increasing traction energy consumption.

[0008] There is a lack of a refined system for quantifying and visualizing carbon emissions. Existing systems struggle to drill down carbon emission indicators to the specific "segment" or "individual vehicle" level. Operations managers cannot intuitively grasp the key bottlenecks with high energy consumption and high carbon emissions in the network, making it difficult to form a systematic energy-saving assessment and decision-making loop.

[0009] Therefore, there is an urgent need to develop an intelligent rail transit network collaborative carbon reduction operation optimization system that can integrate vehicle dynamics modeling, short-term passenger flow prediction, signal coordination strategies, and network-level multi-objective optimization into a unified computing framework, so as to achieve synergistic improvement in energy saving and efficiency. Summary of the Invention

[0010] The purpose of this invention is to address the technical problems in current intelligent rail transit (IRT) operation and scheduling systems, such as the lack of energy consumption prediction methods based on vehicle dynamics, the disconnect between passenger flow prediction and timetable compilation, and the failure of intersection signal priority strategies to coordinate with energy-saving goals, resulting in high operational energy consumption and low efficiency. This invention provides an operational optimization method, system, and equipment for collaborative carbon reduction in the IRT network. Based on the dynamic characteristics of IRT vehicles and the carbon emission factors of the power grid, it constructs a refined section-level energy consumption and carbon emission quantification model. It utilizes a deep learning network to fuse multi-source data for short-term passenger flow prediction across lines, and integrates real-time passenger flow demand, energy consumption and carbon emission indicators, and intersection signal phase (SPaT) time window constraints into a unified multi-objective collaborative optimization framework. Through mixed-integer linear programming under a rolling view, it achieves joint optimization and dynamic closed-loop control of departure intervals, section operating speeds, and right-of-way strategies. This significantly reduces the total operational energy consumption and carbon emissions of the IRT network while ensuring passenger transport demand and service quality, achieving the network-level energy-saving and carbon-reducing operational goals.

[0011] The technical solution of the present invention is as follows:

[0012] An operational optimization method for coordinated carbon reduction in intelligent rail transit networks includes:

[0013] Multi-source operational data of the intelligent rail transit network are collected and standardized to form a standardized operational dataset; the multi-source operational data includes vehicle operation data, passenger flow data, intersection signal data, and basic energy consumption data;

[0014] Based on the vehicle dynamics model, the net energy consumption of vehicles in each line section is calculated using the standardized operation dataset, and the carbon emissions of the section are calculated by combining the power grid carbon emission factor, thus establishing a quantitative characterization of network energy consumption and carbon emissions.

[0015] Based on historical passenger flow sequences and real-time passenger flow input, a network-level passenger flow prediction model is constructed to generate prediction results of the spatiotemporal distribution of passenger flow at each station and section in the future.

[0016] A network collaborative optimization model is constructed, which incorporates the net energy consumption of vehicles, carbon emissions of sections, prediction results of passenger flow spatiotemporal distribution, and intersection signal phase information into a unified calculation framework. The optimization objective is to minimize the total energy consumption and total carbon emissions of the network while taking service quality into account. The optimized network operation map and intersection right-of-way strategy are obtained by solving the model.

[0017] In actual operation, rolling closed-loop control is implemented to monitor the actual passenger flow and operational deviations in real time. When the deviation exceeds the preset threshold, local re-optimization is triggered to generate an adjusted operation plan and issue it for execution.

[0018] Furthermore, the calculation of net vehicle energy consumption for each route segment based on the vehicle dynamics model and using the standardized operational dataset specifically includes:

[0019] Based on the segment speed in the standardized running dataset acceleration Line gradient and vehicle equivalent total mass Calculate the rolling resistance during vehicle operation. air resistance Slope resistance and acceleration resistance ;

[0020] The total traction resistance is calculated based on the individual resistance components. And combined with traction system efficiency Calculate traction power :

[0021]

[0022] Identify vehicle braking conditions and adjust braking power accordingly. With energy feedback efficiency Calculate the feedback energy and analyze the segment's operating time interval. Integrate to obtain the net energy consumption of the vehicle in the aforementioned section. :

[0023]

[0024] in:

[0025] Indicates the braking time interval.

[0026] Furthermore, the construction of the network-level passenger flow prediction model specifically includes:

[0027] A short-term passenger flow prediction model is constructed using a deep learning architecture based on recurrent neural networks. The input of the model is a vector of length [length missing]. The historical passenger flow sequence, the operation control input vector including departure intervals and route operation plans, and the external feature concatenation vector including holidays and weather are used to output the future data. Forecast values ​​of passenger flow demand at each station at each time step;

[0028] Obtain the network topology and transfer probability matrix. Based on the network topology and transfer probability matrix, map the predicted passenger flow demand values ​​of each station to each line segment, and calculate the future time slices of each segment. Predicted cross-sectional passenger flow ;

[0029] Based on the predicted cross-sectional passenger flow With the maximum capacity of the section Construct capacity constraints:

[0030]

[0031] in:

[0032] Let i be the maximum capacity of segment i under a given group length and allowable load factor.

[0033] Furthermore, in the construction of the network collaborative optimization model, the established optimization objective function Represented as:

[0034]

[0035] in:

[0036] The total energy consumption of the network is obtained by summing the net energy consumption of all trains in the network across all sections.

[0037] The total carbon emissions of the network are calculated by summing the section carbon emissions of all trains in all sections of the network.

[0038] The comprehensive service quality indicators include average travel time, average waiting time, and passenger load factor deviation.

[0039] , , These are the weighting coefficients for energy consumption, carbon emissions, and service quality, respectively.

[0040] Furthermore, the intersection signal phase information is incorporated into a unified calculation framework, including:

[0041] The line is calculated by accumulating the travel time of each section. Vehicles on the road arrive at the intersection Time variables ;

[0042] Leading to the intersection Green light window start time and the end time Construct coupling constraints between vehicle arrival time and green light window;

[0043] In the optimization objective function Add SPaT penalty item This allows vehicles to arrive outside the green light window, and penalizes the amount of deviation.

[0044]

[0045] in:

[0046] These are the signal coordination weighting coefficients.

[0047] Furthermore, in the execution of the rolling closed-loop control, real-time monitoring of actual passenger flow and operational deviations specifically includes:

[0048] Calculate any segment within each rolling cycle. In time slice Actual passenger flow With predicted passenger flow Deviation between ;

[0049] Determine the deviation amount Whether the amplitude continuously exceeds the preset ratio threshold, or the degree to which the current energy consumption and carbon emission indicators deviate from the threshold;

[0050] If the triggering conditions are met, a local re-optimization procedure is initiated, which fine-tunes the departure intervals and operating speeds of lines or time periods with excessive deviations while keeping the operation schedules of unaffected lines unchanged.

[0051] Furthermore, it also includes displaying optimization results through a network carbon emission visualization interface, specifically including:

[0052] The global overview area displays the total energy consumption and total carbon emissions of the power grid during the current statistical period, and shows their year-on-year or month-on-month trends relative to historical benchmarks.

[0053] The online map view area displays the intelligent rail lines and sections in the form of geographic topology, and the heat map coloring of each section is based on the calculated net energy consumption or carbon emission intensity of the section to intuitively present the bottleneck sections with high energy consumption or high carbon emissions.

[0054] The time-series curve view area displays the sequence curves of energy consumption and carbon emissions changing over time at the network or line level, and the triggered scheduling events are displayed as markers overlaid on the sequence curves.

[0055] This invention also proposes an intelligent rail transit network collaborative carbon reduction operation optimization system, comprising:

[0056] The data acquisition and processing module is used to collect multi-source operational data of the intelligent rail transit network and perform standardized processing to form a standardized operational dataset; the multi-source operational data includes vehicle operation data, passenger flow data, intersection signal data and basic energy consumption data.

[0057] The energy consumption and carbon emission calculation module is used to calculate the net energy consumption of vehicles in each line section based on the vehicle dynamics model and the standardized operation dataset, and to calculate the carbon emission of the section in combination with the power grid carbon emission factor, so as to establish a quantitative characterization of the network energy consumption and carbon emission.

[0058] The passenger flow prediction module is used to build a network-level passenger flow prediction model based on historical passenger flow sequences and real-time passenger flow input, and generate prediction results of the spatiotemporal distribution of passenger flow at each station and section in the future.

[0059] The network coordination optimization module is used to construct a network coordination optimization model. It incorporates the net energy consumption of vehicles, carbon emissions of sections, prediction results of passenger flow spatiotemporal distribution, and intersection signal phase information into a unified calculation framework. The optimization objective is to minimize the total energy consumption and total carbon emissions of the network while taking service quality into account. The optimized network operation map and intersection right-of-way strategy are obtained by solving the problem.

[0060] The operation scheduling and execution module is used to perform rolling closed-loop control during actual operation, monitor the actual passenger flow and operational deviations in real time, and trigger local re-optimization when the deviation exceeds the preset threshold, generate the adjusted operation plan and issue it for execution.

[0061] Furthermore, the system also includes:

[0062] A network carbon emission visualization module, connected to the energy consumption and carbon emission calculation module and the operation scheduling execution module, is used to generate a visual interactive interface;

[0063] The visual interactive interface is configured to perform the following operations:

[0064] The global overview area displays the total energy consumption and total carbon emissions of the network during the current statistical period;

[0065] The online map view area displays the intelligent rail transit line in the form of geographic topology, and the heat map coloring of each section is based on the calculated net energy consumption or carbon emission intensity of the section.

[0066] The time-series curve view area displays the sequence curves of grid energy consumption and carbon emissions over time, and overlays the triggered scheduling events as markers.

[0067] The present invention also proposes an electronic device, comprising:

[0068] At least one processor; and a memory communicatively connected to said at least one processor;

[0069] The memory stores instructions that can be executed by the at least one processor, and the at least one processor executes the instructions stored in the memory to perform the method described above.

[0070] Compared with existing technologies, the advantages of this invention are:

[0071] 1. This invention constructs a dynamic energy consumption model suitable for intelligent rail transit vehicles, realizing the quantitative characterization of section-level energy consumption and carbon emissions, and providing a directly callable computational basis for subsequent operation diagram optimization.

[0072] 2. This invention introduces a cross-line passenger flow prediction model and incorporates the prediction results into the scheduling optimization process, enabling the operation plan to match passenger flow demand in advance, improving the rationality of transport capacity allocation, and reducing additional energy consumption caused by scheduling delays.

[0073] 3. This invention forms a closed-loop control mechanism by combining energy consumption calculation, carbon emission quantification, passenger flow forecasting, and right-of-way coordination scheduling, enabling the system to update parameters based on real-time operating status, thereby achieving network-level energy-saving operation while meeting service constraints. Attached Figure Description

[0074] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in the embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0075] Figure 1 This is a schematic diagram of the overall structure of the intelligent rail transit network collaborative carbon reduction operation optimization system;

[0076] Figure 2 This is a flowchart of the data acquisition and processing method in the intelligent rail transit network collaborative carbon reduction operation optimization system;

[0077] Figure 3 This is a flowchart of the vehicle energy consumption and carbon emission calculation method in the intelligent rail transit network collaborative carbon reduction operation optimization system;

[0078] Figure 4 This is a schematic diagram illustrating the relationship between multi-source passenger flow prediction and network collaborative optimization in the intelligent rail transit network collaborative carbon reduction operation optimization system.

[0079] Figure 5 This is a flowchart of the closed-loop control of network-based carbon reduction operation in the intelligent rail network collaborative carbon reduction operation optimization system provided in an embodiment of the present invention;

[0080] Figure 6 This is a schematic diagram of the network energy consumption and carbon emission visualization interface structure in the intelligent rail network collaborative carbon reduction operation optimization system provided in an embodiment of the present invention;

[0081] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0082] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0083] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0084] Example 1

[0085] Please see Figure 1 An operational optimization method for intelligent rail transit network collaborative carbon reduction is proposed. This method mainly includes steps such as multi-source operation data acquisition and preprocessing, energy consumption and carbon emission calculation, passenger flow prediction and network collaborative optimization, operation closed-loop control, and network carbon emission visualization. Each step works collaboratively under a unified network coordinate and time benchmark to achieve comprehensive energy saving and emission reduction under multiple lines and multiple intersections.

[0086] In this embodiment, a method for optimizing the operation of intelligent rail transit network for coordinated carbon reduction specifically includes the following steps:

[0087] Step S1: Collect multi-source operation data of the intelligent rail transit network and perform standardized processing to form a standardized operation dataset; the multi-source operation data includes vehicle operation data, passenger flow data, intersection signal data and basic energy consumption data;

[0088] Specifically, such as Figure 2 As shown, the multi-source operational data (i.e., raw data) collected in this embodiment includes vehicle operation data, passenger flow data, intersection signal data, and basic energy consumption data. Among them, vehicle operation data may include vehicle position, speed, acceleration, traction / braking commands, etc.; passenger flow data may include the number of people boarding and alighting at the platform, the passenger load factor of the carriage, etc.; intersection signal data may include phase status, timing scheme, priority request records, etc.; and basic energy consumption data may include traction energy consumption, braking regenerative energy, and auxiliary power consumption, etc.

[0089] After completing multi-source data acquisition, various data types are processed for time alignment and coordinate unification, such as... Figure 2 As shown in the “Time Alignment and Coordinate Unification” section, data with different sampling frequencies and time bases are unified to the same time axis, and vehicle and passenger flow data are mapped to the same network coordinate system or section numbering system to ensure consistent spatial and temporal references for subsequent energy consumption and carbon emission calculations.

[0090] Furthermore, outlier detection and removal are performed on the aligned data, such as... Figure 2 As shown in the “Outlier Detection and Removal” section, outliers can be identified through threshold determination, sliding window statistics, or outlier detection methods based on historical distribution. For example, reasonable value ranges can be defined for variables such as speed, power, and passenger flow count. Data that clearly exceeds physical boundaries or undergoes short-term mutations and cannot be explained by operating conditions can be marked and removed, thereby improving the reliability of subsequent modeling.

[0091] After removing outliers, for unavoidable missing data, methods such as interpolation or historical reconstruction are used to fill in the missing data, such as... Figure 2 As shown in "Missing Data Completion (Interpolation / Historical Reconstruction)," interpolation methods can include linear interpolation, spline interpolation, etc. Historical reconstruction can be carried out by retrieving historical data from similar operating days, similar time periods, or similar passenger flow scenarios to reduce the systematic bias caused by missing data.

[0092] To reduce the impact of high-frequency noise and sensor jitter on energy consumption and carbon emission calculations, the preprocessed time series is filtered and denoised, such as... Figure 2 As shown in "Filtering and Noise Reduction (First-Order Low-Pass, etc.)", a first-order low-pass filter is preferred, and its recursive form can be expressed as:

[0093]

[0094] in, For the current moment The sampled values, For the current moment The filtered output value, The filter coefficients are determined by the sampling frequency and the cutoff frequency.

[0095] Based on the above, the data is aggregated and standardized according to the route and section, such as... Figure 2 As shown in the “Segment Aggregation and Standardization to Form a Standardized Dataset” section, vehicle operation trajectories, passenger flow data, and energy consumption data are aggregated according to the dimensions of “route-segment-time window” to form segment-level speed curves, load levels, and energy consumption records. Standardized operation datasets are formed through normalization, dimensionless processing, and other methods. Finally, a unified data source is output in the “Output Standardized Dataset” node, which provides a foundation for subsequent energy consumption and carbon emission calculations.

[0096] Step S2: Based on the vehicle dynamics model, calculate the net energy consumption of vehicles in each line section using the standardized operation dataset, and calculate the carbon emissions of the section in combination with the power grid carbon emission factor, and establish a quantitative characterization of network energy consumption and carbon emissions.

[0097] like Figure 3 As shown, in this embodiment, the operating data of each line and each section is first read from the standardized operating dataset to obtain the section speed that changes over time. acceleration Line gradient And corresponding load information, where the load can be converted into the vehicle's equivalent mass through vehicle occupancy rate or passenger flow count. .

[0098] Based on this, rolling resistance is calculated using the vehicle dynamics model. air resistance Slope resistance and acceleration resistance The resistance components, when calculated in equal parts, can typically be represented as follows:

[0099]

[0100]

[0101]

[0102]

[0103] in:

[0104] The total mass includes the vehicle's own weight and equivalent load.

[0105] It is the acceleration due to gravity;

[0106] This is the rolling resistance coefficient;

[0107] air density;

[0108] This refers to the air drag coefficient;

[0109] This refers to the windward area.

[0110] Furthermore, the combined total traction resistance And calculate the traction power. Its calculation formula can be expressed as:

[0111]

[0112]

[0113] Right now:

[0114]

[0115] in:

[0116] The traction system efficiency is used to reflect the comprehensive losses of mechanical transmission, motor and converter components. It can be obtained by the operating unit through type testing and historical data calibration, which is routine work for those skilled in the art.

[0117] When the vehicle is braking, this embodiment calculates the braking power and recyclable energy, and uses them as energy feedback items in the net energy consumption calculation. The feedback energy can be expressed as:

[0118]

[0119] in:

[0120] Indicates the braking time interval;

[0121] Braking power is the power generated by the vehicle's kinetic energy during braking.

[0122] Energy recovery efficiency can be determined through vehicle testing or manufacturer parameters and calibrated in operation.

[0123] For each line segment, this embodiment performs time integration on the traction power and the feedback power to obtain the segment's net energy consumption. Its typical form is:

[0124]

[0125] in:

[0126] This refers to the operating time interval for this section;

[0127] Right now:

[0128]

[0129] After obtaining the net energy consumption of the section, the carbon emission factor per unit of electricity of the power grid is used as a basis. Calculate the carbon emissions of this section. The correspondence can be represented as:

[0130]

[0131] in:

[0132] The power structure and power supply composition can be determined by the power grid company or configured by the operating unit according to time periods, and in this embodiment, it can be used as a time-slice. changing function The method of using this parameter configuration is one that can be directly implemented by those skilled in the art.

[0133] For the entire line and the entire intelligent rail transit network, this embodiment obtains the line-level and network-level carbon emissions by summing the carbon emissions of each section, such as... Figure 3 The "cumulative carbon emissions of each section of the network" in the text As shown in the figure, its calculation form can be expressed as:

[0134]

[0135] in:

[0136] The set of all lines in the network;

[0137] For the line The set of segments above;

[0138] For the line Upper section Carbon emissions.

[0139] Using the methods described above, this step achieves refined calculations of energy consumption and carbon emissions from vehicle operation data to section energy consumption and carbon emissions, providing quantitative energy consumption and carbon emission indicators for subsequent network-level collaborative optimization.

[0140] Step S3: Construct a network-level passenger flow prediction model based on historical passenger flow sequences and real-time passenger flow input, and generate prediction results of the spatiotemporal distribution of passenger flow at each station and section in the future.

[0141] like Figure 4 As shown in this embodiment, the input data includes historical passenger flow data (such as OD matrix, card swipe records, boarding statistics, etc.), real-time passenger flow and onboard counting data (such as carriage occupancy rate, platform queue length, etc.), operation diagrams and event data (such as timetables for each line, interchange points and connection relationships, etc.), and section and energy consumption data (such as gradient, speed limit and energy consumption curve, etc.). The above multi-source data together constitute the basis for describing the passenger flow and operation status of the network.

[0142] Regarding online network-level passenger flow forecasting, this embodiment preferably employs a short-term passenger flow forecasting model based on a recurrent neural network, such as... Figure 4 As shown in the "Short-term Passenger Flow Prediction (Multi-source Time Series / Deep Learning)" document, in one specific implementation, the prediction function... The structure employs a two-layer LSTM network plus a fully connected output layer, with an input of length [length missing]. The historical passenger flow sequence, operation control input, and external feature concatenation vector are used to output the future... The predicted value of passenger flow demand at each station at each time step.

[0143] The model can be represented as:

[0144]

[0145] in:

[0146] For length is Passenger flow observation sequence;

[0147] For input sequences related to the train schedule, such as departure intervals and line operation plans;

[0148] External characteristics such as holidays and weather;

[0149] For the future A station-level passenger flow prediction sequence at each time step.

[0150] During the model training phase, this embodiment learns parameters by minimizing the mean squared error or mean absolute percentage error between the predicted values ​​and the historical actual passenger flow. The loss function can be expressed as:

[0151]

[0152] in:

[0153] This represents the number of training samples;

[0154] For the first Sample at time The actual passenger flow;

[0155] This corresponds to the predicted value;

[0156] This embodiment is not limited to the LSTM structure. Those skilled in the art can choose alternative structures such as one-dimensional convolutional networks or spatiotemporal graph neural networks according to the actual data characteristics. Parameter training can also be achieved through conventional deep learning frameworks and optimizers (such as Adam). This is an implementation method that can be completed without creative effort.

[0157] Regarding passenger flow allocation, this embodiment considers the passenger flow diversion and transfer needs at interchange stations, such as... Figure 4 As shown in the section on "Cross-line Passenger Flow Distribution (Interchange Station Diversion, Transfer Demand)," the station-level passenger flow forecast results are mapped to each line segment through the network topology to obtain the predicted cross-sectional passenger flow and load level of each segment at a given time. For any segment... With Time Slice Predicted cross-sectional passenger flow It is calculated by combining passenger flow at upstream and downstream stations and the transfer allocation ratio, and is used for subsequent capacity constraints.

[0158] In this embodiment, the interface between the passenger flow prediction module and the network collaborative optimization module is clearly defined: prediction output This is directly used to constrain the passenger capacity of a section from exceeding the vehicle capacity limit, i.e., it satisfies:

[0159]

[0160] in:

[0161] Let i be the maximum capacity of segment i under a given group length and allowable load factor;

[0162] This embodiment can also deduce the minimum departure frequency for each route based on predicted passenger flow, for example, requiring a minimum departure frequency at a certain station. The average waiting time does not exceed the threshold. Then the departure interval Apply constraints:

[0163]

[0164] This results in the transmission of passenger flow demand constraints to the variables in the operational diagram;

[0165] Through the above passenger flow forecasting and allocation steps, this embodiment obtains the future passenger flow load forecast results for each line and section under different operating schedules and scheduling strategies, providing demand-side constraints and service level evaluation basis for subsequent network collaborative optimization.

[0166] Step S4: Construct a network collaborative optimization model, incorporating the vehicle net energy consumption, section carbon emissions, passenger flow spatiotemporal distribution prediction results, and intersection signal phase information into a unified calculation framework. The optimization objective is to minimize the total energy consumption and total carbon emissions of the network while taking service quality into account. The optimized network operation map and intersection right-of-way strategy are then obtained by solving the problem.

[0167] like Figure 4 As shown, this embodiment uses the "network collaborative optimization engine" as the core module, takes the passenger flow prediction results obtained in the previous step and the section energy consumption and carbon emission indicators output by the energy consumption and carbon emission calculation module of this invention as inputs, and combines the existing operation map and event data as well as the section energy consumption characteristics to construct a multi-line collaborative optimization model.

[0168] In designing the objective function, this embodiment comprehensively considers three factors: grid energy consumption, carbon emission levels, and service quality. A unified objective function is constructed through weighted summation, such as... Figure 5 Solving the network collaborative optimization operation diagram As shown, its form can be expressed as:

[0169]

[0170] in:

[0171] Total energy consumption of the network;

[0172] For total carbon emissions of the network;

[0173] For comprehensive service quality indicators, they can be weighted by factors such as average travel time, average waiting time, and passenger load factor deviation.

[0174] The weighting coefficients are set by the operator based on policy guidance and service requirements. They can be calibrated through multiple rounds of operational simulation or expert experience, and are parameter setting methods that are easy for those skilled in the art to implement.

[0175] , , These are the weighting coefficients for energy consumption, carbon emissions, and service quality, respectively.

[0176] Regarding decision variables, this embodiment includes at least the departure intervals for each line and time period. Target running time for each section Stop times at each station And the target passing speed or passing time variable at key intersections. Used to depict vehicles arriving at intersections At that moment.

[0177] Regarding constraints, this embodiment includes at least operational safety constraints, capacity and congestion constraints, intersection signal coordination constraints, and practical feasibility constraints. The train tracking interval constraint can be expressed as:

[0178]

[0179] in:

[0180] For the line The minimum permissible safety interval; consistency between segment running time and speed can be achieved through:

[0181]

[0182] Expression, in which For segment length, This is the average operating speed of this section, and at the same time... Upper and lower limits are imposed to ensure compliance with line speed limits.

[0183] Regarding capacity and congestion constraints, this embodiment ensures that the predicted passenger flow does not exceed the vehicle capacity by limiting the cross-sectional passenger load factor, that is:

[0184]

[0185] As mentioned earlier, additional constraints can be used to limit the extent to which the occupancy rate deviates from the target value, thereby controlling passenger comfort.

[0186] Regarding signal and right-of-way coordination, this embodiment explicitly introduces intersection green light window constraints to achieve coupling between SPaT and speed. For intersections... A green light window is defined with its start and end times as follows: and For vehicles departing from upstream stations, their arrival time variable Trains can travel from the starting point of this line to the intersection. The calculation is obtained by sequentially summing the running times of each segment, and its calculation form is as follows:

[0187]

[0188] in:

[0189] Indicates the line From the departure point to the intersection The segment sequence.

[0190] This embodiment uses the above-mentioned segment time accumulation method to enable the arrival time to be used as a decision variable and directly coupled with the segment running time.

[0191] This embodiment uses the following constraints:

[0192]

[0193] Limit vehicle arrival times to within the green light window or allow limited offsets. ,in This is the allowable time margin.

[0194] By adjusting the segment travel time and speed variables, the vehicle's arrival time is made as close to the green light window as possible, thereby reducing the energy consumption from ineffective braking and re-acceleration caused by stopping at red lights. For cases where the green light window requirement cannot be fully met, this embodiment adds the following convex penalty term to the objective function to quantify the energy loss caused by arriving late or early:

[0195]

[0196] in:

[0197] These are the signal coordination weighting coefficients.

[0198] This penalty function allows for a continuously differentiable trade-off between optimal energy consumption and feasible signal timing, thereby enabling computable optimization of the green light window.

[0199] In terms of solution methods, this embodiment preferably constructs the objective function and linearized constraints into a mixed-integer linear programming model, and solves it using a commercial optimization solver (such as Gurobi) or the MILP solver module of an open-source optimization library (such as OR-Tools). For large-scale network optimization problems, this embodiment can also adopt a decomposition and coordination strategy, decomposing the overall network optimization problem into line sub-problems and intersection coordination sub-problems, and then solving them through Lagrange relaxation or alternating direction multiplier method. Those skilled in the art can directly use existing optimization solvers to build models and obtain feasible solutions based on the above mathematical structure, without any creative effort.

[0200] To ensure that the above model can be directly processed by the MILP solver, this embodiment further presents a typical method for linearizing green light feasibility. Define 0–1 decision variables:

[0201]

[0202] The green light window constraint can then be linearized as follows:

[0203]

[0204]

[0205] in:

[0206] It is a sufficiently large constant.

[0207] The corresponding penalty can be written as:

[0208]

[0209] This constitutes a standard mixed-integer linear programming structure, which can be directly solved by Gurobi or OR-Tools at multi-line scales.

[0210] By jointly modeling the above objective function and constraints, this embodiment obtains the running graph after multi-path collaborative optimization, as follows: Figure 4 The "Optimized Network Operation Map (Timetables and Routing Schemes for Each Line)" shows the corresponding intersection signaling and right-of-way control strategies, as follows: Figure 4 The document shows the "Intersection Signal and Right-of-Way Strategy (SPaT Configuration, Priority Assignment)" section, and outputs the energy consumption and carbon emission assessment results under this scheduling scheme, as shown below. Figure 4 The results are shown in the "Energy Consumption and Carbon Emission Assessment Results (Multi-dimensional Indicators for Network / Line / Section)".

[0211] Through the above-mentioned collaborative optimization steps, this embodiment achieves integrated optimization of the intelligent rail transit network operation map and intersection priority strategy under the premise of meeting safety and service constraints, thereby reducing the overall energy consumption and carbon emission level of the network while taking into account the passenger travel experience.

[0212] Step S5: During actual operation, execute rolling closed-loop control, monitor the actual passenger flow and operational deviation in real time, and trigger local re-optimization when the deviation exceeds the preset threshold, generate the adjusted operation plan and issue it for execution;

[0213] like Figure 5 As shown, in this embodiment, after the system "starts network operation status monitoring", it periodically collects multi-source data such as real-time passenger flow, operation status, intersection signals, and energy consumption. Figure 5The system collects real-time data (passenger flow, operation, signal, energy consumption) and performs short-term passenger flow forecasting and energy consumption estimation based on this data, obtaining passenger flow and energy consumption trends over a future period in a rolling manner.

[0214] For each rolling cycle, the system invokes the aforementioned network collaborative optimization engine to solve for the updated running graph and right-of-way strategy, corresponding to... Figure 5 The "Solving the Network Collaborative Optimization Operation Diagram" The proposed solution was then subjected to an "operational safety and constraint feasibility check" to ensure that it did not violate hard constraints such as safety intervals, station capacity, and intersection timing.

[0215] After the feasibility check is passed, the system evaluates whether adjustments to the current operating diagram and signal control strategy are needed based on pre-set trigger rules, such as... Figure 5 As shown in the "Trigger Adjustment of Operation Schedule?" section, the triggering rules can be set comprehensively based on factors such as predicted sudden changes in passenger flow, the degree of deviation of current energy consumption and carbon emissions from the threshold, or the severity of operational disturbances. When the adjustment conditions are met, the "Issue Operation Schedule Adjustment and Right-of-Way / Signal Priority Strategy" step is executed, and the optimized operation schedule and priority strategy are issued to the field equipment through the dispatching system and intersection control system.

[0216] After the operational adjustments are completed, the system continues to collect data on the execution effect and evaluates energy consumption and carbon emission indicators, such as... Figure 5 As shown in the “Energy Consumption and Carbon Emission Indicators for Evaluating the Implementation Effect” section, the actual effect of the current cycle’s coordinated carbon reduction measures is analyzed by comparing with the unoptimized plan or historical benchmarks. The results are summarized and displayed at the “Update Network Carbon Visualization Interface to Form a Phase Evaluation” node, and finally enter the “Closed Loop Completed and Enter the Next Cycle Monitoring” state, forming a continuous carbon reduction operation closed loop.

[0217] In this embodiment, the interface between the operation scheduling execution module and the passenger flow prediction module is also explicitly utilized: for any segment With Time Slice The deviation between actual passenger flow and predicted passenger flow can be expressed as:

[0218]

[0219] in:

[0220] This refers to the actual passenger flow obtained from on-site counting equipment;

[0221] These are predicted values.

[0222] The system according to The amplitude of the data determines whether the departure interval or local running time needs to be adjusted. For example, when the deviation of a certain key section continues to exceed the preset proportional threshold, the system triggers local re-optimization to make fine adjustments to the route or time period, thereby improving the system's adaptability to passenger flow fluctuations and changes in the road operating environment.

[0223] In this embodiment, when the judgment result does not meet the adjustment conditions, the system can maintain the current operation diagram and signal strategy unchanged, only update the monitoring and evaluation results, and continue to execute the rolling prediction and optimization process in the next cycle, so as to reduce unnecessary adjustment times and take into account the operational stability.

[0224] Step S6: Display the optimization results through the network carbon emission visualization interface to provide intuitive monitoring and decision support for operation and management personnel, specifically including:

[0225] like Figure 6 As shown, the network carbon emission visualization interface provided in this embodiment is divided into several functional areas, including global overview, network map view, time series curve view, operation status and optimization suggestions, and line / segment ranking. The main body of the interface is used to centrally present the current network energy consumption and carbon emission levels, as well as the implementation effect of collaborative optimization.

[0226] In the global overview area, such as Figure 6 The "Overall Overview - Current Total Network Energy Consumption" section... Total carbon emissions As shown in the “Year-on-Year / Month-on-Month Trend, Threshold Alarm Status”, this embodiment displays the total energy consumption and total carbon emissions of the power grid within the current statistical period, and provides the year-on-year and month-on-month change trends compared with the same period in history. At the same time, it provides over-limit alarm or early warning information based on preset thresholds so that managers can quickly grasp the overall operating status.

[0227] Online map view area, such as Figure 6 As shown in the “Network Map View · Display ART Lines, Stations, and Sections · Sections Colored by Energy Consumption / Carbon Emission Intensity · Supports Zooming and Selecting Lines / Sections”, this embodiment displays intelligent rail lines, stations, and sections in a geographical or topological form, and colors sections according to energy consumption per unit mileage or carbon emission intensity per unit passenger kilometer to achieve spatial positioning of high energy consumption or high carbon emission sections. Users can view detailed indicators of specific lines or sections through zooming and selection operations.

[0228] In the time series curve view area, such as Figure 6As shown in the “Time Series Curve View · Network / Line Energy Consumption and Carbon Emission Time Series · Supports Selecting Time Windows (15min / 1h / 1d) · Can Overlay Operation Charts and Scheduling Event Markers”, this embodiment provides the trend of energy consumption and carbon emissions over time at the network or line level. It supports viewing different time windows such as 15 minutes, 1 hour or 1 day, and can overlay key scheduling events such as operation chart changes and intersection priority issuance on the time series curve in the form of marks, so as to analyze the relationship between a certain scheduling strategy and changes in energy consumption and carbon emissions.

[0229] In the running status and optimization suggestions area, such as Figure 6 As shown in the “Operating Status and Optimization Suggestions, Current Operating Mode and Key Alarms, Energy Consumption / Carbon Emission Deviation from Thresholds, Automatically Generated Energy-Saving Scheduling Suggestion Summary” section, this embodiment summarizes the current network operating mode (e.g., weekday peak, off-peak, weekend modes), key operating alarms, and the degree of deviation of energy consumption and carbon emission comparison thresholds. Based on the results of the collaborative optimization engine, it generates an energy-saving scheduling suggestion summary, providing dispatchers with optimization schemes that can be directly implemented or referenced.

[0230] In the route / segment ranking area, such as Figure 6 As shown in the "Route / Segment Ranking • Sort by Energy Consumption / Carbon Emission per Passenger Kilometer • Supports Filtering by Route and Time Period • Allows Jumping to View Detailed Curves for a Specific Route" section, this embodiment sorts each route or segment based on energy consumption per passenger kilometer or carbon emission intensity. It supports filtering by route, time period, and other conditions. When a user selects a route or segment, they can jump to its detailed time-series curve and spatial distribution interface, thereby helping managers identify routes or segments with significant energy-saving potential.

[0231] Through the above visualization steps, this embodiment presents the complex results of multi-line collaborative carbon reduction optimization in an intuitive form, which facilitates comparative analysis and strategy adjustment by operation and management personnel, thereby further enhancing the feasibility and application value of the method of the present invention in actual engineering scenarios.

[0232] Please see Figure 1 Based on the same inventive concept, this embodiment also proposes an intelligent rail transit network collaborative carbon reduction operation optimization system, the software display interface structure of which can be referred to Figure 6 The above system can be implemented in whole or in part in the form of software, hardware, firmware or any combination thereof, and the deployment and customization by those skilled in the art according to actual engineering conditions are all within the protection scope of this invention.

[0233] In this embodiment, specifically, an intelligent rail transit network collaborative carbon reduction operation optimization system is designed for urban rail transit networks composed of multiple intelligent rail transit lines. Through multi-source data acquisition, vehicle dynamics modeling, carbon emission quantification calculation, passenger flow prediction, and collaborative optimization of timetables, it achieves energy-saving and carbon-reducing scheduling throughout the entire network operation process. The system integrates energy consumption characterization, carbon emission factor models, and dynamic passenger flow distribution into a unified optimization framework, forming a network-level collaborative operation scheme tailored to the characteristics of intelligent rail transit technology. Specifically, it includes:

[0234] The data acquisition and processing module is used to collect multi-source operational data of the intelligent rail transit network and perform standardized processing to form a standardized operational dataset; the multi-source operational data includes vehicle operation data, passenger flow data, intersection signal data and basic energy consumption data.

[0235] The energy consumption and carbon emission calculation module (in this embodiment, it essentially includes a vehicle energy consumption modeling module and a carbon emission calculation module) is used to calculate the net energy consumption of vehicles in each line section based on the vehicle dynamics model and the standardized operation dataset, and to calculate the carbon emission of the section in combination with the power grid carbon emission factor, so as to establish a quantitative characterization of the network energy consumption and carbon emission.

[0236] The passenger flow prediction module is used to build a network-level passenger flow prediction model based on historical passenger flow sequences and real-time passenger flow input, and generate prediction results of the spatiotemporal distribution of passenger flow at each station and section in the future.

[0237] The network coordination optimization module is used to construct a network coordination optimization model. It incorporates the net energy consumption of vehicles, carbon emissions of sections, prediction results of passenger flow spatiotemporal distribution, and intersection signal phase information into a unified calculation framework. The optimization objective is to minimize the total energy consumption and total carbon emissions of the network while taking service quality into account. The optimized network operation map and intersection right-of-way strategy are obtained by solving the problem.

[0238] The operation scheduling and execution module is used to perform rolling closed-loop control during actual operation, monitor the actual passenger flow and operational deviation in real time, and trigger local re-optimization when the deviation exceeds the preset threshold, generate the adjusted operation plan and issue it for execution.

[0239] A network carbon emission visualization module, connected to the energy consumption and carbon emission calculation module and the operation scheduling execution module, is used to generate a visual interactive interface;

[0240] The visual interactive interface is configured to perform the following operations:

[0241] The global overview area displays the total energy consumption and total carbon emissions of the network during the current statistical period;

[0242] The online map view area displays the intelligent rail transit line in the form of geographic topology, and the heat map coloring of each section is based on the calculated net energy consumption or carbon emission intensity of the section.

[0243] The time-series curve view area displays the sequence curves of grid energy consumption and carbon emissions over time, and overlays the triggered scheduling events as markers.

[0244] The modules call each other through data interfaces and a unified scheduler.

[0245] The system constructs state variables that can be used for optimization solutions from the actual operation information of the power grid, generates corresponding carbon reduction operation schemes, and sends the calculation results to the operation center in real time for execution.

[0246] In this embodiment, the data acquisition and processing module of the system will be further described in detail:

[0247] The system first uses a data acquisition and processing module to access multi-dimensional operational data from the intelligent rail network, including station distances, gradients, intersection locations, and signal timings; and vehicle speeds during operation. acceleration Traction and braking power; station passenger flow, transfer volume; traction power consumption and grid carbon emission factors. To ensure the model's computability, the module cleans, standardizes, aligns, and interpolates all data to form a unified spatiotemporal data structure.

[0248] In this embodiment, the vehicle energy consumption modeling module in the system will be further explained:

[0249] The vehicle energy consumption modeling module, based on the tire steering characteristics, lightweight body structure, and road driving environment of ART vehicles, transforms standardized data into dynamic state variables that can be used for energy consumption optimization. For any operating segment, this module calculates the traction power based on the vehicle's speed and acceleration sequence.

[0250]

[0251] Rolling resistance, air resistance, and slope resistance are respectively calculated as follows:

[0252]

[0253] This is determined, and then combined with the energy recovery efficiency during the braking phase. The net energy consumption of the section is obtained as follows:

[0254]

[0255] This module establishes an energy consumption mapping between operating speed curves, track gradients, vehicle loads, and real-time operational behavior.

[0256] In this embodiment, the carbon emission calculation module in the system will be further described in detail:

[0257] This invention combines vehicle net energy consumption with grid carbon emission factors to calculate the carbon emissions for a given area.

[0258]

[0259] By summing the time of all trains and all sections within the network, the total carbon emissions of the network over a given period can be obtained. This module establishes a direct quantitative relationship between vehicle operation behavior and carbon emission indicators, and is the core foundation for the system to achieve the goal of "coordinated carbon reduction".

[0260] In this embodiment, the passenger flow prediction module in the system will be further explained in detail:

[0261] To enable operational scheduling to adapt to changes in passenger flow over time and space, the system models historical passenger flow sequences, real-time passenger flow inputs, and transfer relationships to generate passenger flow forecasts for each station and section in the future. The forecast model comprehensively considers time periodicity, holiday effects, and the topological adjacency relationships between stations to obtain passenger flow estimates for future times. The prediction results can be continuously input into the network collaborative optimization module, enabling the operation plan to optimize capacity configuration in advance and avoid high-energy-consumption scheduling caused by sudden surges in passenger flow.

[0262] In this embodiment, the wire network collaborative optimization module in the system will be further described in detail:

[0263] The network-wide collaborative optimization module is the core of the system. This module integrates energy consumption models, carbon emission models, and passenger flow prediction models into a unified optimization framework, and jointly solves for the departure intervals, interval travel times, station dwell times, and speed control strategies for key sections of the entire network.

[0264] The optimization process aims to reduce the total energy consumption and carbon emissions of the rail network while ensuring service quality, operational safety, and vehicle resource accessibility. The established objective function can be expressed as:

[0265]

[0266] in and The weighting coefficients are used to balance energy consumption and carbon emissions.

[0267] The optimization constraints involve passenger flow satisfaction, minimum departure interval, consistency between speed and travel time, accessibility of green light windows at key intersections, and vehicle dispatching capacity. During the solution process, the system iteratively updates vehicle arrival times at intersections to coordinate speed strategies with signal timing, thereby reducing unnecessary acceleration and deceleration losses.

[0268] In this embodiment, the operation scheduling and execution module of the system will be further described in detail:

[0269] The operational plan output by the optimization module is loaded and implemented by the operation scheduling execution module, including the allocation of the number of vehicles online, the setting of departure intervals during time periods, the operation speed control strategy, and the right-of-way request strategy for connecting to the signal system. This module is also responsible for real-time monitoring of actual passenger flow and operational deviations. When the deviation exceeds a set threshold, it automatically triggers local re-optimization to achieve rolling adjustments to the operation schedule.

[0270] The deviation can be obtained from:

[0271]

[0272] The system uses this deviation information to correct the departure interval and speed curve, improving the system's adaptability to passenger flow fluctuations and changes in the road operating environment.

[0273] In this embodiment, it should also be noted that the above system operates as follows:

[0274] Each module forms a closed loop through data flow and scheduling logic coupling:

[0275] The data acquisition module continuously generates operational status → the energy consumption and carbon emission model measures energy-saving changes in real time → the passenger flow forecasting module generates future demand → the network collaborative optimization module forms an operational plan → the scheduling execution module implements the plan and provides feedback on the results → the feedback re-enters the optimization process.

[0276] This closed-loop mechanism enables the system to continuously minimize energy consumption and carbon emissions, achieving an "on-grid level, dynamic, and collaborative carbon reduction" operation and control mode.

[0277] Based on the same technical concept, embodiments of the present invention also provide an electronic device that can implement the operational optimization method for intelligent rail transit network collaborative carbon reduction provided in the above embodiments of the present invention. In one embodiment, the electronic device can be a server, a terminal device, or other electronic equipment. Figure 7 As shown, the electronic device may include:

[0278] At least one processor and a memory connected to the at least one processor. In this embodiment of the invention, the specific connection medium between the processor and the memory is not limited. Figure 7 The example used is the connection between the processor and memory via a bus. The bus... Figure 7 The connections between other components are indicated by thick lines and are for illustrative purposes only, not as limiting information. Buses can be divided into address buses, data buses, control buses, etc., but for ease of representation, [the specific bus type is not shown here]. Figure 7The processor is represented by a single thick line, but this does not imply that there is only one bus or one type of bus. Alternatively, a processor can also be called a controller; there are no restrictions on the name.

[0279] In this embodiment of the invention, the memory stores instructions executable by at least one processor. By executing the instructions stored in the memory, the at least one processor can perform the aforementioned operational optimization method for intelligent rail transit network-based carbon reduction. The processor can implement... Figure 7 The functions of each module in the device shown.

[0280] The processor is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory and calling data stored in memory, it can monitor the device's various functions and process data, thereby enabling overall monitoring of the device.

[0281] In an alternative design, the processor may include one or more processing units. The processor may integrate an application processor and a modem processor, wherein the application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may also not be integrated into the processor. In some embodiments, the processor and memory may be implemented on the same chip; in some embodiments, they may also be implemented separately on separate chips.

[0282] The processor can be a general-purpose processor, such as a CPU, digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the intelligent rail transit network collaborative carbon reduction operation optimization method disclosed in the embodiments of this invention can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0283] Memory, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory can include at least one type of storage medium, such as flash memory, hard disk, multimedia cards, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), and electrically erasable programmable read-only memory (EPROM). Only memory (EEPROM), magnetic storage, magnetic disks, optical disks, etc. A memory is any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. The memory in embodiments of this invention can also be a circuit or any other device capable of performing storage functions for storing program instructions and / or data.

[0284] By designing and programming the processor, the code corresponding to the intelligent rail transit network collaborative carbon reduction operation optimization method described in the foregoing embodiments can be embedded into the chip, enabling the chip to execute the steps of the method described in the foregoing embodiments during operation. How to design and program the processor is a technique well-known to those skilled in the art and will not be elaborated upon here.

[0285] Example 2

[0286] Based on the method described in Example 1, a specific intelligent rail transit network application scenario and example calculation process are given to illustrate the applicability and carbon reduction effect of the present invention.

[0287] In this embodiment, a city has constructed two intelligent rail transit lines. and ,in As the main artery, it serves commuter traffic between the city's main urban area and new town clusters. As a branch line, it undertakes the task of connecting the new city and the main line. The two lines intersect and transfer at two hub stations.

[0288] During weekday morning rush hours, the operating unit deploys the collaborative carbon reduction operation optimization system of the present invention on the above-mentioned network. First, following step S1 of Embodiment 1, it collects multi-source data from vehicle terminals, passenger flow counting devices, intersection signal controllers and energy consumption metering devices, and forms a standardized operation dataset through time alignment, anomaly removal, missing data completion and filtering noise reduction.

[0289] According to step S2 of Embodiment 1, the system calculates based on standardized data. and The net energy consumption and carbon emissions of each section are calculated and summed to obtain the total network energy consumption under the current operating schedule. and total carbon emissions To improve the sufficiency of disclosure, this embodiment provides... A reproducible example of energy consumption and carbon emission calculations is given for a 1 km uphill section.

[0290] Assuming the vehicle's equivalent total mass:

[0291]

[0292] Average speed:

[0293]

[0294] Average acceleration:

[0295]

[0296] Slope angle:

[0297]

[0298] Rolling resistance coefficient:

[0299]

[0300] Air drag coefficient:

[0301]

[0302] Air density:

[0303]

[0304] The typical resistance can be calculated according to step S2 of Example 1 as follows:

[0305] Rolling resistance:

[0306]

[0307] Air resistance:

[0308]

[0309] Slope resistance:

[0310]

[0311] Acceleration resistance:

[0312]

[0313] Total combined resistance:

[0314]

[0315] In terms of traction system efficiency:

[0316]

[0317] Under these conditions, the average traction power is:

[0318]

[0319] The approximate time it takes for a vehicle to pass through this section is:

[0320]

[0321] Corresponding traction energy consumption:

[0322]

[0323] If the recoverable braking energy is:

[0324]

[0325] Then the net energy consumption of the section:

[0326]

[0327] If the power grid carbon factor for the time period is:

[0328]

[0329] The carbon emissions for this section are:

[0330]

[0331] The above example calculations can be used to compare changes in energy consumption and carbon emissions before and after optimization, providing quantifiable verification of the synergistic carbon reduction effect of this invention.

[0332] Subsequently, following step S3 of Example 1, the system uses historical passenger flow records from the past month and real-time onboard passenger flow counting data to make a short-term forecast of passenger flow for the next two hours during the weekday morning rush hour. Then, through a passenger flow allocation model at interchange stations, the forecasted demand is mapped to various sections of the two lines to obtain the results under different scenarios. and The cross-sectional load level.

[0333] To enable the station passenger flow forecast results to be used for segment capacity constraints, this embodiment further provides the topological mapping relationship between station passenger flow and line segments. Under the network topology, any segment At any moment The predicted cross-sectional passenger flow can be expressed as:

[0334]

[0335] in, For site collection, For the site At any moment The predicted number of passengers boarding. The passenger flow allocation ratio, determined by the route topology and transfer probability, satisfies:

[0336]

[0337] The above formula can be used to directly convert station-level passenger flow forecast results into section-level cross-sectional load, thus establishing a clear interface between the passenger flow forecast module and the collaborative optimization module.

[0338] In step S4 of Example 1, the system constructs a collaborative optimization model with network energy consumption, carbon emissions, and service quality as objectives, and adjusts... and The departure intervals, route schemes, and signal priority strategies at key intersections are determined to obtain a set of collaborative optimization operation graphs that satisfy safety constraints and waiting time upper limits. In the example, under the baseline scheme... The departure interval during the morning rush hour is 10 minutes. After optimization, the departure interval will be reduced to 8 minutes during the busiest hour. During certain periods, the frequency of departures will be appropriately increased to achieve a balance between the overall network capacity and passenger demand. Optimization results show that, compared with the baseline operating schedule, the total energy consumption of the network has been reduced by approximately 8%, total carbon emissions by approximately 9%, and the increase in average waiting time has been controlled within 5%.

[0339] In step S5 of Example 1, the system updates passenger flow forecast and energy consumption estimate on a rolling cycle of 15 minutes during actual operation. When an abnormally concentrated passenger flow is detected on a certain workday (such as the end of a large event), the system automatically triggers the operation plan fine-tuning and right-of-way priority strategy under the premise of meeting operational safety and constraints. By temporarily increasing some short-distance bus services and giving priority to key intersections, local congestion is alleviated and ineffective energy consumption caused by large-scale vehicle addition is avoided.

[0340] According to step S6 of Embodiment 1, the operating unit can view the line-level and section-level energy consumption and carbon emission trends of L1 and L2 in real time through the network carbon visualization interface, compare the monthly total carbon emission changes before and after implementing the method of the present invention, and further identify sections with greater energy-saving potential by combining the "line / section ranking" function, so as to provide a basis for decision-making for subsequent infrastructure transformation and driver energy-saving training.

[0341] As can be seen from this embodiment, the method of the present invention can reduce the overall energy consumption and carbon emissions of the rail network under the premise of ensuring service quality in multi-line collaborative operation scenarios, realize refined carbon emission management and collaborative carbon reduction operation of the intelligent rail network, and has good engineering promotion value.

[0342] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.

[0343] This background section is provided to generally present the context of the invention. The work of the currently named inventors, the work to the extent described in this background section, and aspects of this section that did not constitute prior art at the time of application are neither expressly nor impliedly acknowledged as prior art to the invention.

Claims

1. An operational optimization method for coordinated carbon reduction in intelligent rail transit networks, characterized in that, include: Collect multi-source operational data of the intelligent rail transit network and perform standardized processing to form a standardized operational dataset; The multi-source operational data includes vehicle operation data, passenger flow data, intersection signal data, and basic energy consumption data; Based on the vehicle dynamics model, the net energy consumption of vehicles in each line section is calculated using the standardized operation dataset, and the carbon emissions of the section are calculated by combining the power grid carbon emission factor, thus establishing a quantitative characterization of network energy consumption and carbon emissions. Based on historical passenger flow sequences and real-time passenger flow input, a network-level passenger flow prediction model is constructed to generate prediction results of the spatiotemporal distribution of passenger flow at each station and section in the future. A network collaborative optimization model is constructed, which incorporates the net energy consumption of vehicles, carbon emissions of sections, prediction results of passenger flow spatiotemporal distribution, and intersection signal phase information into a unified calculation framework. The optimization objective is to minimize the total energy consumption and total carbon emissions of the network while taking service quality into account. The optimized network operation map and intersection right-of-way strategy are obtained by solving the model. During actual operation, rolling closed-loop control is implemented to monitor the actual passenger flow and operational deviations in real time. When the deviation exceeds the preset threshold, local re-optimization is triggered to generate an adjusted operation plan and issue it for execution. The construction of the network-level passenger flow prediction model specifically includes: A short-term passenger flow prediction model is constructed using a deep learning architecture based on recurrent neural networks. The input of the model is a vector of length [length missing]. The historical passenger flow sequence, the operation control input vector including departure intervals and route operation plans, and the external feature concatenation vector including holidays and weather are used to output the future data. Forecast values ​​of passenger flow demand at each station at each time step; Obtain the network topology and transfer probability matrix. Based on the network topology and transfer probability matrix, map the predicted passenger flow demand values ​​of each station to each line segment, and calculate the future time slices of each segment. Predicted cross-sectional passenger flow ; Based on the predicted cross-sectional passenger flow With the maximum capacity of the section Construct capacity constraints: in: Let i be the maximum capacity of segment i under a given group length and allowable load factor; In the construction of the network collaborative optimization model, the established optimization objective function Represented as: in: The total energy consumption of the network is obtained by summing the net energy consumption of all trains in the network across all sections. The total carbon emissions of the network are calculated by summing the section carbon emissions of all trains in all sections of the network. The comprehensive service quality indicators include average travel time, average waiting time, and passenger load factor deviation. , , These are the weighting coefficients for energy consumption, carbon emissions, and service quality, respectively. Incorporating intersection signal phase information into a unified calculation framework, including: The line is calculated by accumulating the travel time of each section. Vehicles on the road arrive at the intersection Time variables ; Leading to the intersection Green light window start time and the end time Construct coupling constraints between vehicle arrival time and green light window; In the optimization objective function Add SPaT penalty item This allows vehicles to arrive outside the green light window, and penalizes the amount of deviation. in: These are the signal coordination weighting coefficients.

2. The operational optimization method for coordinated carbon reduction of intelligent rail transit networks according to claim 1, characterized in that, The calculation of net vehicle energy consumption for each route section based on the vehicle dynamics model and the standardized operational dataset specifically includes: Based on the segment speed in the standardized running dataset acceleration Line gradient and vehicle equivalent total mass Calculate the rolling resistance during vehicle operation. air resistance Slope resistance and acceleration resistance ; The total traction resistance is calculated based on the individual resistance components. And combined with traction system efficiency Calculate traction power : Identify vehicle braking conditions and adjust braking power accordingly. With energy feedback efficiency Calculate the feedback energy and analyze the segment's operating time interval. Integrate to obtain the net energy consumption of the vehicle in the aforementioned section. : in: Indicates the braking time interval.

3. The operational optimization method for coordinated carbon reduction of intelligent rail transit networks according to claim 2, characterized in that, In the execution of rolling closed-loop control, real-time monitoring of actual passenger flow and operational deviations specifically includes: Calculate any segment within each rolling cycle. In time slice Actual passenger flow With predicted passenger flow Deviation between ; Determine the deviation amount Whether the amplitude continuously exceeds the preset ratio threshold, or the degree to which the current energy consumption and carbon emission indicators deviate from the threshold; If the triggering conditions are met, a local re-optimization procedure is initiated, which fine-tunes the departure intervals and operating speeds of lines or time periods with excessive deviations while keeping the operation schedules of unaffected lines unchanged.

4. The operation optimization method for coordinated carbon reduction of intelligent rail transit networks according to claim 3, characterized in that, This also includes displaying optimization results through a network carbon emission visualization interface, specifically including: The global overview area displays the total energy consumption and total carbon emissions of the power grid during the current statistical period, and shows their year-on-year or month-on-month trends relative to historical benchmarks. The online map view area displays the intelligent rail lines and sections in the form of geographic topology, and the heat map coloring of each section is based on the calculated net energy consumption or carbon emission intensity of the section to intuitively present the bottleneck sections with high energy consumption or high carbon emissions. The time-series curve view area displays the sequence curves of energy consumption and carbon emissions changing over time at the network or line level, and the triggered scheduling events are displayed as markers overlaid on the sequence curves.

5. An intelligent rail transit network collaborative carbon reduction operation optimization system, characterized in that, An operational optimization method for collaborative carbon reduction of a smart rail transit network as described in any one of claims 1-4 includes: The data acquisition and processing module is used to collect multi-source operational data of the intelligent rail transit network and perform standardized processing to form a standardized operational dataset; the multi-source operational data includes vehicle operation data, passenger flow data, intersection signal data and basic energy consumption data. The energy consumption and carbon emission calculation module is used to calculate the net energy consumption of vehicles in each line section based on the vehicle dynamics model and the standardized operation dataset, and to calculate the carbon emission of the section in combination with the power grid carbon emission factor, so as to establish a quantitative characterization of the network energy consumption and carbon emission. The passenger flow prediction module is used to build a network-level passenger flow prediction model based on historical passenger flow sequences and real-time passenger flow input, and generate prediction results of the spatiotemporal distribution of passenger flow at each station and section in the future. The network coordination optimization module is used to construct a network coordination optimization model. It incorporates the net energy consumption of vehicles, carbon emissions of sections, prediction results of passenger flow spatiotemporal distribution, and intersection signal phase information into a unified calculation framework. The optimization objective is to minimize the total energy consumption and total carbon emissions of the network while taking service quality into account. The optimized network operation map and intersection right-of-way strategy are obtained by solving the problem. The operation scheduling and execution module is used to perform rolling closed-loop control during actual operation, monitor the actual passenger flow and operational deviations in real time, and trigger local re-optimization when the deviation exceeds the preset threshold, generate the adjusted operation plan and issue it for execution.

6. The intelligent rail transit network collaborative carbon reduction operation optimization system according to claim 5, characterized in that, Also includes: A network carbon emission visualization module, connected to the energy consumption and carbon emission calculation module and the operation scheduling execution module, is used to generate a visual interactive interface; The visual interactive interface is configured to perform the following operations: The global overview area displays the total energy consumption and total carbon emissions of the network during the current statistical period; The online map view area displays the intelligent rail transit line in the form of geographic topology, and the heat map coloring of each section is based on the calculated net energy consumption or carbon emission intensity of the section. The time-series curve view area displays the sequence curves of grid energy consumption and carbon emissions over time, and overlays the triggered scheduling events as markers.

7. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which executes the instructions stored in the memory to perform the method as described in any one of claims 1-4.

Images