A highway master-substation optical storage direct current corridor collaborative planning method based on semi-dynamic traffic equalization

Abstract

The present application relates to the technical field of traffic energy integration and new power system collaborative planning, more specifically to a kind of highway main-substation optical storage direct current corridor collaborative planning method based on semi-dynamic traffic equilibrium. Including the following steps: S1, collect the basic information including traffic side information, energy side information and related system;S2, construct a semi-dynamic traffic equilibrium model, and solve iteratively using continuous average method, obtain the EV traffic flow of each period under each typical day, each road section;S3, map the EV traffic flow to the road network node, and generate the multi-typical day charging load space-time distribution of each candidate site by combining the charging trigger probability;S4, combined with the basic information and multi-typical day charging load space-time distribution, establish the optical storage and direct current corridor topology collaborative planning model for highway main-substation, give the objective function and related constraints, after linearization processing, solve, obtain the collaborative planning result of optical storage capacity configuration and direct current corridor topology.

Description

Technical Field

[0001] This invention relates to the field of transportation energy integration and new power system collaborative planning technology, and more specifically to a collaborative planning method for highway master-slave station photovoltaic-storage DC corridor based on semi-dynamic traffic equilibrium. Background Technology

[0002] With the deepening of regional "dual-carbon" goals and the accelerated construction of a low-carbon transportation system, the integration of transportation and energy has become an inevitable trend in energy transformation and the green and low-carbon development of transportation. In recent years, the number of electric vehicles (EVs) and the proportion of long-distance travel along highways have continued to increase. Highway service areas and parking areas, as important carriers for high-intensity, long-distance energy replenishment needs, exhibit high spatiotemporal volatility and random uncertainty in their fast-charging load. Existing methods mostly focus on rough estimations at the static or statistical levels, with less consideration for the deep coupling between traffic and energy and the dynamic mechanism of traffic state temporal evolution. In the highway scenario, the path selection behavior of vehicles in the road network and the longitudinal evolution of traffic states jointly determine the dynamic differences in charging load at each node. This leads to inaccurate characterization of the spatiotemporal distribution of charging load by existing technologies, resulting in redundant energy supply capacity configurations at some nodes and insufficient energy supply capacity at others. Meanwhile, existing plans mostly optimize configurations for individual sites, failing to consider the significant differences between nodes in terms of photovoltaic and energy storage resources, land space, load levels, and grid connection capacity. They also neglect the feasibility of constructing a master-slave heterogeneous energy supply topology by deploying DC lines along existing highway corridors. This results in the inability to achieve spatiotemporal energy sharing between adjacent or master-slave sites, ultimately leading to low overall renewable energy utilization and high construction and operating costs.

[0003] Therefore, how to provide a collaborative planning method for highway master-slave station photovoltaic-storage DC corridor based on semi-dynamic traffic equilibrium is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] In view of this, the present invention provides a collaborative planning method for the master-slave station photovoltaic-storage DC corridor of highways based on semi-dynamic traffic equilibrium, aiming to solve the above-mentioned technical problems.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A collaborative planning method for highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium includes the following steps: S1. Collect basic information including traffic-side information, energy-side information, and related systems; S2. Construct a semi-dynamic traffic equilibrium model and use the continuous averaging method to iteratively solve it to obtain the EV traffic flow of each time period and road segment under each typical day. S3. Map EV vehicle traffic to road network nodes and combine charging trigger probability to generate the spatiotemporal distribution of multiple typical daily charging loads for each candidate site. S4. Combining basic information and the spatiotemporal distribution of charging load on multiple typical days, a collaborative planning model for the photovoltaic-storage and DC corridor topology of highway master-slave stations is established. The objective function and related constraints are given, and the solution is obtained after linearization to obtain the collaborative planning results of photovoltaic-storage capacity configuration and DC corridor topology.

[0006] Furthermore, S1 specifically refers to: The traffic-side information includes the highway network topology, road segment length, free-flow impedance and benchmark traffic capacity of each road segment, travel demand for each origin-destination pair on multiple typical days of weekdays, weekends, and holidays, daily travel coefficient curves and their annual occurrence weights; The energy-side information includes the charging power of the EV and the upper limits of photovoltaic installation, energy storage capacity and power capacity, and electricity purchase capacity of each candidate site. The correlation coefficients include time-of-use electricity prices, unit cost, operation and maintenance ratio coefficients, service life and discount rate of photovoltaic, energy storage and DC corridors, grid loss, load shedding and curtailment penalty coefficients, and unit resistance, transmission power limit and node voltage allowable range of DC corridors.

[0007] Furthermore, S2 specifically includes: The model depicts several consecutive time periods, with the traffic distribution within each period approximately evolving from static traffic equilibrium. The traffic flow between time periods is connected through a residual flow mechanism to characterize the time-varying dynamic characteristics of highway traffic flow. The BPR road resistance function is used to describe the change in road segment travel time with traffic flow. The path selection behavior of vehicles is characterized based on the stochastic user equilibrium criterion. Then, the traffic flow of each road segment is generated according to the network traffic conservation. The MSA algorithm is used to iteratively solve the above model, and the model is calculated separately according to the differentiated travel demand of each typical day. Finally, the spatiotemporal distribution of EV traffic flow on each typical day is obtained.

[0008] Furthermore, S3 specifically includes: For any road network node, the traffic flow of each upstream road segment with that node as the destination is aggregated to obtain the traffic flow mapped to that node. Considering that only some of the vehicles arriving at the node generate charging demand, a time-related charging trigger probability is introduced to characterize the charging behavior of the vehicles, and the traffic flow of the node is mapped to the charging load of the node at each time. The spatiotemporal distribution of EV vehicle traffic on each typical day is mapped to obtain the spatiotemporal distribution of charging load on multiple typical days.

[0009] Furthermore, S4 specifically includes: With the objective function of minimizing the sum of annualized investment cost, operation and maintenance cost and weighted operating cost on multiple typical days, and taking into account the capacity and output constraints of each equipment, the power purchase constraints of the main station, the load shedding constraints, the power flow and voltage constraints, the node power balance constraints and topology constraints, a collaborative planning model for photovoltaic energy storage and DC corridor for highway main-substation is established. Virtual supersource points are introduced and allowed to access each candidate master station node via virtual edges to construct an improved generative forest constraint suitable for multi-master station DC corridors. To address the network loss term related to the squared power flow of the corridor in the objective function, a piecewise linearization method is adopted to transform the original problem into a mixed integer linear programming problem. The Gurobi algorithm is then used to solve the collaborative planning model of photovoltaic storage and DC corridor for highway master-slave stations, and the collaborative planning results of photovoltaic storage capacity configuration and DC corridor topology are obtained.

[0010] Furthermore, the BPR function is used to describe the change in road segment travel time with traffic flow. The path travel time is obtained by summing the travel times of each road segment contained in the path: , , In the formula, For time period Section The passage time, and These are the road's free-flow impedance and reference traffic capacity, respectively. For time period Section Traffic flow For time period Internal OD The The travel time for each route, This aggregates all road sections across the entire network. For path-segment association variables, path Included road sections The value is 1 if the condition is met, otherwise it is 0. Construct the effective path set for each OD pair based on the Dial algorithm. The Logit model is introduced to calculate the selection probability of each effective path, and EV travel demand is proportionally allocated to each path. Then, based on the network traffic conservation constraint, EV traffic for each road segment is generated. , , , , In the formula, For time period Internal path selection The probability, This is the dispersion factor; the larger the value, the more sensitive the car owner is to time costs. for Time period Between EV traffic allocated on each path for Time period EV travel demand between A collection of all OD pairs across the network; Introducing a surplus flow mechanism to semi-dynamically adjust demand for the next time period ensures the conservation of total traffic flow during the rolling time sequence: , , In the formula, for Time period The remaining traffic, The length of the scheduling time. To adjust for vehicle travel demand. To correct the carry-over coefficient.

[0011] Furthermore, for any node The traffic flow of each upstream road segment ending at this node is aggregated to obtain the traffic flow mapped to this node, and the traffic flow arriving at the node is calculated. Only a portion of the vehicles (EVs) generate charging demand. A charging trigger probability is introduced to characterize the vehicle's charging behavior, thereby obtaining the node charging load: , , In the formula, For time period node Traffic flow For nodes The upstream section collection, For a moment node The charging load, For a moment Charging trigger probability, The charging power for EVs.

[0012] Furthermore, the collaborative planning model for the photovoltaic-storage and DC corridor of the highway master-slave station is as follows: , , , , , , , , , In the formula, and These are the sets of all nodes and the set of main site nodes, respectively. , and These are the equivalent annual investment costs for the DC corridor, photovoltaic units, and energy storage units, respectively. The discount rate; , and These refer to the lifespan of the DC corridor, photovoltaic units, and energy storage units, respectively. , and These refer to photovoltaic construction capacity, energy storage capacity, and energy storage power capacity, respectively. For the candidate DC corridor set, DC corridor The corresponding 0-1 variables, Representative construction, DC corridor Length, , , and These are the cost coefficients for DC corridor, photovoltaic, energy storage capacity, and energy storage power capacity, respectively. , and These are the operation and maintenance ratio coefficients for DC corridors, photovoltaics, and energy storage, respectively. , , and These are the typical daily electricity purchase cost, grid loss cost, load shedding penalty, and curtailment penalty. For time-of-use electricity pricing, , and These are network loss cost, load loss penalty, and curtailment penalty coefficient, respectively. For the set of all typical days, This represents the number of days in the year for each typical day. For the power purchase capacity, The impedance of the DC corridor. For the DC corridor current, For unmet load power, This refers to the power of abandoned light.

[0013] Furthermore, the improved generative forest constraint applicable to multi-master DC corridors is as follows: , , , , , , In the formula, The bidirectional directed arc set obtained by expanding the candidate DC corridor set. A virtual flow from node to node. The total number of nodes. The virtual stream delivered from the virtual supersource node to the master node. 0-1 variables represent nodes Whether to connect to the virtual supersource point Let i be the virtual flow obtained by node i from the virtual supersource point.

[0014] Furthermore, a piecewise linearization method is used to process the data, dividing the defined interval of the corridor power flow into... Each segment is of equal length, and segmented power flow variables are introduced. And with the slope of each secant segment An approximation is made so that the network loss cost can be linearly expressed as: .

[0015] Compared to existing technologies, this invention constructs a semi-dynamic traffic equilibrium model, employing stochastic user equilibrium theory to accurately characterize the path selection behavior of EV users, and introducing a residual flow mechanism to describe the temporal evolution of highway traffic conditions. This accurately captures and generates spatiotemporal distribution data of fast-charging loads that highly match the dynamic characteristics of the actual road network. Based on this, a collaborative planning model for the photovoltaic-storage capacity configuration and DC corridor topology of highway master-slave stations is established. The optimal configuration of photovoltaic and energy storage capacities is collaboratively solved with the DC corridor topology design, maximizing the potential for spatiotemporal energy mutual assistance between master and slave stations. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the collaborative planning method for a highway master-slave station photovoltaic-storage DC corridor based on semi-dynamic traffic equilibrium, according to the present invention.

[0018] Figure 2 This is a flowchart illustrating how the MSA algorithm of this invention solves for EV flow. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, this invention proposes a collaborative planning method for highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium, including: Step S101: Collect basic information such as highway traffic network topology, travel demand, resource endowment and correlation coefficients of each candidate station: Traffic-side information is collected, including traffic network topology, free-flow impedance and baseline capacity of each road segment, baseline OD travel demand for three typical days (weekdays, weekends, and holidays), daily travel coefficient curves for each typical day, and their annual occurrence weights; the traffic-side information is used to provide input for the semi-dynamic traffic equilibrium solution in step S102.

[0021] Collect EV charging behavior information, including the charging trigger probability and charging power of the EV at each time moment; the charging behavior information is used to provide input for the generation of charging demand in step S103.

[0022] Collect energy-side information and correlation coefficients. The energy-side information includes the corresponding locations and voltage allowable ranges of master station nodes and sub-station nodes, the upper limit of photovoltaic installation capacity, energy storage power limit, energy storage capacity limit, power purchase capacity limit, and photovoltaic output coefficient curve of each master station node, as well as the candidate DC corridor set, the unit length resistance and transmission power limit of each candidate DC corridor; the correlation coefficients include time-of-use electricity price, cost coefficient, operation and maintenance ratio coefficient, service life and discount rate of each device, and penalty coefficients for network loss, load shedding and curtailment; the energy-side information and correlation coefficients, together with the charging demand generated in step S103, constitute the input boundary of collaborative planning in step S104.

[0023] Step S102: The entire research cycle Divided into The traffic flow distribution within each continuous time period is approximately evolved from static traffic equilibrium. The time periods are connected by a residual flow mechanism, thus establishing a semi-dynamic traffic equilibrium model. The MSA algorithm is used to solve for the traffic flow distribution of each typical day.

[0024] The BPR function is used to describe the change in road segment travel time with traffic flow. The path travel time is obtained by summing the travel times of each road segment contained in the path. , , In the formula, For time period Section The passage time, and These are the road's free-flow impedance and reference traffic capacity, respectively. For time period Section Traffic flow For time period Internal OD The The travel time for each route, This aggregates all road sections across the entire network. For path-segment association variables, path Included road sections The value is 1 if the condition is met, otherwise it is 0. Considering the driver's perception error and bounded rationality regarding road network information, an effective path set for each OD pair is constructed based on the Dial algorithm. The Logit model is introduced to calculate the selection probability of each effective path, and EV travel demand is proportionally allocated to each path. Then, based on the network traffic conservation constraint, EV traffic for each road segment is generated. , , , , In the formula, For time period Internal path selection The probability, This is the dispersion factor; the larger the value, the more sensitive the car owner is to time costs. for Time period Between EV traffic allocated on each path for Time period EV travel demand between A collection of all OD pairs across the network; Due to the long length of highways and the dispersed departure times of vehicles, some vehicles may not be able to complete their journey within the current time period. To address this, a "residual flow" mechanism is introduced to semi-dynamically adjust the demand for the next time period, ensuring the total traffic flow remains constant throughout the rolling time sequence. , , In the formula, for Time period The remaining traffic, The length of the scheduling time. To adjust for vehicle travel demand. To correct the carry-over coefficient.

[0025] For the above model, the MSA algorithm is used for iterative solution. In the nth iteration, the path auxiliary flow is obtained based on the current segment impedance according to Logit, and then the path flow is updated according to the following formula: , Iterate until the convergence condition is met. ,in The error is preset. The solution process is attached. Figure 2 As shown, the solutions are obtained for three typical days—weekdays, weekends, and holidays—based on their respective differentiated travel demands, ultimately yielding the spatiotemporal distribution of EV traffic flow for each typical day.

[0026] Step S103: Based on the EV segment traffic flow obtained in step S102, establish a mapping relationship between segment traffic flow and node charging load.

[0027] For any node The traffic flow of each upstream road segment ending at this node is aggregated to obtain the traffic flow mapped to this node. (Destination Node) Only a portion of the vehicles (EVs) generate charging demand. A charging trigger probability is introduced to characterize the vehicle's charging behavior, thereby obtaining the node charging load: , , In the formula, For time period node Traffic flow For nodes The upstream section collection, For a moment node The charging load, For a moment Charging trigger probability, The charging power for EVs.

[0028] Step S104: Based on the charging demand generated in step S103, establish a collaborative planning model for the photovoltaic-storage and DC corridor topology of the highway master-slave station, perform piecewise linearization on the original problem, call Gurobi to obtain the global optimal solution, and output the collaborative planning results of photovoltaic-storage capacity configuration and DC corridor topology.

[0029] With the objective function of minimizing the sum of annualized investment cost, operation and maintenance cost, and weighted average operating cost on multiple typical days, and taking into account the capacity and output constraints of each device, the power purchase constraints of the main station, the load shedding constraints, the power flow and voltage constraints, the node power balance constraints, and the topology constraints, a collaborative planning model for photovoltaic energy storage and DC corridors for highway master-slave stations is established as follows: , , , , , , , , , In the formula, and These are the sets of all nodes and the set of main site nodes, respectively. , and These are the equivalent annual investment costs for the DC corridor, photovoltaic units, and energy storage units, respectively. The discount rate; , and These refer to the lifespan of the DC corridor, photovoltaic units, and energy storage units, respectively. , and These refer to photovoltaic construction capacity, energy storage capacity, and energy storage power capacity, respectively. For the candidate DC corridor set, DC corridor The corresponding 0-1 variables, Representative construction, DC corridor Length, , , and These are the cost coefficients for DC corridor, photovoltaic, energy storage capacity, and energy storage power capacity, respectively. , and These are the operation and maintenance ratio coefficients for DC corridors, photovoltaics, and energy storage, respectively. , , and These are the typical daily electricity purchase cost, grid loss cost, load shedding penalty, and curtailment penalty. For time-of-use electricity pricing, , and These are network loss cost, load loss penalty, and curtailment penalty coefficient, respectively. For the set of all typical days, This represents the number of days in the year for each typical day. For the power purchase capacity, The impedance of the DC corridor. For the DC corridor current, For unmet load power, This refers to the power of abandoned light.

[0030] Considering that photovoltaic and energy storage at each master station node are limited by geographical space and sunlight conditions, and that substations cannot be equipped with photovoltaic and energy storage devices, the following constraints should be met: , , In the formula, , and These are the upper limits for the photovoltaic construction capacity of the main station, the upper limit for energy storage capacity, and the upper limit for energy storage power capacity, respectively. This is the set of sub-station nodes.

[0031] Considering the different converter access capabilities of each master station node, resulting in differences in power purchase capacity among the stations; and the fact that substations do not have power purchase capacity, the following constraints should be met: , , In the formula, Main site The upper limit of electricity purchasing capacity.

[0032] Considering that the sum of actual photovoltaic output and curtailed power is constrained by installed capacity and time-series irradiance, the following constraints should be met: , In the formula: Photovoltaic power generation capacity, This represents the output coefficient of photovoltaic power.

[0033] Considering the charging and discharging behavior of the energy storage system during the scheduling cycle, the following mathematical model is constructed for constraint: , , , , , In the formula, , These represent the charging and discharging power of the energy storage, and For energy storage charging and discharging efficiency, Main site On a typical day Inside State of charge at time t, The variable is 0-1, controlling whether the energy storage can charge and discharge simultaneously. For discharging.

[0034] Considering that a penalty needs to be added to the objective function when the partial load is not satisfied, the following constraints should be satisfied: , , , In the formula, For the site On a typical day Inside The charging load at any time, This indicates the charging load power that has been satisfied.

[0035] The power balance constraint must be satisfied by both the master station and the slave station at any given time: , , In the formula, and These are sets of routes that start and end at specific nodes. The power transmitted via the DC corridor.

[0036] DC corridor power transmission must meet voltage and power flow constraints. It is assumed that the DC node voltage operates near its rated value, and a voltage squared variable is used. ,exist Under these conditions, the Big M method is used to combine linearized power flow constraints with DC corridor construction variables. Couple the nodes and further constrain their operating voltages: , , , In the formula, when When the condition is met, the above formula represents the linearized power flow constraint for the DC corridor; conversely, the large M term releases this constraint, preventing unbuilt lines from participating in power flow calculations. This represents the upper limit of the DC corridor's transmission power. and These represent the upper and lower limits of the node voltage amplitude, respectively.

[0037] To ensure that the planned DC corridor network has a reasonable radial connectivity structure and avoid isolated substations or redundant loops, virtual super-source points are introduced, allowing them to access candidate master station nodes via virtual edges. The following improved generating forest constraint suitable for multi-master station DC corridors is constructed: , , , , , , In the formula, The bidirectional directed arc set obtained by expanding the candidate DC corridor set. A virtual flow from node to node. The total number of nodes. The virtual stream delivered from the virtual supersource node to the master node. 0-1 variables represent nodes Whether to connect to the virtual supersource point, Let i be the virtual flow obtained by node i from the virtual supersource point.

[0038] Since the network loss term in the objective function contains the square of the current term, in order to reduce model complexity and improve solution efficiency, a piecewise linearization method is used to transform the original model into a mixed-integer linear programming problem: First, rewrite the network loss cost in power form: , Will The defined interval is divided into K segments of equal length, each segment having a length of: , By introducing segmented power flow variables, the slope of the secant line in each segment is used to approximate the linear slope of that segment: , , , Therefore, the network loss cost can be written as: , After the above processing, the original problem is transformed into a mixed integer linear programming problem. The Gurobi algorithm is then used to solve the above model to obtain the collaborative planning results of the photovoltaic storage capacity configuration and the DC corridor topology.

[0039] This invention fully considers the path selection behavior of EV users in the road network and the impact of traffic state evolution over time on the spatiotemporal distribution of node loads, effectively overcoming the problem of low accuracy in characterizing the spatiotemporal differences of charging loads in existing technologies. Simultaneously, considering the varying resource endowments and equipment construction conditions of different service areas and rest areas along highways, a DC corridor is introduced to achieve site interconnection and source-load mutual support, improving the level of new energy consumption. This invention collaboratively optimizes the photovoltaic and energy storage capacity configuration and DC corridor topology of resource-differentiated sites, realizing cross-temporal and spatiotemporal consumption of photovoltaic and energy storage resources among multiple sites and global scheduling optimization, providing a scientific decision-making basis for the collaborative construction of green energy networks on highways.

[0040] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0041] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A collaborative planning method for highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium, characterized in that, Includes the following steps: S1. Collect basic information including traffic-side information, energy-side information, and related systems; S2. Construct a semi-dynamic traffic equilibrium model and use the continuous averaging method to iteratively solve it to obtain the EV traffic flow of each time period and road segment under each typical day. S3. Map EV vehicle traffic to road network nodes and combine charging trigger probability to generate the spatiotemporal distribution of multiple typical daily charging loads for each candidate site. S4. Combining basic information and the spatiotemporal distribution of charging load on multiple typical days, a collaborative planning model for the photovoltaic-storage and DC corridor topology of highway master-slave stations is established. The objective function and related constraints are given, and the solution is obtained after linearization to obtain the collaborative planning results of photovoltaic-storage capacity configuration and DC corridor topology.

2. The method for collaborative planning of highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium as described in claim 1, characterized in that, Specifically, S1 is: The traffic-side information includes the highway network topology, road segment length, free-flow impedance and benchmark traffic capacity of each road segment, travel demand for each origin-destination pair on multiple typical days of weekdays, weekends, and holidays, daily travel coefficient curves and their annual occurrence weights; The energy-side information includes the charging power of the EV and the upper limits of photovoltaic installation, energy storage capacity and power capacity, and electricity purchase capacity of each candidate site. The correlation coefficients include time-of-use electricity prices, unit cost, operation and maintenance ratio coefficients, service life and discount rate of photovoltaic, energy storage and DC corridors, grid loss, load shedding and curtailment penalty coefficients, and unit resistance, transmission power limit and node voltage allowable range of DC corridors.

3. The method for collaborative planning of highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium as described in claim 1, characterized in that, Specifically, S2 is: The model depicts several consecutive time periods, with the traffic distribution within each period approximately evolving from static traffic equilibrium. The traffic flow between time periods is connected through a residual flow mechanism to characterize the time-varying dynamic characteristics of highway traffic flow. The BPR road resistance function is used to describe the change in road segment travel time with traffic flow. The path selection behavior of vehicles is characterized based on the stochastic user equilibrium criterion. Then, the traffic flow of each road segment is generated according to the network traffic conservation. The MSA algorithm is used to iteratively solve the above model, and the model is calculated separately according to the differentiated travel demand of each typical day. Finally, the spatiotemporal distribution of EV traffic flow on each typical day is obtained.

4. The method for collaborative planning of highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium as described in claim 1, characterized in that, Specifically, S3 is: For any road network node, the traffic flow of each upstream road segment with that node as the destination is aggregated to obtain the traffic flow mapped to that node. Considering that only some of the vehicles arriving at the node generate charging demand, a time-related charging trigger probability is introduced to characterize the charging behavior of the vehicles, and the traffic flow of the node is mapped to the charging load of the node at each time. The spatiotemporal distribution of EV vehicle traffic on each typical day is mapped to obtain the spatiotemporal distribution of charging load on multiple typical days.

5. The method for collaborative planning of highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium as described in claim 1, characterized in that, Specifically, S4 is: With the objective function of minimizing the sum of annualized investment cost, operation and maintenance cost and weighted operating cost on multiple typical days, and taking into account the capacity and output constraints of each equipment, the power purchase constraints of the main station, the load shedding constraints, the power flow and voltage constraints, the node power balance constraints and topology constraints, a collaborative planning model for photovoltaic energy storage and DC corridor for highway main-substation is established. Virtual supersource points are introduced and allowed to access each candidate master station node via virtual edges to construct an improved generative forest constraint suitable for multi-master station DC corridors. To address the network loss term related to the squared power flow of the corridor in the objective function, a piecewise linearization method is adopted to transform the original problem into a mixed integer linear programming problem. The Gurobi algorithm is then used to solve the collaborative planning model of photovoltaic storage and DC corridor for highway master-slave stations, and the collaborative planning results of photovoltaic storage capacity configuration and DC corridor topology are obtained.

6. The method for collaborative planning of highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium as described in claim 3, characterized in that, The BPR function is used to describe the change in road segment travel time with traffic flow. The path travel time is obtained by summing the travel times of each road segment contained in the path. ； ； In the formula, For time period Section The passage time, and These are the road's free-flow impedance and reference traffic capacity, respectively. For time period Section Traffic flow For time period Internal OD The The travel time for each route, This aggregates all road sections across the entire network. For path-segment association variables, path Included road sections The value is 1 if the condition is met, otherwise it is 0. Construct the effective path set for each OD pair based on the Dial algorithm. The Logit model is introduced to calculate the selection probability of each effective path, and EV travel demand is proportionally allocated to each path. Then, based on the network traffic conservation constraint, EV traffic for each road segment is generated. ； ； ； ； In the formula, For time period Internal path selection The probability, This is the dispersion factor; the larger the value, the more sensitive the car owner is to time costs. for Time period Between EV traffic allocated on each path for Time period EV travel demand between It is a collection of all OD pairs across the network; Introducing a surplus flow mechanism to semi-dynamically adjust demand for the next time period ensures the conservation of total traffic flow during the rolling time sequence: ； ； In the formula, for Time period The remaining traffic, The length of the scheduling time. To adjust for vehicle travel demand. To correct the carry-over coefficient.

7. The method for collaborative planning of highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium as described in claim 4, characterized in that, For any node The traffic flow of each upstream road segment ending at this node is aggregated to obtain the traffic flow mapped to this node, and the traffic flow arriving at the node is calculated. Only a portion of the vehicles (EVs) generate charging demand. A charging trigger probability is introduced to characterize the vehicle's charging behavior, thereby obtaining the node charging load: ； ； In the formula, For time period node Traffic flow For nodes The upstream section collection, For a moment node The charging load, For a moment Charging trigger probability, The charging power for EVs.

8. The method for collaborative planning of highway master-slave station photovoltaic-storage DC corridors based on semi-dynamic traffic equilibrium as described in claim 5, characterized in that, The collaborative planning model for the photovoltaic energy storage and DC corridor of the highway master-slave station is as follows: ； ； ； ； ； ； ； ； ； In the formula, and These are the sets of all nodes and the set of main site nodes, respectively. , and These are the equivalent annual investment costs for the DC corridor, photovoltaic units, and energy storage units, respectively. The discount rate; , and These refer to the lifespan of the DC corridor, photovoltaic units, and energy storage units, respectively. , and These refer to photovoltaic construction capacity, energy storage capacity, and energy storage power capacity, respectively. For the candidate DC corridor set, DC corridor The corresponding 0-1 variables, Represents construction, DC corridor Length, , , and These are the cost coefficients for DC corridor, photovoltaic, energy storage capacity, and energy storage power capacity, respectively. , and These are the operation and maintenance ratio coefficients for DC corridors, photovoltaics, and energy storage, respectively. , , and These are the typical daily electricity purchase cost, grid loss cost, load shedding penalty, and curtailment penalty. For time-of-use electricity pricing, , and These are network loss cost, load loss penalty, and curtailment penalty coefficient, respectively. For the set of all typical days, This represents the number of days in the year for each typical day. For the power purchase capacity, The impedance of the DC corridor. For the DC corridor current, For unmet load power, This refers to the power of abandoned light.

9. A collaborative planning method for highway master-slave station photovoltaic-storage DC corridor based on semi-dynamic traffic equilibrium, as described in claim 8, is characterized in that... The improved generative forest constraint applicable to multi-master DC corridors is: ； ； ； ； ； ； In the formula, The bidirectional directed arc set obtained by expanding the candidate DC corridor set. A virtual flow from node to node. The total number of nodes. The virtual stream delivered from the virtual supersource node to the master node. 0-1 variables represent nodes Whether to connect to the virtual supersource point, Let i be the virtual flow obtained by node i from the virtual supersource point.

10. A collaborative planning method for highway master-slave station photovoltaic-storage DC corridor based on semi-dynamic traffic equilibrium, as described in claim 9, is characterized in that... A piecewise linearization method is used to process the data, dividing the definition interval of the corridor power flow into three parts. Each segment is of equal length, and segmented power flow variables are introduced. And with the slope of each secant segment An approximation is made so that the network loss cost can be linearly expressed as: 。

Images