Expressway virtual power plant low-carbon resource scheduling method and device, equipment and medium

Abstract

The application discloses a highway virtual power plant low-carbon resource scheduling method and device, equipment and medium, relates to the technical field of resource scheduling, and comprises the following steps: forming a multi-source heterogeneous data set by collecting multiple types of data, building a highway virtual power plant point chain network aggregation structure, predicting traffic flow in combination with traffic and environmental data, relying on traffic flow and topological measurement to calculate the energy demand of each node, dynamically generating three types of scheduling boundaries of flexible load, adjustable resource and backup in combination with energy operation data, and carrying out low-carbon scheduling in multiple time scales in combination with market carbon data. Through the coupling relationship between traffic and energy data, the scheduling boundary is dynamically updated, the low-carbon, economic and high-speed energy supply service are taken into account, the new energy consumption and energy supply reliability are improved, and the complex highway operation scenes such as traffic flow fluctuation, congestion and holidays are adapted.

Description

Technical Field

[0001] This invention relates to the field of resource scheduling technology, and in particular to a low-carbon resource scheduling method, apparatus, equipment and medium for a virtual power plant on a highway. Background Technology

[0002] Currently, existing technologies for resource aggregation and scheduling of virtual power plants along highways mainly fall into the following categories: The first category constructs virtual power plants or chain-like microgrids based on distributed energy sources, energy storage systems, charging and swapping facilities, and infrastructure loads along highways. This forms a "point-chain-grid" energy control architecture suitable for the narrow road-area scenarios of highways, and to a certain extent considers distributed energy consumption, energy storage operation, charging load management, and power supply and distribution coordinated control. The second category primarily focuses on predicting electric vehicle charging loads at highway service areas or public charging stations. It uses traffic flow simulation, vehicle trajectory analysis, travel behavior modeling, queuing models, spatiotemporal feature extraction, or deep learning models to predict future trends in service area charging demand or charging load. The third category addresses source-load uncertainty and low-carbon operation requirements by employing methods such as scenario generation and reduction, stochastic optimization, distributed bar optimization, multi-timescale rolling optimization, and reinforcement learning to perform economic scheduling, low-carbon scheduling, or electricity-carbon coordinated scheduling of virtual power plants or microgrids. It also gradually introduces carbon emission constraints, electricity pricing mechanisms, ancillary services, and market interaction mechanisms.

[0003] However, existing technologies suffer from insufficient coupling between dynamic traffic flow forecasting and virtual power plant resource aggregation and scheduling on highways. They typically treat the energy demand of road nodes as ordinary power loads for forecasting and scheduling, rarely using traffic operation information as a direct basis for virtual power plant resource aggregation, scheduling boundary updates, and operational strategy adjustments. Existing technologies also fail to adequately characterize flexible load boundaries under traffic service constraints, often treating service area charging / swapping loads, HVAC loads, lighting loads, and tunnel ventilation loads as fixed loads or static flexible loads, lacking methods to dynamically determine the adjustable load range by combining traffic flow changes and traffic service constraints. Furthermore, existing technologies lack sufficient dynamic characterization of the adjustable resource boundaries and reserve demand boundaries of virtual power plants on highways, often using static resource capacity, fixed adjustable ratios, or empirical safety margins to describe resource regulation capabilities and reserve capacity requirements, making dynamic adjustments difficult based on scenarios such as holiday peaks, severe weather, and localized congestion. Finally, existing low-carbon scheduling methods struggle to balance economic efficiency, safety, and highway traffic service quality.

[0004] Therefore, how to effectively couple the dynamic forecast results of highway traffic flow with the virtual power plant resource aggregation and scheduling, so that the scheduling plan can reflect the impact of traffic flow changes on energy demand and resource regulation capacity in a timely manner, and achieve highway virtual power plant resource aggregation and scheduling that takes into account low carbon emissions, economy, safety and traffic service quality at multiple time scales such as day-ahead, intraday and real-time, has become an urgent problem to be solved. Summary of the Invention

[0005] The main objective of this application is to provide a low-carbon resource scheduling method, device, equipment, and medium for a virtual power plant on highways. The aim is to solve the technical problem of how to effectively couple the dynamic prediction results of highway traffic flow with the resource aggregation and scheduling of the virtual power plant, so that the scheduling plan can reflect the impact of traffic flow changes on energy demand and resource regulation capacity in a timely manner.

[0006] To achieve the above objectives, this application proposes a low-carbon resource scheduling method for a virtual power plant on highways, comprising: Acquire spatial topology data of highways, traffic operation data, energy operation data, energy facility distribution data, external environment data, and market and carbon emission related data to obtain a multi-source heterogeneous dataset; Based on the highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset, a point-chain-network aggregation structure for a highway virtual power plant is constructed. Based on the traffic operation data and external environment data of the multi-source heterogeneous dataset, the traffic flow of each road segment, service area and key node within the future set time domain is dynamically predicted to obtain the dynamic traffic flow prediction result. Based on the dynamic traffic flow prediction results and the point-chain network aggregation structure, the energy demand of each road node in the future set time domain is determined. The set of road node scheduling boundary parameters is dynamically determined based on the energy demand and energy operation data in the multi-source heterogeneous dataset. The set of road node scheduling boundary parameters includes flexible load boundary, resource adjustable boundary, and reserve demand boundary. Based on the set of road node scheduling boundary parameters and market and carbon emission related data in the multi-source heterogeneous dataset, low-carbon resource aggregation scheduling is performed at multiple time scales, including day-ahead, intraday, and real-time, to obtain the resource aggregation scheduling results of the highway virtual power plant.

[0007] In one embodiment, the step of constructing a point-chain network aggregation structure of a highway virtual power plant based on highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset includes: The location information of service areas, toll stations, tunnels, and charging / battery swapping stations are extracted from the spatial topology data of highways to obtain a set of road domain node locations; The distributed photovoltaic configuration information, energy storage device configuration information, charging pile configuration information, and battery swapping facility configuration information within each road node are extracted from the energy facility distribution data to obtain the node resource configuration set; Based on the set of road domain node locations and the set of node resource configurations, highway service areas, toll stations, tunnels, and charging / battery swapping stations with energy facilities are identified as road domain nodes, thus obtaining the set of road domain nodes. Based on the power connection and energy coordination relationships between adjacent road nodes in the road node set, the power exchange capacity and transmission capacity between adjacent nodes are determined, and the link parameter set is obtained. Based on the set of road domain nodes and the set of link parameters, a point-chain network aggregation structure for a highway virtual power plant is constructed, with road domain nodes as points, link parameters as chains, and multi-node collaborative control as the network.

[0008] In one embodiment, the step of dynamically predicting traffic flow for each road segment, service area, and key node within a future set time domain based on traffic operation data and external environment data from the multi-source heterogeneous dataset, and obtaining dynamic traffic flow prediction results, includes: Historical traffic flow data, real-time traffic flow data, average vehicle speed data, road occupancy data, and road topology data are extracted from the traffic operation data in the multi-source heterogeneous dataset to obtain traffic flow time series data; Holiday information data, meteorological data, and emergency data are extracted from the external environment data in the multi-source heterogeneous dataset to obtain traffic flow scenario data; Based on road topology data, service areas, toll stations, interchange hubs, and key road sections are identified as spatiotemporal prediction nodes. Weighted spatiotemporal association edges are constructed based on the road segment distance and traffic capacity between adjacent spatiotemporal prediction nodes to obtain the spatiotemporal graph structure of the highway. Based on the traffic flow time series data, the temporal traffic status of each road node is trended using a preset time series prediction model to obtain the node temporal trend characteristics. Based on the temporal trend characteristics of the nodes and the spatiotemporal graph structure of the highway, spatial correlation propagation and node state updates are performed through a preset graph neural network prediction model to obtain the spatial coupling characteristics of the nodes. Based on the traffic flow scenario data, the corresponding traffic flow correction coefficients are extracted from the preset scenario correction factor library, and the node spatial coupling features are subjected to scenario adaptive weighted correction to obtain the dynamic prediction result of traffic flow.

[0009] In one embodiment, the step of determining the energy demand of each road node in a future set time domain based on the dynamic traffic flow prediction results and the point-chain network aggregation structure includes: The predicted vehicle arrivals, vehicle type structure, vehicle dwell time and congestion status of each road node are extracted from the dynamic traffic flow prediction results to obtain the node traffic state parameters. Based on the predicted vehicle arrival volume and vehicle type structure in the node traffic state parameters, combined with the preset electric vehicle penetration rate parameters and the preset vehicle remaining battery power distribution parameters, the charging load demand and battery swapping load demand of each road node are determined, and the charging and battery swapping load prediction results are obtained. Based on the predicted vehicle arrivals in the node traffic state parameters and the meteorological data in the external environment data, the service area infrastructure load demand of each road node is determined, and the service area basic load prediction results are obtained. Based on the predicted traffic flow, average vehicle speed, and congestion status in the node traffic state parameters, the tunnel electromechanical load demand and road lighting load demand of each road node are determined, and the predicted results of tunnel and road electromechanical loads are obtained. Based on the predicted charging and swapping load, the predicted service area load, and the predicted tunnel and road electromechanical load, the energy demand of each road node in the future within a set time domain is generated.

[0010] In one embodiment, the step of dynamically determining the set of road node scheduling boundary parameters based on the energy demand and energy operation data in the multi-source heterogeneous dataset includes: The charging and swapping load demand, air conditioning and heating load demand, and lighting load demand of each road node are extracted from the energy demand to obtain the node flexible load demand set; Based on the set of flexible load requirements of the nodes and the constraints of vehicle waiting time, road illumination requirements, and tunnel ventilation safety requirements, the adjustable range of tunnel ventilation load, adjustable range of charging and swapping load, adjustable range of HVAC load, and adjustable range of lighting load for each road node are determined, and the flexible load boundary is obtained. Distributed photovoltaic power output data, energy storage charge status data, and charging and swapping facility status data of each road node are extracted from the energy operation data of the multi-source heterogeneous dataset to obtain node resource operation status data; Based on the node resource operation status data and flexible load boundary, the adjustable range of distributed photovoltaic, energy storage and load side of each road node are determined to obtain the resource adjustable boundary. Based on the predicted vehicle arrivals and event status data in the traffic flow dynamic prediction results, and the distributed photovoltaic output data in the energy operation data of the multi-source heterogeneous dataset, the node load fluctuation and energy output fluctuation are calculated through a preset uncertainty quantification model to obtain the node uncertainty parameters. Based on the node uncertainty parameters and the resource adjustability boundary, the uncertainty reserve requirements, event reserve requirements, and peak reserve requirements of each road domain node are determined, resulting in a set of road domain node scheduling boundary parameters.

[0011] In one embodiment, the step of performing low-carbon resource aggregation scheduling at multiple time scales (day-ahead, intraday, and real-time) based on the road domain node scheduling boundary parameter set and market and carbon emission-related data in the multi-source heterogeneous dataset to obtain the highway virtual power plant resource aggregation scheduling result includes: Based on the set of road zone node scheduling boundary parameters and market and carbon emission related data, a comprehensive scheduling objective function with operating cost, carbon emission cost and traffic service quality loss cost as optimization objectives is constructed before the operation date, and a day-ahead baseline scheduling plan is generated. During the operating day, real-time traffic operation data, real-time energy operation data, and real-time market and carbon emission related data are acquired according to a preset rolling cycle to obtain the intraday real-time operation data. Based on the intraday real-time operation data, the daily baseline scheduling plan is rolled over and revised, the set of scheduling boundary parameters for the road nodes in subsequent periods is re-determined, and the energy storage charging and discharging plan, charging and swapping power arrangement and flexible load response plan are adjusted to obtain the intraday revised scheduling plan; During the real-time operation phase, real-time traffic operation data and real-time energy operation data are collected in a preset short period and compared with the intraday revised scheduling plan to obtain real-time operation deviation data. The intraday revised scheduling plan is corrected in real time based on the real-time operational deviation data. By adjusting the energy storage output, charging and swapping power and flexible load adjustment, the resource aggregation and scheduling result of the highway virtual power plant is obtained.

[0012] In one embodiment, the step of acquiring highway spatial topology data, traffic operation data, energy operation data, energy facility distribution data, external environmental data, and market and carbon emission-related data to obtain a multi-source heterogeneous dataset includes: Traffic operation data is obtained by collecting highway topology, traffic flow, vehicle type structure, service area vehicle entry and exit volume and congestion status through roadside detection equipment and vehicle detectors. Energy operation data is obtained by collecting data on distributed photovoltaic output, energy storage charge status, charging and swapping facility operation status, and node power supply and distribution capacity at each road node through energy monitoring terminals. Meteorological data is collected through meteorological monitoring equipment, and holiday and weekday information data are obtained through a calendar interface to obtain external environmental data. By obtaining time-of-use electricity prices, demand response orders, and ancillary service prices through market data interfaces, and by obtaining carbon emission factor and carbon cost data through carbon emission databases, market and carbon emission related data can be obtained. The traffic operation data, energy operation data, external environment data, and market and carbon emission related data are time-aligned and format-standardized to obtain a multi-source heterogeneous dataset.

[0013] Furthermore, to achieve the above objectives, this application also proposes a low-carbon resource scheduling device for a highway virtual power plant, the highway virtual power plant low-carbon resource scheduling device comprising: The acquisition module is used to acquire highway spatial topology data, traffic operation data, energy operation data, energy facility distribution data, external environment data, and market and carbon emission related data to obtain multi-source heterogeneous datasets; The construction module is used to construct the point-chain network aggregation structure of the highway virtual power plant based on the highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset. The prediction module is used to dynamically predict the traffic flow of each road segment, service area and key node within a future set time domain based on the traffic operation data and external environment data of the multi-source heterogeneous dataset, and obtain the dynamic prediction result of traffic flow. The energy demand determination module is used to determine the energy demand of each road node in a future set time domain based on the dynamic traffic flow prediction results and the point-chain network aggregation structure. The scheduling boundary determination module is used to dynamically determine the set of scheduling boundary parameters for road nodes based on the energy demand and energy operation data in the multi-source heterogeneous dataset. The set of scheduling boundary parameters for road nodes includes flexible load boundary, resource adjustable boundary, and reserve demand boundary. The execution module is used to perform low-carbon resource aggregation scheduling at multiple time scales, including day-ahead, intraday, and real-time, based on the set of road domain node scheduling boundary parameters and market and carbon emission-related data in the multi-source heterogeneous dataset, to obtain the resource aggregation scheduling result of the highway virtual power plant.

[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the low-carbon resource scheduling method for the highway virtual power plant described above.

[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the highway virtual power plant low-carbon resource scheduling method described above.

[0016] This application constructs a high-speed virtual power plant point-chain network aggregation structure by collecting various types of data to form a multi-source heterogeneous dataset. It combines traffic and environmental data to predict traffic flow, calculates the energy demand of each node based on traffic flow and topology, and dynamically generates three types of scheduling boundaries—flexible load, adjustable resources, and reserves—in conjunction with energy operation data. It also conducts low-carbon scheduling at multiple time scales by integrating market carbon data. This application achieves a coupling relationship between traffic and energy data, dynamically updates scheduling boundaries, balances low-carbon and economical aspects with high-speed energy replenishment services, improves the reliability of new energy consumption and supply, and adapts to complex high-speed operation scenarios such as traffic fluctuations, congestion, and holidays. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the first embodiment of the low-carbon resource scheduling method for highway virtual power plants in this application. Figure 2 This is a flowchart illustrating the second embodiment of the low-carbon resource scheduling method for highway virtual power plants in this application. Figure 3 This is a schematic diagram of the module structure of the low-carbon resource scheduling device for the highway virtual power plant in this application. Figure 4 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the low-carbon resource scheduling method of the highway virtual power plant in this application embodiment.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0021] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0022] Highways are home to a large number of distributed photovoltaic, energy storage, charging and swapping devices, as well as flexible electromechanical loads in tunnels and service areas. Virtual power plants can enable unified management of these diverse energy resources, facilitating the low-carbon transformation of the transportation sector. Electricity demand along highways is highly influenced by traffic flow, weather, holidays, and congestion events. However, existing dispatching schemes only use traffic flow data as supplementary information for charging load forecasting, employ static parameters for dispatching boundaries, and fail to form a multi-scale collaborative dispatching closed loop driven by traffic flow. This makes it difficult to simultaneously address energy supply security, operating costs, carbon emissions, and vehicle refueling service guarantees.

[0023] Based on the above, this application also provides a low-carbon resource scheduling method for a highway virtual power plant, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the low-carbon resource scheduling method for a virtual power plant on highways in this application.

[0024] In this embodiment, the low-carbon resource scheduling method for the highway virtual power plant includes steps S10 to S60: Step S10: Obtain highway spatial topology data, traffic operation data, energy operation data, energy facility distribution data, external environment data, and market and carbon emission related data to obtain a multi-source heterogeneous dataset.

[0025] It should be noted that highway spatial topology data refers to the geographical location information of road network nodes such as service areas, toll stations, tunnels, and interchanges, as well as spatial structure information such as the connection relationships between nodes, road segment lengths, and the number of lanes. Traffic operation data refers to data reflecting the traffic operation status of highways, including historical traffic flow, real-time traffic flow, vehicle type structure, vehicle entry and exit volumes at service areas, average vehicle speed, road occupancy, congestion status, accident information, and construction information. Energy operation data refers to the real-time operation status data of energy systems at each road network node, including distributed photovoltaic power output, energy storage status of charge, energy storage charging and discharging power, charging pile operation status, battery swapping station operation status, comprehensive load of service areas, and electromechanical load of tunnels. Energy facility distribution data refers to information on the types, capacity, location, and connection relationships of energy facilities configured at each road network node along the highway, including distributed photovoltaic installed capacity, rated capacity of energy storage devices, number and power levels of charging piles, and configuration of battery swapping facilities. External environmental data refers to data on external conditions that affect traffic flow and energy supply and demand, including weather type, temperature, light intensity, wind speed, rainfall, holiday information, weekday information, special event information, and emergency information. Market and carbon emission related data refers to data related to electricity market operation and carbon emission management, including time-of-use pricing, demand response orders, ancillary service prices, carbon emission factors, carbon costs, and carbon budget constraint parameters.

[0026] Further, step S10 includes: First, collecting highway topology, traffic flow, vehicle type structure, service area vehicle entry / exit volume, and congestion status through roadside detection equipment and vehicle detectors to obtain traffic operation data. Specifically, the roadside detection equipment includes microwave detectors, video detectors, and geomagnetic detectors, deployed at various sections and key nodes along the highway. Microwave detectors detect vehicle presence and speed by emitting microwave signals and receiving reflected signals. Video detectors extract traffic flow and vehicle type structure information from video surveillance footage using image recognition algorithms. Geomagnetic detectors count traffic flow by detecting changes in the geomagnetic field caused by passing vehicles. Vehicle detectors are deployed at highway entrance and exit toll stations to obtain entry and exit traffic flow for each vehicle type. Service area vehicle entry / exit volume is collected by vehicle counting devices deployed at service area entrances and exits. Congestion status is identified by roadside video detectors combined with a preset congestion threshold (average vehicle speed below 20 km / h and road occupancy greater than 80%). The collected traffic operation data is transmitted to the virtual power plant dispatch system via roadside communication units at a preset upload cycle (5 minutes).

[0027] Secondly, energy operation data is obtained by collecting distributed photovoltaic power output, energy storage status of charge, charging and swapping facility operation status, and node power supply and distribution capacity at each road network node through energy monitoring terminals. Specifically, the energy monitoring terminals include photovoltaic inverter monitoring devices, energy storage management systems, charging pile controllers, and distribution automation terminals. The photovoltaic inverter monitoring devices read the real-time output power of distributed photovoltaics through the Modbus communication protocol. The energy storage management system obtains the status of charge and charging / discharging power of the energy storage devices through the battery management system interface. The charging pile controller obtains the operation status and charging power of each charging pile through the charging protocol. The distribution automation terminal obtains the node power supply and distribution capacity and load data through power line carrier communication. The energy operation data of each road network node is transmitted to the virtual power plant dispatch system through the energy monitoring terminals at a preset upload cycle (1 minute).

[0028] Next, meteorological data is collected through meteorological monitoring equipment, and holiday and workday information data are obtained through a calendar interface to obtain external environmental data. Specifically, the meteorological monitoring equipment includes temperature and humidity sensors, light intensity sensors, wind speed sensors, and rainfall sensors, deployed at meteorological monitoring stations along highways. The temperature and humidity sensors collect ambient temperature and relative humidity, the light intensity sensors collect solar irradiance data, the wind speed sensors collect wind speed and direction data, and the rainfall sensors collect rainfall intensity data. Holiday and workday information data are obtained by calling the national statutory holiday data interface, including holiday type, holiday date, and adjusted workday date. The meteorological data is transmitted to the virtual power plant dispatch system through the meteorological monitoring equipment at a preset upload cycle (10 minutes), and the holiday and workday information data are automatically updated at midnight every day through the calendar interface.

[0029] Then, time-of-use (TOU) electricity prices, demand response orders, and ancillary service prices are obtained through market data interfaces, and carbon emission factor and carbon cost data are obtained through carbon emission databases to obtain market and carbon emission-related data. Specifically, TOU electricity prices are obtained through the power trading center data interface, including peak-hour, normal-hour, and off-peak electricity prices and their corresponding time periods; demand response orders are obtained through the grid dispatch data interface, including response time, response capacity, and response price; and ancillary service prices are obtained through the power market operation system interface, including frequency regulation service prices and reserve service prices. Carbon emission factors are obtained through the national average carbon dioxide emission factor database for electricity published by the Ministry of Ecology and Environment, and carbon cost data is obtained through the carbon emission trading market data interface, including carbon emission quota prices and carbon trading costs. Market and carbon emission-related data are obtained and updated through the respective data interfaces before each daily operation date.

[0030] Finally, time alignment and format standardization were performed on traffic operation data, energy operation data, external environmental data, and market and carbon emission-related data to obtain a multi-source heterogeneous dataset. Specifically, time alignment was performed using the timestamps of the virtual power plant dispatch system as a benchmark, interpolating and aligning the four types of data according to a preset time resolution (5 minutes). Missing data was filled in using linear interpolation or forward imputation. Format standardization converted the four types of data into a preset JSON format, including fields such as data type identifier, road node number, timestamp, data item name, and data value, forming a structured multi-source heterogeneous dataset. The multi-source heterogeneous dataset is stored in the time-series database of the virtual power plant dispatch system for subsequent steps.

[0031] Step S20: Construct a point-chain network aggregation structure for a virtual power plant on a highway based on the highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset.

[0032] It should be noted that the point-chain-network aggregation structure refers to the hierarchical aggregation model of the virtual power plant on the highway. Here, a point represents a single road domain node, a chain represents the power connection, energy coordination or information communication relationship between adjacent road domain nodes, and a network represents the energy aggregation system along the highway composed of multiple road domain nodes and links.

[0033] It should be noted that step S20 includes: First, extracting service area location information, toll station location information, tunnel location information, and charging / swapping station location information from the highway spatial topology data to obtain a set of road domain node locations. The highway spatial topology data contains the latitude and longitude coordinates, station numbers, and relative distances to adjacent facilities. By parsing the geographic information fields in the spatial topology data, the latitude and longitude of the center point of each service area, the entrance and exit latitude and longitude of toll stations, the start and end points of tunnels, and the construction location latitude and longitude of charging / swapping stations are extracted to form a set of road domain node locations. Each node record in the road domain node location set includes a node number, node type (service area / toll station / tunnel / charging / swapping station), node name, and the latitude and longitude coordinates of the node's center point.

[0034] Secondly, distributed photovoltaic (PV) configuration information, energy storage device configuration information, charging pile configuration information, and battery swapping facility configuration information within each roadside node are extracted from the energy facility distribution data to obtain a node resource configuration set. The energy facility distribution data includes the type, capacity, quantity, and operational status of existing and under-construction energy facilities at each roadside node. By matching the node numbers in the energy facility distribution data with the node numbers in the roadside node location set, the distributed PV installed capacity, rated capacity and maximum charging / discharging power of energy storage devices, number of charging piles and power level per pile, and number of battery compartments and battery swapping power of battery swapping facilities within each roadside node are extracted to form a node resource configuration set. Each node record in the node resource configuration set includes the node number, PV installed capacity, rated energy storage capacity, maximum charging / discharging power of energy storage, number of charging piles, power per pile, number of battery compartments in battery swapping facilities, and battery swapping power.

[0035] Next, based on the set of road domain node locations and the set of node resource configurations, highway service areas, toll stations, tunnels, and charging / swapping stations with energy facility distributions are identified as road domain nodes, resulting in a set of road domain nodes. Each node in the set of road domain node locations is traversed, and the energy facility configuration information for the corresponding node is checked in the set of node resource configurations. If it exists, the node is marked as a valid road domain node; otherwise, it is marked as an invalid node and removed. Valid road domain nodes must meet preset minimum energy facility configuration conditions: distributed photovoltaic installed capacity greater than 0, or rated capacity of energy storage devices greater than 0, or number of charging piles greater than 0, or number of battery compartments for battery swapping facilities greater than 0. Through the above filtering, service areas, toll stations, tunnels, and charging / swapping stations that simultaneously possess geographical location information and energy facility configuration information are identified as road domain nodes, forming a set of road domain nodes.

[0036] Then, based on the power connection and energy coordination relationships between adjacent road domain nodes in the road domain node set, the power exchange capacity and transmission capacity between adjacent nodes are determined, resulting in a link parameter set. The power connection relationships are obtained by parsing the highway power supply and distribution network topology data, including the voltage level, line length, and conductor type of the distribution lines between adjacent road domain nodes. The rated current carrying capacity of the line is obtained by querying a preset conductor current carrying capacity table based on the conductor type, and the line transmission capacity is calculated in conjunction with the line voltage level. The energy coordination relationships are obtained by parsing the coordinated control strategies in the virtual power plant energy management system, including the maximum allowable power exchange value and power exchange direction constraints between adjacent road domain nodes. Each link record in the link parameter set includes the link number, starting node number, ending node number, line transmission capacity, maximum power exchange value, and power exchange direction constraints.

[0037] Finally, based on the set of road domain nodes and the set of link parameters, a point-chain network aggregation structure for a highway virtual power plant is constructed, with road domain nodes as points, link parameters as chains, and multi-node collaborative control as the network. Each road domain node in the set of road domain nodes is mapped as a vertex in the graph structure, and each link in the set of link parameters is mapped as an edge in the graph structure. The weight of the edge is determined by the smaller value between the line transmission capacity and the maximum power exchange value. On this basis, energy resources such as distributed photovoltaics, energy storage devices, charging piles, and battery swapping facilities within the road domain nodes are used as vertex attributes, node power supply and distribution capacity and node load demand are used as vertex constraints, and link transmission capacity and power exchange direction constraints are used as edge constraints, forming a weighted graph structure containing vertex attributes, vertex constraints, and edge constraints. Through the virtual power plant energy management system, multiple road domain nodes in the weighted graph structure are collaboratively controlled to realize power exchange, reserve sharing, and joint scheduling among nodes, thus constructing a highway virtual power plant point-chain network aggregation structure with road domain nodes as points, link parameters as chains, and multi-node collaborative control as the network.

[0038] Based on this, a cross-node power support and backup sharing mechanism is established. When real-time operational deviation data indicates that the target road area node experiences a sudden increase in charging and swapping load, limited node power supply capacity, energy storage charge status close to the lower limit, or insufficient backup capacity, cross-node power support, backup sharing, or charging and swapping power limit adjustments are executed according to the link transmission capacity, maximum power exchange value, and power exchange direction constraints between adjacent road area nodes.

[0039] Furthermore, a cross-node advance support mechanism driven by traffic flow propagation is established. Based on the predicted traffic volume, distance, and average speed of upstream road segments, the arrival time window for downstream road nodes in the future is determined. When there is concentrated traffic flow, holiday peaks, or accident detours on upstream road segments, the arrival pressure of downstream service areas or charging / swapping stations in the future is predicted based on the upstream vehicle arrival volume, average speed, and distance. If it is determined that a downstream node will experience concentrated energy replenishment demand, the reserve demand boundary of that node is increased in advance, and the energy storage, link exchangeable power, or external power purchase capacity of adjacent nodes are scheduled in advance to achieve cross-node power support and reserve sharing. In this example, the advance adjustment strategy for energy storage state of charge is as follows: when the downstream node is predicted to have a concentrated energy replenishment demand in the future, the reserve demand boundary of that node is increased in advance, and the energy storage, link exchangeable power, or external power purchase capacity of adjacent nodes are scheduled in advance to achieve cross-node power support and reserve sharing. When a sudden increase in charging load occurs after a certain period, it should be done in advance. During certain periods, the state of charge (SBC) of adjacent upstream nodes will be adjusted from the current value to a preset support SBC (0.5-0.7) to reserve discharge capacity. The strategy for advance adjustment of link exchangeable power is as follows: based on the predicted surge in charging load at downstream nodes and the discharge capacity of adjacent nodes' energy storage, adjustments will be made in advance. During this period, the link's switchable power will be increased from its current value to a preset supported power value (60%-80% of the link's transmission capacity). The strategy for adjusting reserve capacity configuration in advance is as follows: based on the predicted vehicle arrival volume at downstream nodes and the preset peak reserve capacity coefficient (1.2-1.5), adjustments will be made in advance. Increase the peak standby capacity configuration for downstream nodes and adjacent nodes during the time period.

[0040] Step S30: Based on traffic operation data and external environment data from multi-source heterogeneous datasets, dynamically predict the traffic flow of each road segment, service area and key node within the future set time domain, and obtain the dynamic prediction result of traffic flow.

[0041] It should be noted that step S30 includes: First, extracting historical traffic flow data, real-time traffic flow data, average vehicle speed data, road occupancy data, and road topology data from the traffic operation data in the multi-source heterogeneous dataset to obtain traffic flow time-series data. Historical traffic flow data is a sequence of traffic flow data collected from each road node within a preset time window (30 days) at a preset time resolution (5 minutes). Real-time traffic flow data is a sequence of traffic flow data within a preset time window (2 hours) prior to the current moment. Average vehicle speed data is the average speed of vehicles collected by roadside detection equipment within the corresponding time period. Road occupancy data is the proportion of time the road is occupied by vehicles within the corresponding time period. Road topology data includes road segment number, road segment start and end nodes, road segment length, number of lanes, and design capacity. The above data is organized according to node number and timestamp to form a multi-dimensional time-series data sequence for each road node, serving as traffic flow time-series data.

[0042] Secondly, holiday information data, meteorological data, and emergency data are extracted from the external environment data in the multi-source heterogeneous dataset to obtain traffic flow scenario data. Holiday information data includes holiday type (Spring Festival, National Day, etc.) and the number of days until the holiday; meteorological data includes weather type (sunny, rainy, snowy, foggy, etc.), temperature, light intensity, wind speed, and rainfall intensity; emergency data includes event type (traffic accidents, construction control, severe weather warnings, etc.), event location, and event level. These data are aligned with the traffic flow time-series data according to their timestamps to form traffic flow scenario data, which is used to characterize the external scenario conditions under which traffic flow operates in the current and future periods.

[0043] Next, service areas, toll stations, interchanges, and key road sections are identified as spatiotemporal prediction nodes based on road topology data. Weighted spatiotemporal association edges are then constructed based on the distance and capacity between adjacent spatiotemporal prediction nodes, resulting in the spatiotemporal graph structure of the highway. The determination of spatiotemporal prediction nodes is based on the following criteria: service areas have vehicle entry / exit and refueling needs; toll stations have traffic flow statistics functions; interchanges have traffic flow diversion and merging characteristics; and key road sections are traffic flow monitoring sections or frequently congested road sections. The weights of the weighted spatiotemporal association edges are calculated as follows: in Represents a node and nodes Weighted spatiotemporal correlation edge weights between them; Indicates the road segment capacity between adjacent nodes; Indicates the distance between adjacent nodes; This represents the preset normalization coefficient. A larger weight indicates a stronger spatial correlation between nodes. Using spatiotemporal prediction nodes as vertices of the graph structure and weighted spatiotemporal correlation edges as edges, an adjacency matrix is ​​constructed as follows: The node feature matrix is The spatiotemporal structure of highways .

[0044] Next, based on the traffic flow time-series data, a preset time-series prediction model is used to extract the temporal traffic state trend of each road node, obtaining the node's temporal trend features. In this example, the preset time-series prediction model uses a Long Short-Term Memory (LSTM) network. The LSTM model contains three gating structures: an input gate, a forget gate, and an output gate, which control the memory and forgetting of information through the gating mechanism. The historical traffic flow sequence of each road node is input into the LSTM model, and the input is mapped to a preset hidden dimension (64 dimensions) through the embedding layer. Temporal features are extracted through two layers of LSTM units, each containing 128 hidden units, with a dropout rate set to 0.2. The output layer is mapped to the prediction time domain length through a fully connected layer, obtaining the predicted traffic flow trend value of each road node in the future set time domain, which serves as the node's temporal trend feature.

[0045] Then, based on the temporal trend features of nodes and the spatiotemporal graph structure of the highway, spatial correlation propagation and node state updates are performed using a pre-defined graph neural network prediction model to obtain the spatial coupling features of nodes. In this example, the pre-defined graph neural network prediction model adopts a graph convolutional network (GCN). The GCN model extracts features from the graph structure data and updates the node states through graph convolutional layers. The temporal trend features of nodes are used as the initial node features of the highway spatiotemporal graph structure and input into a two-layer GCN model. Each layer of GCN performs graph convolution operations using the following formula: in Indicates the first The node feature matrix of the layer GCN; This represents the initial node feature matrix (i.e., the node temporal trend features); This represents the adjacency matrix with added self-loops. Represents the identity matrix. express The degree matrix; Indicates the first The learnable weight matrix of the layer; This represents the ReLU activation function. The first GCN layer has an output dimension of 128, and the second GCN layer has an output dimension of 64. Spatial association propagation between nodes and node state updates are achieved through graph convolution operations, resulting in node spatial coupling features.

[0046] Finally, based on the traffic flow scenario data, the corresponding traffic flow correction coefficients are extracted from the preset scenario correction factor library. Scenario-adaptive weighted correction is then applied to the node spatial coupling features to obtain dynamic traffic flow prediction results. The preset scenario correction factor library includes five scenarios: regular scenarios, holiday peak scenarios, severe weather scenarios, accident and congestion scenarios, and construction control scenarios. Each scenario corresponds to a set of traffic flow correction coefficients (range 0.5-2.0). The correction coefficients are determined based on historical data statistical analysis. For example, the correction coefficient for holiday peak scenarios is 1.5 (indicating a 50% increase in traffic flow compared to regular scenarios), and the correction coefficient for severe weather scenarios is 0.7 (indicating a 30% decrease in traffic flow compared to regular scenarios). Based on the holiday information, weather type, and event type in the traffic flow scenario data, the scenario classification in the preset scenario correction factor library is matched, and the corresponding traffic flow correction coefficients are extracted. Weighted correction is then applied to the node spatial coupling features. in For nodes During the period Traffic flow dynamic prediction results; In representing the spatial coupling characteristics of nodes During the period eigenvalues; This represents the traffic flow correction coefficient. Through scenario-adaptive weighted correction, dynamic traffic flow prediction results are obtained that adapt to different external scenario conditions. The dynamic traffic flow prediction results specifically include predicted vehicle arrivals, vehicle type structure, vehicle dwell time, congestion status, and event status.

[0047] Furthermore, the traffic flow driving boundary adjustment coefficient is determined based on the set of traffic flow driving parameters. When the predicted vehicle arrival volume increases, the predicted vehicle dwell time decreases, the queuing risk increases, or the event status indicates congestion, accidents, severe weather, construction control, etc., the range of downward adjustment of charging and swapping load is narrowed, the minimum allowable power of charging and swapping load is increased, the reserve demand boundary is raised, and the priority of adjacent road nodes participating in power support is increased. When the traffic flow is stable, the predicted vehicle dwell time is long, and the queuing risk is low, the range of transferable charging and swapping load and the flexible load adjustment space are expanded to improve the distributed new energy consumption capacity.

[0048] Specifically, the flexible load boundary reconfiguration is determined based on predicted vehicle arrivals, predicted vehicle dwell times, queuing risk, and event status. When predicted vehicle arrivals exceed the historical average, predicted vehicle dwell times are lower than the historical average, queuing risk increases, or event status indicates congestion, accidents, severe weather, or construction control, the minimum allowable power of the charging and swapping load increases from 60% to 80%-90% of the rated power, and the range for downward adjustment of the charging and swapping load is correspondingly reduced. When traffic flow is stable, predicted vehicle dwell times exceed the historical average, and queuing risk is low, the charging and swapping load can delay charging periods from the current period to 2-3 future periods, and the range for transferable charging and swapping load is correspondingly expanded. The adjustable range of HVAC and lighting loads is adjusted synchronously according to changes in service area vehicle entry volume and personnel scale. When vehicle entry volume increases, the minimum allowable power of HVAC load increases by 10%-20%, and the minimum allowable power of lighting load increases by 5%-15%.

[0049] Resource adjustable boundary reconfiguration is determined based on predicted vehicle arrivals, service area vehicle entry volumes, and event status. When predicted vehicle arrivals exceed 80% of the historical maximum arrivals, or when event status indicates congestion, accidents, or other similar situations, the energy storage discharge priority of adjacent road nodes is upgraded from normal to priority, and the link's exchangeable power is increased from the current value to 70%-90% of the link's transmission capacity. When traffic flow is stable and predicted vehicle arrivals are below 50% of the historical maximum arrivals, the energy storage discharge priority of adjacent road nodes reverts to normal, the link's exchangeable power is restored to 40%-60% of the link's transmission capacity, and the adjustable energy storage range is prioritized for local renewable energy consumption, with the proportion of distributed photovoltaic power generation increasing from 60% to 80%-90%.

[0050] The reserve demand boundary is reconstructed based on predicted vehicle arrivals, congestion status, and event status. When predicted vehicle arrivals exceed 80% of the historical maximum arrivals, congestion status is congested or severe congestion, or event status indicates accidents, severe weather, or other similar conditions, the uncertainty reserve demand, event reserve demand, and peak reserve demand are increased by amplification factors of 1.2–2.0, respectively. When traffic flow is stable, congestion status is smooth, and event status is normal, all types of reserve demand are restored to the baseline value or reduced by a reduction factor of 0.8–1.0.

[0051] Step S40: Based on the dynamic traffic flow forecast results and the point-chain network aggregation structure, determine the energy demand of each road node in the future set time domain.

[0052] It should be noted that energy demand refers to the total power load required for the operation of each node of the highway in a future set time domain, including charging and swapping load, service area infrastructure load, tunnel electromechanical load, road lighting load, and monitoring and communication load, etc.

[0053] Further, step S40 includes: First, extracting the predicted vehicle arrivals, vehicle type structure, vehicle dwell time, and congestion status of each road node from the traffic flow dynamic prediction results to obtain node traffic state parameters. The predicted vehicle arrivals are the total number of vehicles predicted to arrive at each road node within a future set time domain, statistically analyzed according to a preset time resolution (1 hour). Vehicle type structure includes types such as small passenger cars, large passenger cars, small trucks, medium-sized trucks, and large trucks, expressed as the proportion of each type. Vehicle dwell time is the average dwell time of predicted vehicles within the service area of ​​the road node after arrival, determined through historical data statistical analysis. Congestion status is the predicted congestion level of adjacent road segments of each road node, divided into four levels: smooth traffic, slow traffic, congested, and severe congestion, determined by comparing the road occupancy data in the traffic flow dynamic prediction results with a preset congestion judgment threshold (road occupancy greater than 80% is considered congested).

[0054] Secondly, based on the predicted vehicle arrival volume and vehicle type structure in the node traffic state parameters, combined with the preset electric vehicle penetration rate parameters and the preset vehicle remaining battery power distribution parameters, the charging load demand and battery swapping load demand of each road node are determined, resulting in the predicted charging and battery swapping load. The preset electric vehicle penetration rate parameter is determined based on historical data statistical analysis, with a value range of 0.1-0.5 (i.e., 10%-50%). Different vehicle types correspond to different penetration rates; for example, small passenger cars have a higher penetration rate (0.3-0.5), while large trucks have a lower penetration rate (0.1-0.2). The preset vehicle remaining battery power distribution parameter is determined based on charging behavior survey data, using a normal distribution model, with a mean of 0.3 (i.e., 30%) and a standard deviation of 0.1. The charging load demand is calculated using the following formula: in Represents a node During the period The predicted charging demand; Represents a node During the period The predicted arrival of the first Number of vehicles of this type; Indicates the first Electric vehicle penetration rate for this type of vehicle; Indicates the first Vehicles at nodes The probability of charging occurring; Indicates the first Battery capacity of this type of vehicle; Indicates the first Target state of charge for similar vehicles; Indicates the first Vehicles arrive at the node Time period The state of charge; Indicates the first For vehicles of this type, the single-vehicle charging demand is set to 0 if the state of charge upon arrival has reached or exceeded the target charging state of charge. The battery swapping load demand is estimated based on the proportion of vehicles with battery swapping needs among the predicted vehicle arrivals (preset battery swapping proportion 0.05-0.1) and the battery swapping capacity per vehicle (preset 50-80 kWh).

[0055] Next, based on the predicted vehicle arrival volume in the node traffic state parameters and the meteorological data in the external environment data, the service area infrastructure load demand of each road node is determined, resulting in the service area basic load prediction results. The service area infrastructure load includes lighting load, HVAC load, commercial electricity load, and water supply and drainage load. Lighting load is estimated based on the predicted vehicle arrival volume and service area area, calculated according to preset unit area lighting power and service area building area. HVAC load is estimated based on temperature and humidity data from the meteorological data, combined with the population size determined by the predicted vehicle arrival volume, using a preset air conditioning load calculation model (cooling load coefficient method). The summer cooling load coefficient is taken as 0.08-0.12 kW / person, and the winter heating load coefficient is taken as 0.05-0.08 kW / person. Commercial electricity load and water supply and drainage load are estimated based on the predicted vehicle arrival volume and preset per capita commercial electricity consumption index (0.3-0.5 kW / person) and per capita water consumption index (0.01-0.02 cubic meters / person).

[0056] Then, based on the predicted traffic flow, average vehicle speed, and congestion status parameters of each road node, the tunnel electromechanical load demand and road lighting load demand are determined, resulting in predicted tunnel and road electromechanical loads. Tunnel electromechanical loads include tunnel lighting load, tunnel ventilation load, and tunnel monitoring load. Tunnel lighting load is calculated based on tunnel length, predicted traffic flow, and a preset lighting power density (5-10 watts / square meter), with different lighting levels used during the day and night. Tunnel ventilation load is determined based on tunnel length, predicted traffic flow, average vehicle speed, and congestion status. Ventilation load increases by 30%-50% under congested conditions compared to uncongested conditions, and is estimated using a preset ventilation load calculation model (demand volume method). Road lighting load is calculated based on road segment length, predicted traffic flow, and a preset road lighting power (200-400 watts / light), with nighttime lighting load controlled in stages according to traffic volume. Monitoring load is estimated based on the number of monitoring devices at each road node and the power of a single device (50-100 watts).

[0057] Finally, based on the predicted charging and battery swapping loads, the predicted service area loads, and the predicted tunnel and road electromechanical loads, the energy demand for each road node in the future within a specified time period is generated. The charging load demand, battery swapping load demand, service area lighting load, HVAC load, commercial electricity load, water supply and drainage load, tunnel lighting load, tunnel ventilation load, tunnel monitoring load, and road lighting load for each road node in the corresponding time period are summed to obtain the total energy demand for each road node in the future within the specified time period.

[0058] Step S50: Dynamically determine the set of boundary parameters for road node scheduling based on energy demand and energy operation data in the multi-source heterogeneous dataset.

[0059] It should be noted that the set of boundary parameters for road zone node scheduling includes flexible load boundaries, resource adjustable boundaries, and reserve demand boundaries. Flexible load boundaries refer to the power range within a set time period that can be adjusted upwards, downwards, transferred, or interrupted for adjustable loads such as charging / swapping loads, HVAC loads, lighting loads, and tunnel ventilation loads at each road zone node, under the conditions of meeting vehicle energy replenishment service needs, road safety operation requirements, and equipment operation constraints. Resource adjustable boundaries refer to the range of upward adjustment capabilities, downward adjustment capabilities, peak shifting capabilities, and power support capabilities that each road zone node can provide within a set time period, under the conditions of meeting constraints on power supply security, equipment capacity, energy storage state of charge, and charging / swapping service capabilities. Reserve demand boundaries refer to the range of reserve capacity that needs to be reserved to ensure the safe and stable operation of the highway virtual power plant under the influence of uncertainties such as traffic flow forecast deviations, renewable energy output fluctuations, sudden increases in charging loads, and equipment failures.

[0060] Further, step S50 includes: First, extracting the charging / swapping load demand, HVAC load demand, and lighting load demand for each road segment node from the energy demand, to obtain a set of flexible load demands for the nodes. The charging / swapping load demand is the sum of the charging load demand and the swapping load demand calculated in step four. The HVAC load demand is determined based on the service area building area, the preset air conditioning load index per unit area (80-120 watts / square meter), and temperature data from meteorological data. The lighting load demand is determined based on the lighting area of ​​the service area, toll station, and tunnel, the preset lighting power density per unit area (10-20 watts / square meter), and day / night conditions. The above three types of load demand are organized according to road segment nodes and time periods to form a set of flexible load demands for the nodes.

[0061] Secondly, based on the set of flexible load demands at each node, vehicle waiting time constraints, road illumination requirements, and tunnel ventilation safety requirements, the adjustable ranges of tunnel ventilation load, charging / swapping load, HVAC load, and lighting load for each road node are determined, thus obtaining the flexible load boundaries. The adjustable range of tunnel ventilation load is determined based on tunnel length, predicted traffic volume, average vehicle speed, congestion status, and tunnel ventilation safety requirements. Tunnel ventilation load includes jet fan ventilation load and shaft ventilation load. When the ventilation load is reduced, it is necessary to ensure that the carbon monoxide concentration in the tunnel does not exceed the preset threshold (250 ppm) and the smoke concentration does not exceed the preset threshold (0.0075 ppm). When the real-time monitored concentration approaches 90% of the threshold, the downward adjustment stops. When the ventilation load is increased, it is determined based on the rated power of the fan and the current number of operating fans. The upward adjustment range is the difference between the current operating power and the rated total power. The vehicle waiting time constraint is the preset maximum waiting time (15-30 minutes). When the predicted vehicle queuing waiting time exceeds this constraint, the downward adjustment range of the charging / swapping load is limited. Road illuminance requirements are determined according to the "Highway Lighting Design Standard" (JTG / T D70 / 2-01). The maintained illuminance of the main line lighting on highways shall not be less than 10-20 lux. When the illuminance is lower than this requirement, the downward adjustment range of the lighting load is limited. Tunnel ventilation safety requirements are determined according to the "Highway Tunnel Ventilation Design Specification" (JTG / T D70 / 2-02). The carbon monoxide concentration in the tunnel shall not exceed the preset threshold (250 ppm). When the concentration approaches this threshold, the downward adjustment range of the ventilation load is limited. The adjustable range of the charging / swapping load is determined by the following formula: in Represents a node During the period The power dispatching capacity of the charging and swapping load; This indicates the minimum allowable charging power (taking 50%-80% of the rated power, considering vehicle waiting time constraints). This indicates the maximum allowable charging power (considering charging pile capacity constraints, it is taken as 100% of the rated power). The adjustable range of HVAC load is determined based on the indoor and outdoor temperature difference and personnel comfort requirements, with a downward adjustment range of 20%-40% of the rated load and an upward adjustment range of 10%-20% of the rated load. The adjustable range of lighting load is determined based on road illuminance requirements and day / night conditions, with a downward adjustment range of 30%-50% of the rated load (during off-peak hours at night) and an upward adjustment range of 0%-10% of the rated load.

[0062] Next, distributed photovoltaic (PV) output data, energy storage state of charge (SOC) data, and charging / swapping facility status data for each road node are extracted from the energy operation data in the multi-source heterogeneous dataset to obtain node resource operation status data. Distributed PV output data consists of real-time output power and short-term output prediction values ​​collected by the PV inverters at each road node. Energy storage SOC data consists of the current SOC value (range 0-1) and preset upper and lower limits (lower limit 0.2, upper limit 0.9) collected by the energy storage management system. Charging / swapping facility status data consists of the number of available charging piles, their occupancy status, and fault status collected by the charging pile controller.

[0063] Next, based on the node resource operation status data and flexible load boundaries, the adjustable ranges of distributed photovoltaic (PV), energy storage, and load-side components for each road network node are determined, resulting in the resource adjustable boundaries. The adjustable range of distributed PV is determined based on the predicted PV output and curtailment limits; the upward adjustment range is limited by the rated capacity of the PV inverter, while the downward adjustment range is limited by the minimum PV output (0 or tracking output). The adjustable range of energy storage is determined using the following formula: in Represents a node During the period Discharge capability; Represents a node During the period Rechargeable capability; and These are the rated maximum discharge power and charging power of the energy storage device, respectively. and These are discharge efficiency and charging efficiency, respectively. Indicates the rated capacity of energy storage; Indicates the current state of charge; and These are the lower and upper limits of the state of charge, respectively; This indicates the length of the scheduling period. The adjustable range on the load side is the adjustable range within the flexible load boundary.

[0064] Based on the predicted vehicle arrivals and event status data from the traffic flow dynamic prediction results, and the distributed photovoltaic output data from the energy operation data in the multi-source heterogeneous dataset, the node load fluctuation and energy output fluctuation are calculated using a preset uncertainty quantification model to obtain the node uncertainty parameters. In this example, the preset uncertainty quantification model uses the Monte Carlo simulation method, which calculates the standard deviation of node load fluctuation and energy output fluctuation by randomly sampling the predicted vehicle arrivals and photovoltaic output (sampling times 1000). Event status data includes status indicators such as holiday peaks, accidents, congestion, severe weather, and construction control. When the event status indicator is 1, the standard deviation of load fluctuation is increased by a preset event fluctuation coefficient (1.2-1.5).

[0065] Then, based on the node uncertainty parameters and resource adjustability boundaries, the uncertain reserve requirements, event reserve requirements, and peak reserve requirements for each road domain node are determined, resulting in the set of road domain node scheduling boundary parameters. The uncertain reserve requirements are calculated using the following formula: in This indicates uncertain backup demand; This represents the confidence coefficient, which is determined based on historical prediction error statistics. The standard deviation of node load fluctuation; This represents the standard deviation of the node's energy output fluctuation. Event reserve requirements are determined based on event status data; when the event status flag is 1, Otherwise, it is 0. The preset event backup capacity is determined based on 10%-20% of the node's supply capacity.

[0066] Peak-hour reserve demand is determined by comparing the predicted vehicle arrivals with a preset peak threshold (80% of the historical maximum arrivals). When the predicted arrivals exceed this threshold, Otherwise, it is 0. Pre-set peak reserve capacity (determined based on 15%-25% of the node's supply capacity). Total reserve requirement. for: The flexible load boundary, resource adjustable boundary, and reserve demand boundary are organized according to road domain nodes and time periods to obtain the set of road domain node scheduling boundary parameters.

[0067] Step S60: Based on the set of road node scheduling boundary parameters and market and carbon emission related data in the multi-source heterogeneous dataset, perform low-carbon resource aggregation scheduling at multiple time scales, including day-ahead, intraday, and real-time, to obtain the highway virtual power plant resource aggregation scheduling result.

[0068] It should be noted that day-ahead scheduling refers to the optimization process of generating a baseline operating plan based on data such as traffic flow forecasts, load forecasts, renewable energy output forecasts, electricity price information, and carbon emission factors, with the target of the next day or a set time domain, before the operating day. Intraday scheduling refers to the optimization process of continuously revising the day-ahead baseline scheduling plan based on changes in real-time traffic flow, actual load, energy storage status, renewable energy output, and market signals within the operating day. Real-time correction scheduling refers to the optimization process of adjusting the virtual power plant operating plan in real time based on changes in traffic flow, charging load, equipment status, and scheduling execution deviations collected at minute-level or shorter intervals during the real-time operation phase.

[0069] Specifically, during the day-ahead phase, based on the set of road node scheduling boundary parameters and market and carbon emission related data, a comprehensive scheduling objective function is constructed with operating costs, carbon emission costs, and traffic service quality loss costs as optimization objectives. This generates a day-ahead baseline scheduling plan. The traffic service quality loss costs include one or more of the following: penalty costs for failing to meet minimum vehicle refueling needs, penalty costs for average vehicle waiting time exceeding a threshold, penalty costs for insufficient service area comfort, penalty costs for failing to meet road lighting safety constraints, and penalty costs for failing to meet tunnel ventilation safety constraints. During the intraday phase, real-time traffic operation data, real-time energy operation data, and real-time market and carbon emission related data are acquired according to a preset rolling cycle. The day-ahead baseline scheduling plan is then rolled over and revised, the set of road node scheduling boundary parameters for subsequent time periods is redefined, and the energy storage charging and discharging plan, charging and swapping power arrangement, and flexible load response plan are adjusted to obtain the intraday revised scheduling plan. During the real-time phase, based on real-time traffic operation data and real-time energy operation data collected over a preset short cycle, the deviation is compared with the intraday revised scheduling plan. Real-time correction is performed by adjusting energy storage output, charging and swapping power, and flexible load adjustment to obtain the highway virtual power plant resource aggregation scheduling result.

[0070] Based on this, the dispatch results are sent to energy storage devices, charging and swapping facilities, distributed photovoltaic systems, flexible load control terminals, or node energy management terminals for execution. Energy storage devices, according to the energy storage charging and discharging power and time period instructions in the dispatch results, control the battery charging and discharging power and direction through the energy storage management system. When the instruction is to discharge, the energy storage system outputs power to the load at the road node or adjacent nodes. When the instruction is to charge, the energy storage system absorbs power from distributed photovoltaic systems or the external power grid and provides real-time feedback on the energy storage's state of charge to the virtual power plant dispatch system. Charging and swapping facilities, according to the charging and swapping power arrangements and time period instructions in the dispatch results, adjust the charging power and charging time period through the charging pile controller or swapping station management system. When the instruction is to reduce power, the charging power is reduced or the charging start time is delayed. When the instruction is to increase power, the charging power is increased within the rated capacity range of the charging pile, and the charging pile's operating status and vehicle charging progress are provided real-time feedback to the virtual power plant dispatch system. Distributed photovoltaic (PV) systems, based on the PV consumption plan in the dispatch results, control the output power and grid connection mode of PV inverters. When the node load is low and the energy storage is fully charged, they adjust the inverter output or supply excess power to adjacent nodes to avoid curtailment, and provide real-time PV output data to the virtual power plant dispatch system. Flexible load control terminals, based on the flexible load adjustment amount and time period instructions in the dispatch results, adjust the operating power of corresponding equipment through HVAC controllers, lighting controllers, or tunnel ventilation controllers. When the instruction is to reduce load, they reduce the equipment operating power or adjust the operating period; when the instruction is to increase load, they increase the operating power within the equipment's rated capacity and traffic service safety constraints, and provide real-time feedback of equipment operating status and load data to the virtual power plant dispatch system. Node energy management terminals, based on the inter-node collaborative power and reserve capacity configuration instructions in the dispatch results, coordinate the operating status of various energy resources within their node. When adjacent nodes request power support, they control the discharge of the node's energy storage or increase the purchased power through the node energy management terminal, supplying power to adjacent nodes, and providing real-time feedback of node power balance status and energy flow data to the virtual power plant dispatch system.

[0071] This embodiment collects various types of data to form a multi-source heterogeneous dataset, constructs a high-speed virtual power plant point-chain network aggregation structure, combines traffic and environmental data to predict traffic flow, calculates the energy demand of each node based on traffic flow and topology, and dynamically generates three types of scheduling boundaries—flexible load, adjustable resources, and reserves—in conjunction with energy operation data. It also incorporates market carbon data to conduct low-carbon scheduling at multiple time scales. This application achieves a coupling relationship between traffic and energy data, dynamically updates scheduling boundaries, balances low-carbon and economical aspects with high-speed energy replenishment services, improves the reliability of new energy consumption and supply, and adapts to complex high-speed operation scenarios such as traffic fluctuations, congestion, and holidays.

[0072] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 The low-carbon resource scheduling method for the virtual power plant on highways further includes steps S201-S205 in step S60: Step S201: Based on the set of boundary parameters for road node scheduling and market and carbon emission related data, construct a comprehensive scheduling objective function with operating cost, carbon emission cost and traffic service quality loss cost as optimization objectives before the operation date, and generate a day-ahead baseline scheduling plan.

[0073] Specifically, the comprehensive scheduling objective function is: in This refers to the overall operational cost within the scheduling cycle; This indicates that the optimization objective is to minimize the overall operational cost. Indicates time period The cost of purchasing and selling electricity; Indicates time period The operating cost of energy storage; Indicates time period The costs of unused solar, wind, or other renewable energy sources; Indicates time period The cost of flexible load adjustment or demand response; Indicates time period The carbon emission costs; Indicates time period The penalty costs for poor traffic service quality, such as penalties caused by excessively long vehicle waiting times, failure to meet minimum vehicle refueling requirements, and insufficient comfort in service areas; Indicates time period The costs of configuring standby capacity or the penalty costs for insufficient standby are calculated as follows: Electricity purchase and sale costs are calculated based on time-of-use pricing and purchased electricity; energy storage operation costs are calculated based on the number of charge / discharge cycles and capacity loss coefficient (0.001-0.005 yuan / kWh); curtailment costs are calculated based on the amount of curtailed solar power and the photovoltaic grid connection price; flexible load adjustment costs are calculated based on the adjusted electricity and a preset unit adjustment cost (0.1-0.5 yuan / kWh); carbon emission costs are calculated based on purchased electricity, carbon emission factors, and carbon prices; traffic service quality penalty costs are calculated based on the duration of vehicle waiting time exceeding a preset threshold (15 minutes) and a preset unit penalty cost (5-10 yuan / minute); and standby capacity configuration costs are calculated based on standby capacity and a preset unit standby cost (0.05-0.2 yuan / kWh). Constraints include flexible load boundary constraints, resource adjustability boundary constraints, and standby demand boundary constraints in the road node scheduling boundary parameter set, as well as node power balance constraints, energy storage state of charge constraints, and inter-node link transmission capacity constraints. The above optimization problem is solved by a mixed-integer linear programming solver (such as CPLEX or Gurobi) to obtain the day-ahead baseline scheduling plan.

[0074] Step S202: During the operating day, real-time traffic operation data, real-time energy operation data, and real-time market and carbon emission related data are acquired according to a preset rolling cycle to obtain the intraday real-time operation data.

[0075] It should be noted that real-time traffic operation data includes actual traffic flow, vehicle arrivals, vehicle type composition, and congestion status for the current period, collected via roadside detection equipment and vehicle detectors at a preset collection frequency (5 minutes). Real-time energy operation data includes actual output of distributed photovoltaic systems, current state of charge of energy storage, actual operating power of charging and swapping facilities, and actual load of nodes, collected via energy monitoring terminals at a preset collection frequency (1 minute). Real-time market and carbon emission related data includes actual electricity prices, demand response orders, and carbon prices for the current period, obtained via market data interfaces at a preset update frequency (15 minutes).

[0076] Step S203: Based on the real-time operation data within the day, the daily baseline scheduling plan is rolled out and revised. The set of boundary parameters for the scheduling of road nodes in subsequent periods is redefined, and the energy storage charging and discharging plan, the charging and swapping power arrangement, and the flexible load response plan are adjusted to obtain the intraday revised scheduling plan.

[0077] Specifically, the real-time intraday operational data is compared with the predicted data for the corresponding time period in the previous day's baseline scheduling plan to calculate traffic flow prediction deviation and load prediction deviation. When the deviation between the actual traffic volume and the predicted value exceeds the preset rolling correction threshold (20%), the traffic flow dynamic prediction in step three and the energy demand determination and scheduling boundary determination processes in steps four to six are re-executed to update the road node scheduling boundary parameter set for subsequent time periods. Based on this, with the remaining time period as the optimization cycle, the updated road node scheduling boundary parameter set as the constraint, and minimizing the comprehensive operating cost of the remaining time period as the objective, a mixed-integer linear programming solver is used to re-solve the problem, adjusting the energy storage charging and discharging plan, charging and swapping power arrangement, and flexible load response plan to obtain the intraday revised scheduling plan.

[0078] Step S204: During the real-time operation phase, real-time traffic operation data and real-time energy operation data collected in a preset short cycle are compared with the daily revised scheduling plan to obtain real-time operation deviation data.

[0079] It should be noted that real-time operational deviation data includes sudden increases in charging load, sudden drops in photovoltaic output, offsets of energy storage state of charge approaching constraint boundaries, and node power supply capacity limitations. When any deviation in the real-time operational deviation data exceeds a preset real-time correction threshold (10%), the real-time correction process is triggered.

[0080] Step S205: Based on real-time operational deviation data, the intraday revised scheduling plan is corrected in real time. By adjusting the energy storage output, charging and swapping power, and flexible load adjustment, the resource aggregation and scheduling results of the highway virtual power plant are obtained.

[0081] Specifically, when the charging load suddenly increases, priority is given to utilizing the energy storage discharge reserve and flexible load reduction space. If this is still insufficient, the purchased power is increased. When the photovoltaic output drops sharply, priority is given to utilizing the energy storage discharge reserve and flexible load reduction space. If this is still insufficient, the purchased power is increased and demand response is initiated. When the energy storage state of charge approaches the constraint boundary, the energy storage charging and discharging power is adjusted to keep it within a safe operating range. When the node's power supply capacity is limited, power support is provided through transmission between adjacent nodes or the charging and swapping power limit is adjusted. Real-time correction is solved quickly through a preset rule engine or a preset optimization model (such as a linear programming model). The solution time is controlled within a preset time limit (1 minute). The output of distributed photovoltaic absorption power, energy storage charging and discharging power, charging and swapping facility power, flexible load adjustment, inter-node collaborative power, reserve capacity, purchased power, and operating costs for each road node in each time period serves as the result of resource aggregation and scheduling for the highway virtual power plant.

[0082] This embodiment establishes a global and forward-looking operational framework by constructing a comprehensive scheduling objective function and generating a baseline scheduling plan during the day-ahead phase, providing a foundation for subsequent rolling corrections and real-time adjustments. By rolling corrections of the day-ahead plan based on real-time operational data during the intraday phase, scheduling boundary parameters are redefined and the scheduling plan is adjusted, improving the scheduling plan's adaptability to traffic flow changes and energy output fluctuations, and reducing the impact of day-ahead forecast deviations on actual operation. Through rapid correction based on short-cycle data collected during the real-time phase, short-term disturbances such as sudden increases in charging load, sharp drops in photovoltaic output, and energy storage exceeding limits are addressed promptly, ensuring the safe and stable operation of the highway virtual power plant. Through coordinated scheduling at the day-ahead, intraday, and real-time time scales, multi-time-scale closed-loop optimization driven by traffic flow forecasting is achieved, improving the accuracy and response speed of scheduling decisions, reducing operating costs and carbon emissions, while simultaneously ensuring the quality of traffic services such as vehicle refueling and road safety.

[0083] Based on the first embodiment of this application, this application also provides a low-carbon resource scheduling device for a highway virtual power plant. Please refer to... Figure 3 The device includes: The acquisition module 10 is used to acquire highway spatial topology data, traffic operation data, energy operation data, energy facility distribution data, external environment data, and market and carbon emission related data to obtain a multi-source heterogeneous dataset. Module 20 is used to construct the point-chain network aggregation structure of the highway virtual power plant based on the highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset. The prediction module 30 is used to dynamically predict the traffic flow of each road segment, service area and key node within a future set time domain based on traffic operation data and external environment data from multi-source heterogeneous datasets, and obtain the dynamic prediction result of traffic flow. The energy demand determination module 40 is used to determine the energy demand of each road node in a future set time domain based on the dynamic prediction results of traffic flow and the point-chain network aggregation structure. The scheduling boundary determination module 50 is used to dynamically determine the set of scheduling boundary parameters for road nodes based on energy demand and energy operation data in a multi-source heterogeneous dataset. The set of scheduling boundary parameters for road nodes includes flexible load boundary, resource adjustable boundary, and reserve demand boundary. The execution module 60 is used to perform low-carbon resource aggregation scheduling at multiple time scales, including day-ahead, intraday, and real-time, based on the set of road domain node scheduling boundary parameters and market and carbon emission related data in the multi-source heterogeneous dataset, to obtain the resource aggregation scheduling results of the highway virtual power plant.

[0084] The low-carbon resource scheduling device for highway virtual power plants provided in this application, employing the low-carbon resource scheduling method for highway virtual power plants in the above embodiments, can solve the technical problem of how to effectively couple the dynamic prediction results of highway traffic flow with the resource aggregation and scheduling of virtual power plants, enabling the scheduling plan to reflect the impact of traffic flow changes on energy demand and resource regulation capacity in a timely manner. Compared with the prior art, the beneficial effects of the low-carbon resource scheduling device for highway virtual power plants provided in this application are the same as those of the low-carbon resource scheduling method for highway virtual power plants provided in the above embodiments, and other technical features in the low-carbon resource scheduling device for highway virtual power plants are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0085] This application provides a low-carbon resource scheduling device for a highway virtual power plant. The device includes: 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 are executed by the at least one processor to enable the at least one processor to perform the low-carbon resource scheduling method for the highway virtual power plant described in Embodiment 1 above.

[0086] The following is for reference. Figure 4 This document illustrates a structural schematic diagram of a low-carbon resource scheduling device suitable for implementing embodiments of this application's highway virtual power plant. The low-carbon resource scheduling device for the highway virtual power plant in this application's embodiments may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The illustrated highway virtual power plant low-carbon resource scheduling device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments in this application.

[0087] like Figure 4As shown, the low-carbon resource scheduling equipment for a highway virtual power plant may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the highway virtual power plant low-carbon resource scheduling equipment. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the highway virtual power plant low-carbon resource scheduling equipment to exchange data with other devices wirelessly or via wired communication. Although various types of highway virtual power plant low-carbon resource scheduling equipment are shown in the figures, it should be understood that implementation or possession of all of them is not required. More or fewer of them may be implemented alternatively.

[0088] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0089] The low-carbon resource scheduling equipment for highway virtual power plants provided in this application, employing the low-carbon resource scheduling method for highway virtual power plants described in the above embodiments, can solve the technical problem of how to effectively couple the dynamic prediction results of highway traffic flow with the resource aggregation and scheduling of virtual power plants, enabling the scheduling plan to reflect the impact of traffic flow changes on energy demand and resource regulation capacity in a timely manner. Compared with the prior art, the beneficial effects of the low-carbon resource scheduling equipment for highway virtual power plants provided in this application are the same as those of the low-carbon resource scheduling method for highway virtual power plants provided in the above embodiments, and other technical features of this low-carbon resource scheduling equipment are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0090] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0091] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0092] This application provides a computer-readable medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the highway virtual power plant low-carbon resource scheduling method in the above embodiments.

[0093] The computer-readable medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor devices, or any combination thereof. More specific examples of computer-readable media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable medium may be any tangible medium containing or storing a program that can be executed by instructions, used by a device, or used in conjunction with it. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0094] The aforementioned computer-readable medium may be included in the low-carbon resource scheduling equipment of the highway virtual power plant; or it may exist independently and not be assembled into the low-carbon resource scheduling equipment of the highway virtual power plant.

[0095] The aforementioned computer-readable medium carries one or more programs that, when executed by the highway virtual power plant low-carbon resource scheduling device, enable the device to write computer program code for performing the operations of this application in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0096] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of this application. In this regard, all blocks in the flowcharts or block diagrams may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that all blocks in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using dedicated hardware-based implementations that perform the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0097] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0098] The readable medium provided in this application is a computer-readable medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-described low-carbon resource scheduling method for highway virtual power plants. This solves the technical problem of how to effectively couple the dynamic prediction results of highway traffic flow with the resource aggregation and scheduling of virtual power plants, enabling the scheduling plan to reflect the impact of traffic flow changes on energy demand and resource regulation capabilities in a timely manner. Compared with the prior art, the beneficial effects of the computer-readable medium provided in this application are the same as those of the low-carbon resource scheduling method for highway virtual power plants provided in the above embodiments, and will not be elaborated upon here.

[0099] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the low-carbon resource scheduling method for a highway virtual power plant as described above.

[0100] The computer program product provided in this application solves the technical problem of how to effectively couple the dynamic prediction results of highway traffic flow with the resource aggregation and scheduling of virtual power plants, so that the scheduling plan can reflect the impact of traffic flow changes on energy demand and resource regulation capacity in a timely manner. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the low-carbon resource scheduling method for highway virtual power plants provided in the above embodiments, and will not be repeated here.

[0101] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A low-carbon resource scheduling method for a virtual power plant on a highway, characterized in that, The method includes: Acquire spatial topology data of highways, traffic operation data, energy operation data, energy facility distribution data, external environment data, and market and carbon emission related data to obtain a multi-source heterogeneous dataset; Based on the highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset, a point-chain-network aggregation structure for a highway virtual power plant is constructed. Based on the traffic operation data and external environment data of the multi-source heterogeneous dataset, the traffic flow of each road segment, service area and key node within the future set time domain is dynamically predicted to obtain the dynamic traffic flow prediction result. Based on the dynamic traffic flow prediction results and the point-chain network aggregation structure, the energy demand of each road node in the future set time domain is determined. The set of road node scheduling boundary parameters is dynamically determined based on the energy demand and energy operation data in the multi-source heterogeneous dataset. The set of road node scheduling boundary parameters includes flexible load boundary, resource adjustable boundary, and reserve demand boundary. Based on the set of road node scheduling boundary parameters and market and carbon emission related data in the multi-source heterogeneous dataset, low-carbon resource aggregation scheduling is performed at multiple time scales, including day-ahead, intraday, and real-time, to obtain the resource aggregation scheduling results of the highway virtual power plant.

2. The method as described in claim 1, characterized in that, The step of constructing the point-chain network aggregation structure of the highway virtual power plant based on the highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset includes: The location information of service areas, toll stations, tunnels, and charging / battery swapping stations are extracted from the spatial topology data of highways to obtain a set of road domain node locations; The distributed photovoltaic configuration information, energy storage device configuration information, charging pile configuration information, and battery swapping facility configuration information within each road node are extracted from the energy facility distribution data to obtain the node resource configuration set; Based on the set of road domain node locations and the set of node resource configurations, highway service areas, toll stations, tunnels, and charging / battery swapping stations with energy facilities are identified as road domain nodes, thus obtaining the set of road domain nodes. Based on the power connection and energy coordination relationships between adjacent road nodes in the road node set, the power exchange capacity and transmission capacity between adjacent nodes are determined, and the link parameter set is obtained. Based on the set of road domain nodes and the set of link parameters, a point-chain network aggregation structure for a highway virtual power plant is constructed, with road domain nodes as points, link parameters as chains, and multi-node collaborative control as the network.

3. The method as described in claim 1, characterized in that, The step of dynamically predicting traffic flow for each road segment, service area, and key node within a future set time domain based on traffic operation data and external environment data from the multi-source heterogeneous dataset, and obtaining the dynamic traffic flow prediction result, includes: Historical traffic flow data, real-time traffic flow data, average vehicle speed data, road occupancy data, and road topology data are extracted from the traffic operation data in the multi-source heterogeneous dataset to obtain traffic flow time series data; Holiday information data, meteorological data, and emergency data are extracted from the external environment data in the multi-source heterogeneous dataset to obtain traffic flow scenario data; Based on road topology data, service areas, toll stations, interchange hubs, and key road sections are identified as spatiotemporal prediction nodes. Weighted spatiotemporal association edges are constructed based on the road segment distance and traffic capacity between adjacent spatiotemporal prediction nodes to obtain the spatiotemporal graph structure of the highway. Based on the traffic flow time series data, the temporal traffic status of each road node is trended using a preset time series prediction model to obtain the node temporal trend characteristics. Based on the temporal trend characteristics of the nodes and the spatiotemporal graph structure of the highway, spatial correlation propagation and node state updates are performed through a preset graph neural network prediction model to obtain the spatial coupling characteristics of the nodes. Based on the traffic flow scenario data, the corresponding traffic flow correction coefficients are extracted from the preset scenario correction factor library, and the node spatial coupling features are subjected to scenario adaptive weighted correction to obtain the dynamic prediction result of traffic flow.

4. The method as described in claim 1, characterized in that, The step of determining the energy demand of each road node in a future set time domain based on the dynamic traffic flow prediction results and the point-chain network aggregation structure includes: The predicted vehicle arrivals, vehicle type structure, vehicle dwell time and congestion status of each road node are extracted from the dynamic traffic flow prediction results to obtain the node traffic state parameters. Based on the predicted vehicle arrival volume and vehicle type structure in the node traffic state parameters, combined with the preset electric vehicle penetration rate parameters and the preset vehicle remaining battery power distribution parameters, the charging load demand and battery swapping load demand of each road node are determined, and the charging and battery swapping load prediction results are obtained. Based on the predicted vehicle arrivals in the node traffic state parameters and the meteorological data in the external environment data, the service area infrastructure load demand of each road node is determined, and the service area basic load prediction results are obtained. Based on the predicted traffic flow, average vehicle speed, and congestion status in the node traffic state parameters, the tunnel electromechanical load demand and road lighting load demand of each road node are determined, and the predicted results of tunnel and road electromechanical loads are obtained. Based on the predicted charging and swapping load, the predicted service area load, and the predicted tunnel and road electromechanical load, the energy demand of each road node in the future within a set time domain is generated.

5. The method as described in claim 1, characterized in that, The step of dynamically determining the set of road node scheduling boundary parameters based on the energy demand and energy operation data in the multi-source heterogeneous dataset includes: The charging and swapping load demand, air conditioning and heating load demand, and lighting load demand of each road node are extracted from the energy demand to obtain the node flexible load demand set; Based on the set of flexible load requirements of the nodes and the constraints of vehicle waiting time, road illumination requirements, and tunnel ventilation safety requirements, the adjustable range of tunnel ventilation load, adjustable range of charging and swapping load, adjustable range of HVAC load, and adjustable range of lighting load for each road node are determined, and the flexible load boundary is obtained. Distributed photovoltaic power output data, energy storage charge status data, and charging and swapping facility status data of each road node are extracted from the energy operation data of the multi-source heterogeneous dataset to obtain node resource operation status data; Based on the node resource operation status data and flexible load boundary, the adjustable range of distributed photovoltaic, energy storage and load side of each road node are determined to obtain the resource adjustable boundary. Based on the predicted vehicle arrivals and event status data in the traffic flow dynamic prediction results, and the distributed photovoltaic output data in the energy operation data of the multi-source heterogeneous dataset, the node load fluctuation and energy output fluctuation are calculated through a preset uncertainty quantification model to obtain the node uncertainty parameters. Based on the node uncertainty parameters and the resource adjustability boundary, the uncertainty reserve requirements, event reserve requirements, and peak reserve requirements of each road domain node are determined, resulting in a set of road domain node scheduling boundary parameters.

6. The method as described in claim 1, characterized in that, The step of performing low-carbon resource aggregation scheduling at multiple time scales (day-ahead, intraday, and real-time) based on the set of road node scheduling boundary parameters and market and carbon emission-related data in the multi-source heterogeneous dataset to obtain the resource aggregation scheduling result of the highway virtual power plant includes: Based on the set of road zone node scheduling boundary parameters and market and carbon emission related data, a comprehensive scheduling objective function with operating cost, carbon emission cost and traffic service quality loss cost as optimization objectives is constructed before the operation date, and a day-ahead baseline scheduling plan is generated. During the operating day, real-time traffic operation data, real-time energy operation data, and real-time market and carbon emission related data are acquired according to a preset rolling cycle to obtain the intraday real-time operation data. Based on the intraday real-time operation data, the daily baseline scheduling plan is rolled over and revised, the set of scheduling boundary parameters for the road nodes in subsequent periods is re-determined, and the energy storage charging and discharging plan, charging and swapping power arrangement and flexible load response plan are adjusted to obtain the intraday revised scheduling plan; During the real-time operation phase, real-time traffic operation data and real-time energy operation data are collected in a preset short period and compared with the intraday revised scheduling plan to obtain real-time operation deviation data. The intraday revised scheduling plan is corrected in real time based on the real-time operational deviation data. By adjusting the energy storage output, charging and swapping power and flexible load adjustment, the resource aggregation and scheduling result of the highway virtual power plant is obtained.

7. The method as described in claim 1, characterized in that, The steps of acquiring highway spatial topology data, traffic operation data, energy operation data, energy facility distribution data, external environmental data, and market and carbon emission-related data to obtain a multi-source heterogeneous dataset include: Traffic operation data is obtained by collecting highway topology, traffic flow, vehicle type structure, service area vehicle entry and exit volume and congestion status through roadside detection equipment and vehicle detectors. Energy operation data is obtained by collecting data on distributed photovoltaic output, energy storage charge status, charging and swapping facility operation status, and node power supply and distribution capacity at each road node through energy monitoring terminals. Meteorological data is collected through meteorological monitoring equipment, and holiday and weekday information data are obtained through a calendar interface to obtain external environmental data. By obtaining time-of-use electricity prices, demand response orders, and ancillary service prices through market data interfaces, and by obtaining carbon emission factor and carbon cost data through carbon emission databases, market and carbon emission related data can be obtained. The traffic operation data, energy operation data, external environment data, and market and carbon emission related data are time-aligned and format-standardized to obtain a multi-source heterogeneous dataset.

8. A low-carbon resource scheduling device for a virtual power plant on a highway, characterized in that, The device includes: The acquisition module is used to acquire highway spatial topology data, traffic operation data, energy operation data, energy facility distribution data, external environment data, and market and carbon emission related data to obtain multi-source heterogeneous datasets; The construction module is used to construct the point-chain network aggregation structure of the highway virtual power plant based on the highway spatial topology data and energy facility distribution data in the multi-source heterogeneous dataset. The prediction module is used to dynamically predict the traffic flow of each road segment, service area and key node within a future set time domain based on the traffic operation data and external environment data of the multi-source heterogeneous dataset, and obtain the dynamic prediction result of traffic flow. The energy demand determination module is used to determine the energy demand of each road node in a future set time domain based on the dynamic traffic flow prediction results and the point-chain network aggregation structure. The scheduling boundary determination module is used to dynamically determine the set of scheduling boundary parameters for road nodes based on the energy demand and energy operation data in the multi-source heterogeneous dataset. The set of scheduling boundary parameters for road nodes includes flexible load boundary, resource adjustable boundary, and reserve demand boundary. The execution module is used to perform low-carbon resource aggregation scheduling at multiple time scales, including day-ahead, intraday, and real-time, based on the set of road domain node scheduling boundary parameters and market and carbon emission-related data in the multi-source heterogeneous dataset, to obtain the resource aggregation scheduling result of the highway virtual power plant.

9. A low-carbon resource scheduling device for a virtual power plant on a highway, characterized in that, The device includes: a memory, a processor, and a highway virtual power plant low-carbon resource scheduling program stored in the memory and running on the processor, the highway virtual power plant low-carbon resource scheduling program being configured to implement the steps of the highway virtual power plant low-carbon resource scheduling method as described in any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores a low-carbon resource scheduling program for a virtual power plant on highways. When the processor executes the low-carbon resource scheduling program for a virtual power plant on highways, it implements the steps of the low-carbon resource scheduling method for a virtual power plant on highways as described in any one of claims 1-7.

Images