A charging infrastructure planning and charging scheduling collaborative optimization method and system and storage medium

Abstract

The application provides a charging infrastructure planning and charging scheduling cooperative optimization method and system and a storage medium, wherein parameters of a pre-constructed CPO-EVCI double-layer model are initialized, a new CPO connected to a CPO relationship network is entered in each round of evolution of the CPO layer, and strategy updating is performed to obtain a strategy updating result; total charging demand in each preset planning period is predicted in the EVCI layer, site scale and capacity parameters of each CPO are calculated, charging station regulation and control information and typical day operation income under different investment cost scenarios are output based on a charging scheduling optimization model. The method and system provided by the application depict strategy learning and expansion differences of CPOs in a competitive environment on a long time scale, and perform typical day operation scheduling on a short time scale, so that the cooperative optimization of facility planning and charging scheduling is achieved, dynamic resource scheduling is improved, and resource utilization is improved.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging infrastructure planning and operation scheduling technology, and in particular to a collaborative optimization method, system and storage medium for charging infrastructure planning and charging scheduling. Background Technology

[0002] With the continuous growth of electric vehicle ownership, public electric vehicle charging infrastructure has become a key energy and transportation infrastructure supporting the electrification transformation of transportation. Fast charging stations, due to their high power and high turnover characteristics, have become an important guarantee for urban transportation systems. However, the construction investment of fast charging stations is high, the equipment power level is high, the payback period is long, and their centralized access will significantly change the load characteristics of the distribution network. Coupled with the fluctuation of charging demand over a certain period of time, the distribution network is more prone to problems such as capacity constraints, local overload, and power quality risks during peak hours.

[0003] To achieve coordinated optimization of electric vehicle charging infrastructure planning and charging scheduling, existing technologies typically couple long-term planning with short-term scheduling in two or more stages. However, this approach, by assuming a single planner or single entity to make decisions, struggles to express the competition, learning, and expansion processes of multiple charging point operators coexisting in the long term. Furthermore, it generally ignores the direct relationship between electric vehicle charging infrastructure and charging operators, making it difficult to reflect the long-term impact of various subsidy combinations on site layout and the ratio of fast to slow charging.

[0004] Therefore, the existing technology needs further improvement. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide users with a collaborative optimization method, system and storage medium for charging infrastructure planning and charging scheduling, overcoming the defects of existing collaborative optimization methods that are unable to reflect the long-term changes in the market environment and structure and cannot realize differentiated collaborative optimization strategies.

[0006] The technical solution adopted by this invention to solve the technical problem is as follows: In a first aspect, the present invention provides a method for the coordinated optimization of charging infrastructure planning and charging scheduling, comprising: Obtain CPO information for the target region, initialize the parameters of the pre-constructed CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer; wherein, the two-layer model includes: CPO layer and EVCI layer; wherein, the CPO layer represents the competition or information interaction relationship between each CPO as a CPO relationship network; The CPO layer connects newly entered CPOs in each round of evolution to the CPO relationship network according to pre-connection rules, and updates the strategy of the CPO relationship network using an evolutionary game mechanism to obtain the strategy update result of each round of evolution; Under the constraint of the strategy update results in each round of evolution, the EVCI layer predicts the total charging demand in each preset planning period of the target area, and calculates the number of new fast charging stations and slow charging stations and investment costs for each CPO, so as to obtain the site scale and capacity parameters of each CPO in each round of evolution. Based on the scale and capacity parameters of each CPO site in each round of evolution, a typical day's charging scheduling optimization model is constructed. With the goal of maximizing CPO profits, the charging power and queue allocation of electric vehicles in different time periods and different types of charging stations are evaluated to obtain charging station control information and typical day operating revenue under different investment cost scenarios.

[0007] Optionally, the steps of obtaining CPO information of the target region and initializing the parameters of the pre-built CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer include: Obtain the number of CPOs in the target area, the market share of each CPO, the information transparency among CPOs, the initial FCS ratio strategy of each CPO, and the number of fast charging stations and slow charging stations in each CPO. Configure market share weights for each CPO in the CPO relationship network, and configure information transparency weights for the connections between CPOs.

[0008] Optionally, the steps of connecting newly entering CPOs in each round of evolution to the CPO relationship network according to pre-connection rules, and updating the strategy of the CPO relationship network using an evolutionary game mechanism to obtain the strategy update result for each round of evolution include: In each round of evolution, newly entering CPOs are connected to at least one CPO in the CPO relationship network according to the priority connection rule to obtain an updated CPO relationship network; After the structure of the CPO relationship network is updated, each CPO selects a strategy imitation object according to the neighbor selection mechanism; wherein, the neighbor selection mechanism takes the proximity of the connection relationship between two CPOs and the weight of information transparency as factors to consider when selecting a strategy imitation object; Each CPO calculates the strategy imitation probability using the Fermi function based on the difference in revenue between itself and the strategy imitation target, and determines its own fast charging station ratio based on the strategy imitation probability.

[0009] Optionally, under the constraint of the strategy update results in each round of evolution, the EVCI layer predicts the total charging demand in the target area within each preset planning period, and calculates the number of new fast charging stations and slow charging stations and investment costs for each CPO, obtaining the site size and capacity parameters of each CPO in each round of evolution. Under the constraints of the strategy update results, the traffic demand within the target area is calculated using a preset traffic demand prediction model. Based on the aforementioned transportation demand, the total charging demand within the target area is calculated. Based on the total charging demand, the number of new fast charging stations and slow charging stations for each CPO is calculated using a preset energy system model. The number of new fast charging stations and slow charging stations is then added to the existing stock of each CPO to obtain the site size and capacity parameters of each CPO.

[0010] Optionally, the step of calculating the traffic demand within the target area using a preset traffic demand prediction model under the constraints of the strategy update result includes: The traffic demand forecasting model is constructed based on regional socio-economic variables and traffic supply variables. The regional socio-economic variables include per capita GDP, population size, and per capita built-up area. The traffic supply variables include per capita road supply and generalized travel costs.

[0011] Optionally, the step of calculating the number of new fast charging stations and slow charging stations for each CPO based on the total charging demand and using a preset energy system model includes: The energy system model aims to minimize the total cost of new EVCIs over a long time scale, and outputs the number of new FCSs, the number of new slow charging stations, and the cost for each CPO. The constraints of the energy system model include one or more of the following: supply and demand balance constraints, strategy consistency constraints, budget constraints, and cost composition constraints. Among them, the strategy consistency constraints ensure that the proportion of FCSs corresponding to the new stations falls within the strategy range given by the CPO layer.

[0012] Optionally, the step of constructing a typical day's charging scheduling optimization model based on the scale and capacity parameters of each CPO site in each round of evolution, with the goal of maximizing CPO profits, and obtaining charging station control information and typical day operating revenue under different investment cost scenarios by allocating charging power and queues for electric vehicles at different time periods and different types of charging stations, includes: Based on the scale and capacity parameters of each CPO site, a set of typical days and the weights of each set of typical days are constructed. Based on the intraday time scale of a selected typical day, a charging scheduling optimization model for the typical day is constructed. With the goal of maximizing CPO profits, the optimized operating revenue and costs of each CPO are obtained based on the aforementioned charging scheduling optimization model. The annual profit or revenue of each CPO is calculated by weighting the operating revenue and costs of each CPO. The annual profit or revenue of each CPO is fed back to the CPO layer to update the market share weight of each CPO, and the updated market share weight of each CPO is used as the revenue input for the next round of strategy evolution. The strategy update process is repeated on the CPO relationship network based on the updated revenue input. The strategy update results are obtained, the number of new fast charging stations and slow charging stations and the investment cost of each CPO are calculated, and the site size and capacity parameters of each CPO are obtained. With the site size and capacity parameters of each CPO as constraints and the goal of maximizing CPO profits, the optimized operating revenue and cost of each CPO are obtained based on the charging scheduling optimization model until the preset evolution termination condition is met. The charging station control information and typical daily operating revenue under different investment cost scenarios are obtained.

[0013] Optionally, the strategy update results include: the FCS ratio and subsidy policy parameters for each charging station; the subsidy policy parameters include: construction-driven subsidies and demand / operation-driven subsidies, wherein construction-driven subsidies include investment subsidies and equipment subsidies, and demand / operation-driven subsidies include operation subsidies and user charging subsidies.

[0014] Secondly, the present invention provides a collaborative optimization system for charging infrastructure planning and charging scheduling, comprising: An initialization module is used to acquire CPO information of the target area and initialize the parameters of a pre-constructed CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer; wherein, the two-layer model includes: a CPO layer and an EVCI layer; wherein, the CPO layer represents the competition or information interaction relationship between each CPO as a CPO relationship network; The strategy evolution module is used by the CPO layer to connect newly entered CPOs in each round of evolution to the CPO relationship network according to pre-connection rules, and to update the strategy of the CPO relationship network using an evolutionary game mechanism to obtain the strategy update result of each round of evolution; The site information calculation module is used by the EVCI layer to predict the total charging demand in each preset planning period of the target area under the constraint of the strategy update result in each round of evolution, and to calculate the number of new fast charging stations and slow charging stations and investment costs of each CPO, so as to obtain the site scale and capacity parameters of each CPO in each round of evolution. The charging scheduling optimization module is used to construct a typical day's charging scheduling optimization model based on the scale and capacity parameters of each CPO site in each round of evolution. With the goal of maximizing CPO profits, it obtains charging station control information and typical day operating revenue under different investment cost scenarios by allocating charging power and queuing for electric vehicles in different time periods and different types of charging stations.

[0015] Thirdly, the present invention provides a computer storage medium, which is a computer-readable storage medium, wherein a computer program is stored in the computer storage medium, and when the computer program is executed by a computer, the computer is used to execute the collaborative optimization method for charging infrastructure planning and charging scheduling as described above.

[0016] Beneficial effects: This invention provides a collaborative optimization method, system, and storage medium for charging infrastructure planning and charging scheduling. By acquiring CPO information for a target area, parameters are initialized on a pre-constructed CPO-EVCI two-layer model, resulting in initialized CPO and EVCI layers. The CPO layer connects newly entering CPOs to the CPO relationship network according to pre-defined connection rules and updates the CPO relationship network with strategies, obtaining strategy update results. Under the constraints of the strategy update results, the EVCI layer predicts the total charging demand in each preset planning period of the target area and calculates the number of new fast-charging and slow-charging stations and investment costs for each CPO, obtaining the site size and capacity parameters for each CPO. Based on the site size and capacity parameters of each CPO, a typical day's charging scheduling optimization model is constructed. With the goal of maximizing CPO profits, the charging power and queue allocation of electric vehicles across different time periods and different types of charging stations are analyzed to obtain charging station control information and typical day operating revenue under different investment cost scenarios. The method and system provided by this invention construct a CPO-EVCI two-layer model over a longer time scale to characterize the strategy learning and expansion differences of CPOs in a competitive environment, avoiding single-entity decision-making. Furthermore, it performs typical daily operation scheduling over a short time scale so that changes in operational constraints, service levels, and revenues can be reflected in the long-term evolution process in a traceable manner. This provides quantifiable technical support for differentiated subsidy design and the efficient and orderly development of EVCI. Attached Figure Description

[0017] Figure 1 The steps of the collaborative optimization method for charging infrastructure planning and charging scheduling provided by this invention are as follows: Figure 2 The schematic diagram of the collaborative optimization system for charging infrastructure planning and charging scheduling provided by the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] With the continuous growth of EV (Electric Vehicle) ownership, public EVCI (Electric Vehicle Charging Infrastructure) has become a key energy and transportation infrastructure supporting the electrification transformation of transportation. Especially in high-frequency, high-intensity travel scenarios such as ride-hailing and taxis, charging efficiency directly affects vehicle operating efficiency. Fast Charging Stations (FCS) have become an important guarantee for urban transportation systems due to their high power and high turnover characteristics. However, FCS construction involves high investment, high equipment power levels, and long payback periods. Furthermore, their centralized access significantly alters the load characteristics of the distribution network. Coupled with hourly or even minute-level fluctuations in charging demand, this makes the distribution network more prone to capacity constraints, localized overloads, and power quality risks during peak hours. To promote EVCI development, the government has introduced various policy tools, including investment subsidies, equipment subsidies, operating subsidies, and user charging subsidies. However, under different regions, market stages, and operating entity structures, the actual transmission effect of subsidy policies and their impact on long-term market structure evolution still lack explainable and quantifiable analytical basis.

[0020] Existing technologies still have shortcomings in evaluating subsidy policies and modeling the evolution of EVCI. On the one hand, some studies use macro-planning or energy system optimization frameworks to predict future demand for charging facilities, but they often treat infrastructure expansion as a centralized or homogeneous decision-making process. This makes it difficult to characterize the competitive learning, strategy diffusion, and market structure changes caused by the entry of new players among charging CPOs (Charge Point Operators), and also fails to reflect the impact of differences in CPO market share and information exchange on station construction decisions.

[0021] Furthermore, from a research paradigm perspective, coupling long-term planning with short-term scheduling in two or more stages has become a common approach for EVCI planning and charging dispatch management. The first stage typically determines site selection, capacity grading, and capacity expansion strategies. The second stage performs operational verification and scheduling optimization of charging load and on-site resources under typical daily conditions to reflect the impact of grid constraints, service satisfaction, and operating cost-benefit on planning decisions. Related research has established a certain foundation in charging station planning, service satisfaction constraints, and coupled modeling of distribution networks and road networks.

[0022] However, the aforementioned two-stage multi-timescale framework still needs improvement when used for policy evaluation and market structure evolution analysis. First, many models assume a single planner or single entity makes decisions, making it difficult to express the competition, learning, and expansion processes when multiple CPOs coexist on a long-term scale, and also failing to characterize the effects of differences in information exchange and market share on strategy diffusion and structural evolution. Second, existing coupled modeling often focuses on static indicators such as the number, capacity, or location of stations, lacking a long-term evolutionary characterization of the fast / slow charging structure ratio—a key indicator reflecting supply structure and service efficiency. Third, although macro-transport and energy system models (such as AIM / Enduse) can provide assessments of transportation energy demand and carbon reduction pathways on multiple scenarios and annual scales, their outputs are often difficult to reproducibly connect with CPO-level strategy updates, station-level revenue feedback, and subsequent short-term operational scheduling, thus limiting the interpretation and analysis of policy mix effects and their engineering applications.

[0023] Therefore, it is necessary to propose a multi-timescale technical solution for multiple subsidy scenarios, which can characterize the CPO decision-making and EVCI structural evolution over a longer timescale, and introduce charging scheduling evaluation of typical operating scenarios over a shorter timescale. This will enable changes in operating constraints, service levels, and revenues to be reflected in the long-term evolution process in a traceable manner, thereby providing quantifiable technical support for differentiated subsidy design and the efficient and orderly development of EVCI.

[0024] To overcome the problems in the prior art, this application proposes a collaborative optimization method, system, and storage medium for charging infrastructure planning and charging scheduling. This method and system construct a two-layer collaborative model between the Charger Point Owner (CPO) and its associated Electric Vehicle Integration Center (EVCI) over a longer timescale, characterizing the CPO's strategy evolution and site expansion behavior under market competition and policy incentives. Simultaneously, a charging scheduling model for typical operating scenarios is introduced over a shorter timescale, aiming to maximize operating profits by scheduling and verifying in-station resources and the charging process. The operational benefits are then fed back to the long-term evolution process, thereby enabling an interpretable and quantifiable assessment of the evolution of the fast / slow / charging structure ratio and the effects of single and combined subsidy policies.

[0025] The following description, in conjunction with the accompanying drawings, provides a more detailed explanation of the collaborative optimization method, system, and storage medium for charging infrastructure planning and charging scheduling disclosed in this embodiment.

[0026] Firstly, this invention provides a collaborative optimization method for charging infrastructure planning and charging scheduling, such as... Figure 1 As shown, it includes: Step S1: Obtain CPO information for the target region, initialize the parameters of the pre-constructed CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer; wherein, the two-layer model includes: CPO layer and EVCI layer; wherein, the CPO layer represents the competition or information interaction relationship between each CPO as a CPO relationship network.

[0027] Obtain CPO information for the target region. This CPO information includes: the number of CPOs, the market share of each CPO, the information transparency among CPOs, the initial FCS ratio strategy of each CPO, and the number of existing fast charging stations and slow charging stations of each CPO.

[0028] In detail, the steps of obtaining CPO information of the target region and initializing the parameters of the pre-built CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer include: Step S101: Obtain the number of CPOs in the target area, the market share of each CPO, the information transparency among CPOs, the initial FCS ratio strategy of each CPO, and the number of fast charging stations and slow charging stations in each CPO.

[0029] Step S102: Configure market share weights for each CPO in the CPO relationship network, and configure information transparency weights for the connections between each CPO.

[0030] The CPO-EVCI two-layer model in this embodiment includes an interconnected CPO layer and an EVCI layer. In the CPO layer, the competitive information or interaction relationships between CPOs within the target area are abstracted into a CPO relationship network. This describes the relationships between the various CPOs; where For the CPO set, This is the set of connections between CPOs. A market share weight is configured for each CPO. Transparency weights for configuration information of each connection This is used to characterize differences in market influence and differences in the intensity of information dissemination.

[0031] This invention constructs a two-layer complex network evolutionary game framework consisting of a CPO layer and an EVCI layer, which can characterize the differences in strategy learning and expansion of CPOs in a competitive environment, avoiding the bias caused by simplifying market players into a single centralized planner.

[0032] Preferably, the fast charging ratio strategy for each CPO is defined as follows: It can be represented in the form of a discrete strategy set or a continuous interval, as shown in Table 1.

[0033] Table 1. Optional CPO strategies based on FCS ratio

[0034] In this step of the CPO-EVCI two-layer model, the competition and information interaction relationships among multiple CPOs are abstracted into a complex network, and the fast charging ratio strategy of each CPO is regarded as a set of evolutionary game strategies to initialize the parameters of the two-layer model. The method in this embodiment uses the fast / slow charging structure ratio as the core evaluation index, which can directly reflect the long-term changes in the market environment and structure, and provide a quantitative basis for fast charging guidance policies, structural optimization, and resource allocation.

[0035] Step S2: The CPO layer connects the newly entered CPOs in each round of evolution to the CPO relationship network according to the pre-connection rules, and updates the strategy of the CPO relationship network using an evolutionary game mechanism to obtain the strategy update result of each round of evolution.

[0036] In each evolution cycle, the CPO layer introduces new CPOs to simulate market participants entering the market. Based on a preset priority connection mechanism, the new CPOs are connected with existing CPOs in the relationship network to obtain an updated CPO relationship network.

[0037] Specifically, the steps of connecting newly entering CPOs to the CPO relationship network according to pre-defined connection rules and updating the policy of the CPO relationship network to obtain the policy update result include: Step S21: The newly entered CPO establishes a connection with at least one CPO in the CPO relationship network according to the priority connection rule, so as to obtain an updated CPO relationship network.

[0038] Newly added CPOs are assigned an initial fast-charging ratio strategy upon entry (preferably set to a "medium" level, but can also be set to a preset constant or randomly initialized), and establish an information interaction connection with existing CPOs according to priority connection rules. Preferably, the probability of a new CPO establishing a connection with an existing CPO is... Market share weight of the CPO and its network connectivity Positive correlation can be represented as (1): (1) The CPO relationship network is completed using the above connection rules. The structural updates make it easier for existing CPOs with a large market share or in an information hub position to form information connections with new entities.

[0039] Step S22: After the structure of the CPO relationship network is updated, each CPO selects a strategy imitation object according to the neighbor selection mechanism; wherein, the neighbor selection mechanism takes the distance between two CPOs and the weight of information transparency as factors to consider when selecting a strategy imitation object.

[0040] Step S23: Each CPO calculates the strategy imitation probability using the Fermi function based on the difference in revenue between itself and the strategy imitation target, and determines its own fast charging station ratio based on the strategy imitation probability.

[0041] After constructing the new network, for any one Within its neighbor set, a potential policy imitation target is selected. To reflect the impact of differences in information transparency among CPOs on policy diffusion, preferably, the imitation target selection probability is determined by edge weights. With network connectivity The decision is made jointly, and can be expressed as formula (2): (2) in, The larger the value, the more likely it is to be a large number. and The higher the transparency of information between them, The greater the probability of being chosen as a model, the higher the chance of being imitated.

[0042] Although evolutionary game theory methods can be used to describe strategy evolution, existing evolutionary game theory methods are still based on the assumption of uniform mixing or static networks. The payoff functions and constraints are relatively simplified, making it difficult to form a consistent closed-loop description with actual site construction costs, policy subsidy mechanisms, and site-level construction constraints (such as capacity limits). This affects the explanatory power and transferability of long-term policy effects.

[0043] To ensure that strategy updates have an explainable economic driver, this embodiment defines the revenue of the CPO in the current evolutionary round as the net revenue of its fast and slow charging stations for that year. In this embodiment, The revenue is determined by the construction costs, operating electricity purchase costs, maintenance costs, and charging service revenue of its affiliated fast and slow charging stations. Furthermore, the annual equivalent profit is normalized according to the station scale to form revenue that can be used for strategic comparison. Among these, construction-driven subsidies apply to station construction and equipment investment, while demand / operation-driven subsidies apply to CPO electricity purchase costs and user-side settlement prices (or settlement compensation).

[0044] In this embodiment, The site collection is divided into fast charging type. With slow charging type The remaining sites at the end of the current evolutionary cycle are respectively and .in The new increment output by step S302 This is obtained by adding the stock from the previous period, which satisfies the following condition: (3) In this embodiment, the annualized cost per station Annualized cost of initial investment Operating electricity purchase cost Maintenance costs The composition is as shown in equation (4): (4) The initial investment annualized cost is calculated using the life-cycle method, and investment subsidies are introduced. To reflect the government's role in reducing upfront investments in equipment, installation, and land, we obtain equation (5): (5) in, For the initial equipment purchase cost, For the initial installation costs, This refers to the initial land investment costs. Regarding equipment subsidies, this embodiment supports two main mechanisms: one is a proportional subsidy (regardless of power level), using... The second is to provide subsidies based on power differentiation, using... The equipment purchase cost is calculated according to formula (6): (6) In the formula, For the cost of a single-station benchmark equipment, The rated power parameters are for the corresponding station type. These are the quantity parameters corresponding to this station type after the planning update. The operating electricity purchase cost is related to the station's annual equivalent charging volume; this embodiment introduces an operating subsidy. This is used to represent the subsidy deduction for CPO electricity purchase costs. In this embodiment, the typical day set is... , typical day The corresponding weight is (This represents the number of calendar days represented by the typical day), and the set of discrete time periods within a typical day is... The time step is Therefore, under typical daily time-of-use pricing conditions, the annual equivalent operating cost of a single station can be expressed as follows: (7) in, This indicates the time-of-use electricity price. Indicates station type A single station on a typical day Time period The charging capacity is calculated from the output of step S401. The maintenance cost is estimated as a fixed percentage of the annualized cost of the initial investment, and can be expressed as equation (8): (8) In this embodiment, the revenue of the CPO depends on the user's charging demand. Furthermore, user preferences for new technologies should not remain entirely constant. Therefore, user demand is calculated to demonstrate the attractiveness of EVCI and user preferences. Specifically, a discrete choice model with stochastic utility is employed to estimate user choices, thereby satisfying the demand for fast and slow charging. This estimation method is widely used in transportation to describe consumer behavioral choices. A typical day is defined. During the period Upper type The charging price is User charging replenishment Therefore, the actual charging electricity price paid by the user is: (9) The benefits of users choosing fast or slow charging can be represented as follows: (10) (11) here, It is the average electricity price of FCS. It is the average electricity price sold at slow charging stations. It is the effect of battery degradation caused by fast charging. This refers to the utility of waiting time caused by slow charging. A, b, c, and e are the coefficients of the various factors mentioned above, while d and f are constants. This involves user charging subsidies, which reduce charging prices for users to study the impact of user demand on FCS (Fulfilled Computer System) construction. Furthermore, The utility of users choosing fast charging increases with the number of FCSs in the network. The benefits of users choosing slow charging are detailed in (12) and (13): (12) (13) here, This indicates the total number of charging stations. This indicates the proportion of FCS in the network. This represents the energy service utility coefficient. Therefore, based on the Logit form, the probability of a user choosing fast / slow charging can be obtained: (14) (15) Based on the above analysis, the revenue of fast and slow charging stations can be represented as follows: (16) in For station type The annual equivalent charging capacity per station is calculated by weighting the time-of-use electricity consumption on a typical day. The average charging capacity per station during time-sharing is calculated from (38): (17) For station type The equivalent annual electricity sales price per unit station: (18) Based on the above revenue and costs, this embodiment will The annualized net income per station is defined by equation (19): (19) Furthermore, to avoid the difficulty in comparing revenue fluctuations caused by the small scale of newly entered CPOs, this embodiment uses a site-weighted average revenue based on the number of CPOs as the revenue input for strategy updates. Its average return is calculated according to formula (20): (20) In detail, the policy update based on the Fermi rule in this step includes the following steps: In obtaining the benefits of the CPO itself and its imitation objects Benefits Afterwards, CPO The fast charging ratio strategy is updated probabilistically according to the Fermi rule. Preferably, the strategy adoption probability can be expressed as: ;(twenty one) in This is an irrational parameter used to characterize the randomness of policy updates. When When triggered, The fast charging ratio strategy is updated to emulate the target strategy, or it is updated to align with the target strategy by a preset update range, thereby achieving strategy diffusion and evolution. After the strategy update is completed, the fast charging ratio strategy constraint given by the CPO layer for that year is formed, which serves as the input for the site-level planning solution.

[0045] Step S3: Under the constraint of the strategy update results in each round of evolution, the EVCI layer predicts the total charging demand in each preset planning period of the target area, and calculates the number of new fast charging stations and slow charging stations and investment costs for each CPO, so as to obtain the site scale and capacity parameters of each CPO in each round of evolution.

[0046] After completing step S2, the CPO layer has obtained the network structure, policy state, and revenue-based policy update results for this round of evolution. Then, it proceeds to step S3, where the traffic demand forecasting model is coupled with the AIM / Enduse model at the EVCI layer. Under the policy constraints and subsidy parameter constraints given by the CPO layer, the number of new fast-charging stations is calculated. With the number of newly added slow charging stations The number of each CPO site is updated to provide constraints for subsequent charging scheduling and revenue calculation.

[0047] In this step, based on the charging demand obtained from traffic demand forecasting, the CPO strategy is used as a structural constraint input to construct and solve a linear optimization model based on AIM / Enduse, obtaining the number of new fast charging stations and new slow charging stations added by each CPO in this evolution cycle; the number of new stations is accumulated into the existing number of stations of each CPO for subsequent operation scheduling and revenue calculation.

[0048] In detail, the EVCI layer, under the constraints of the strategy update results, predicts the total charging demand in the target area within each preset planning period, and calculates the number and investment cost of new fast charging stations and slow charging stations for each CPO, obtaining the site scale and capacity parameters of each CPO. Step S301: Under the constraints of the strategy update result, calculate the traffic travel demand in the target area using a preset traffic demand prediction model.

[0049] In one implementation, the traffic demand forecasting model is first constructed, specifically as follows: In each planning cycle To obtain the passenger transportation demand in the target area and further infer the electrification charging demand, a regional passenger transportation demand prediction model is used to calculate the total passenger trip volume.

[0050] The step of calculating the traffic demand within the target area using a preset traffic demand prediction model under the constraints of the strategy update result includes: The traffic demand forecasting model is constructed based on regional socio-economic variables and traffic supply variables. The regional socio-economic variables include per capita GDP, population size, and per capita built-up area. The traffic supply variables include per capita road supply and generalized travel costs.

[0051] Specifically, passenger transportation demand Determined by both regional socioeconomic variables and transportation supply variables, it can be expressed as: ; (twenty two) in, GDP per capita For per capita road supply, The per capita built-up area For the generalized cost of travel, For population size, These are the parameters for demand forecasting.

[0052] Step S302: Based on the transportation demand, calculate the total charging demand within the target area.

[0053] Based on traffic demand Further utilize EV penetration rate Average electricity consumption per trip Calculate the total charging demand (annual electricity demand) of the target area during the planning period. ,Right now: ;(twenty three) Step S303: Based on the total charging demand, use a preset energy system model to calculate the number of new fast charging stations and slow charging stations for each CPO, and add the number of new fast charging stations and slow charging stations to the existing stock of each CPO to obtain the site scale and capacity parameters of each CPO.

[0054] This step combines traffic demand forecasting models with energy system models to determine the expected number of EVCIs required to meet charging demand in each planning cycle and the corresponding cost of each CPO.

[0055] In detail, the step of calculating the number of new fast charging stations and slow charging stations for each CPO based on the total charging demand and using a preset energy system model includes: The energy system model aims to minimize the total cost of new EVCIs over a long time scale, and outputs the number of new FCSs, the number of new slow charging stations, and the cost for each CPO. The constraints of the energy system model include one or more of the following: supply and demand balance constraints, strategy consistency constraints, budget constraints, and cost composition constraints. Among them, the strategy consistency constraints ensure that the proportion of FCSs corresponding to the new stations falls within the strategy range given by the CPO layer.

[0056] The AIM / Enduse model is a bottom-up energy system model capable of long-term, multi-year, and multi-scenario simulations with high temporal resolution. This capability allows the model to integrate evolving technological costs, policy interventions, and socioeconomic trends, as detailed below: The AIM / Enduse model is used to optimize the number of fast and slow charging stations based on a detailed technology database to minimize the total cost of the EVCI network. This database includes information on each charging technology option, such as lifetime, initial construction cost, operating and maintenance costs, and energy efficiency. Specifically, the construction of charging station types is formulated as a linear optimization problem with several practical constraints to meet a given total charging demand. To achieve the overall goal of minimizing total cost, dual constraints of upper-level strategy and initial construction cost are imposed to optimize and obtain the expected number of new EVCIs to meet internal charging demand, as shown in (24)-(30): ;(twenty four) st (25) (26) (27) (28) (29) (30) The objective function (24) represents the total cost of the two types of charging stations. Constraints (25)-(27) provide the model for the three basic components of the total cost function, respectively. Constraint (28) reveals the supply-demand balance, which means that the number of newly added fast and slow charging stations must meet the predicted load demand. Constraint (29) limits the proportion of lower-level FCSs to be consistent with the construction strategy of the corresponding upper-level CPOs. Here, and represent the upper and lower limits of the FCS proportion strategy, respectively. Constraint (30) describes the funding constraints for each CPO, indicating that the construction cost in each CPO does not exceed its available budget.

[0057] Traffic demand forecasting models feed charging demand into the AIM / Enduse model, which estimates the number of new fast and slow charging stations. EVCI construction decisions and revenue for each CPO are fed back to the upper levels to assist the respective CPOs in assessing the feasibility of their strategies and making adaptive updates for the next evolutionary cycle.

[0058] Step S4: Based on the scale and capacity parameters of each CPO site in each round of evolution, construct a charging scheduling optimization model for a typical day. With the goal of maximizing CPO profits, calculate the charging power and queue allocation of electric vehicles in different time periods and different types of charging stations to obtain charging station control information and typical daily operating revenue under different investment cost scenarios.

[0059] This step is used to implement typical daily operation scheduling on a short time scale. It constructs a set of typical days and their weights to represent operation scenarios with different seasons and load patterns. Within each typical day, a charging scheduling optimization model is established. With the goal of maximizing CPO profits, the allocation of charging resources, charging power, and service processes within the station are scheduled and verified. Under the premise of meeting the station capacity and related operational constraints, the operation revenue and service level indicators of each CPO under typical day conditions are obtained. Based on the typical day weights, the operation results are converted into the annual revenue or profit indicators of the CPO.

[0060] In detail, based on the scale and capacity parameters of each CPO site, a charging scheduling optimization model for a typical day is constructed. With the goal of maximizing CPO profits, the steps involved in obtaining charging station control information and typical daily operating revenue under different investment cost scenarios, by allocating charging power and queueing among electric vehicles at different time periods and different types of charging stations, include: Step S401: Construct a set of typical days and the weights of each set of typical days based on the scale and capacity parameters of each CPO site.

[0061] Charging scheduling optimization on a short time scale uses the updated site stock and capacity parameters in step S401 as constraints to construct a typical daily set and its weights.

[0062] Step S402: Based on the intraday time scale of the selected typical day, construct a charging scheduling optimization model for the typical day.

[0063] A charging scheduling optimization model is established at the intraday time scale of a typical day.

[0064] Step S403: With the goal of maximizing CPO profits, based on the charging scheduling optimization model, obtain the optimized operating revenue and cost of each CPO.

[0065] With the goal of maximizing CPO profits, the system comprehensively considers time-of-use pricing, user queuing characteristics, and distribution network capacity limitations to optimize the charging power and allocation of vehicles at different times and different types of charging stations, thereby obtaining the operating revenue and costs of each station and each CPO. Step S404: Calculate the annual profit or revenue of each CPO by weighting the operating revenue and costs of each CPO.

[0066] Step S405: Feed back the annual profit or revenue of each CPO to the CPO layer to update the market share weight of each CPO, and use the updated market share weight of each CPO as the revenue input for the next round of strategy evolution.

[0067] The annual revenue of each CPO obtained in step S404 is fed back to the CPO layer in step S2 to update the market share weight of each CPO. And serve as the profit input for the next round of strategy evolution.

[0068] Step S406: Repeatedly update the strategy of the CPO relationship network based on the updated revenue input to obtain the strategy update result, calculate the number of new fast charging stations and slow charging stations and investment costs for each CPO, obtain the site size and capacity parameters of each CPO, take the site size and capacity parameters of each CPO as constraints, take the CPO profit maximization as the objective, and based on the charging scheduling optimization model, obtain the optimized operating revenue and cost of each CPO, until the preset evolution termination condition is met, and obtain the charging station control information and typical daily operating revenue under different investment cost scenarios.

[0069] The process repeats steps S2 to S406 until the preset evolution cycle limit is reached, outputting the FCS ratio evolution curve, final FCS ratio index, and typical daily operating revenue under different single subsidy and combined subsidy scenarios.

[0070] In one implementation, a queue energy state variable is introduced to characterize the daily demand accumulation caused by arrivals and queuing. Indicates time period Unmet charging needs will be addressed by introducing unserviceable energy. As a feasibility relaxation, its size is constrained by a penalty term. Define decision variables. For typical days time period Internal, CPO At the node station type The average charging power per station, and the corresponding time-of-use charging capacity per station, are denoted as: .make This represents the traffic demand forecast and station allocation rules obtained on a typical day. Time period Arrival at CPO ,node Station type The energy required for charging. Therefore, the scheduling model for step S401 can be written as: (31) st (32) (33) (34) (35) (36) (37) (38) Wherein, equation (31) represents the planning period. The goal of maximizing annual equivalent operating profit; The unserviceable energy penalty coefficient is used to suppress the passage of energy. The escape constraint leads to an unrealistic feasible solution. Equations (32) and (33) respectively give the power-to-electricity conversion relationship and the upper limit of power per station. Equations (34)-(36) describe the energy required to reach the destination. The intraday queuing accumulation process is described, with the queue being cleared at the beginning and end of a typical day to characterize the intraday closed loop. Formula (37) represents the capacity constraint of the distribution network nodes. Represents a node On a typical day Time period The maximum available capacity that can be used for charging loads.

[0071] The strategy update results in this embodiment include: the FCS ratio and subsidy policy parameters of each charging station; the subsidy policy parameters include: construction-driven subsidies and demand / operation-driven subsidies, wherein construction-driven subsidies include investment subsidies and equipment subsidies, and demand / operation-driven subsidies include operation subsidies and user charging subsidies.

[0072] This disclosure pertains to the field of electric vehicle charging infrastructure (EVCI) planning and operation scheduling technology, proposing a hierarchical planning and scheduling method for charging stations that considers multi-strategy subsidies and the evolution of operational behavior. Addressing the common problem in existing research that neglects the affiliation between EVCI and charging operators (CPOs), making it difficult to characterize the long-term impact of various subsidy combinations on site layout and the ratio of fast and slow charging stations, this invention constructs a two-layer model of "upper-layer CPO—lower-layer EVCI". The upper layer describes the dynamic evolution of the CPO network over multiple years based on complex network evolution game theory, including the entry of new CPOs, information propagation and strategy imitation among existing CPOs, and evolves the fast charging station (FCS) ratio strategy and construction scale of each CPO under various single and combined subsidy scenarios such as investment subsidies, equipment subsidies, operation subsidies, and user charging subsidies. The lower layer couples a traffic demand model with an AIM / Enduse energy system model, predicting the total charging demand for each year under the constraints of the upper-layer strategy and calculating the number of new fast and slow charging stations and investment costs for each CPO.

[0073] Building upon this foundation, the present invention further introduces a second-stage charging scheduling module based on typical days: Several representative typical days are selected, and the station capacity obtained from the upper-level planning is used as a constraint. A charging scheduling optimization model is established for each typical day, comprehensively considering time-of-use pricing, differences in fast and slow charging power, user arrival and queuing characteristics, and distribution network capacity limitations. This optimizes the charging power and queuing allocation of electric vehicles (EVs) across different time periods and different types of charging stations, resulting in more refined operating costs and revenues for each CPO. The above operating results are weighted according to typical days and converted into annual revenue, which is then fed back into the upper-level CPO evolutionary game to update CPO market share and strategies, achieving interactive iteration between planning decisions and intraday operations.

[0074] The method disclosed in this embodiment can simultaneously characterize the evolution of CPO strategy, EVCI spatial layout, and charging scheduling process during a typical day, revealing the long-term impact of different construction-driven subsidies and demand / operation-driven subsidies and their combinations on the evolution of FCS ratio, providing systematic and implementable technical support for governments to formulate differentiated subsidy combination strategies and optimize FCS planning and scheduling.

[0075] Secondly, this invention provides a collaborative optimization system for charging infrastructure planning and charging scheduling, such as... Figure 2 As shown, it includes: The initialization module 210 is used to obtain CPO information of the target area and initialize the parameters of the pre-constructed CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer; wherein, the two-layer model includes: CPO layer and EVCI layer; wherein, the CPO layer represents the competition or information interaction relationship between each CPO as a CPO relationship network; its function is as described in step S1.

[0076] The strategy evolution module 220 is used to connect newly entered CPOs in each round of evolution to the CPO relationship network according to pre-connection rules, and to update the strategy of the CPO relationship network using an evolutionary game mechanism to obtain the strategy update result of each round of evolution; its function is as described in step S2.

[0077] The site information calculation module 230 is used by the EVCI layer to predict the total charging demand in each preset planning period of the target area under the constraint of the strategy update result in each round of evolution, and to calculate the number of new fast charging stations and slow charging stations and investment costs of each CPO, so as to obtain the site scale and capacity parameters of each CPO in each round of evolution; its function is as described in step S3.

[0078] The charging scheduling optimization module 240 is used to construct a typical day's charging scheduling optimization model based on the scale and capacity parameters of each CPO site in each round of evolution. With the goal of maximizing CPO profits, it allocates the charging power and queueing of electric vehicles among different types of charging stations in different time periods to obtain charging station control information and typical day operating revenue under different investment cost scenarios. Its function is as described in step S4.

[0079] This disclosure pertains to the field of electric vehicle charging infrastructure (EVCI) planning and operation scheduling technology, proposing a hierarchical planning and scheduling method for charging stations that considers multi-strategy subsidies and the evolution of operational behavior. Addressing the common problem in existing research that neglects the affiliation between EVCI and charging operators (CPOs), making it difficult to characterize the long-term impact of various subsidy combinations on site layout and the ratio of fast and slow charging stations, this invention constructs a two-layer model of "upper-layer CPO—lower-layer EVCI". The upper layer describes the dynamic evolution of the CPO network over multiple years based on complex network evolution game theory, including the entry of new CPOs, information propagation and strategy imitation among existing CPOs, and evolves the fast charging station (FCS) ratio strategy and construction scale of each CPO under various single and combined subsidy scenarios such as investment subsidies, equipment subsidies, operation subsidies, and user charging subsidies. The lower layer couples a traffic demand model with an AIM / Enduse energy system model, predicting the total charging demand for each year under the constraints of the upper-layer strategy and calculating the number of new fast and slow charging stations and investment costs for each CPO.

[0080] Building upon this foundation, the present invention further introduces a second-stage charging scheduling module based on typical days: Several representative typical days are selected, and the station capacity obtained from the upper-level planning is used as a constraint. A charging scheduling optimization model is established for each typical day, comprehensively considering time-of-use pricing, differences in fast and slow charging power, user arrival and queuing characteristics, and distribution network capacity limitations. This optimizes the charging power and queuing allocation of electric vehicles (EVs) across different time periods and different types of charging stations, resulting in more refined operating costs and revenues for each CPO. The above operating results are weighted according to typical days and converted into annual revenue, which is then fed back into the upper-level CPO evolutionary game to update CPO market share and strategies, achieving interactive iteration between planning decisions and intraday operations.

[0081] The embodiments disclosed herein can simultaneously characterize the evolution of CPO strategy, EVCI spatial layout, and charging scheduling process during a typical day, revealing the long-term impact of different construction-driven subsidies and demand / operation-driven subsidies and their combinations on the evolution of FCS ratio, providing systematic and implementable technical support for governments to formulate differentiated subsidy combination strategies and optimize FCS planning and scheduling.

[0082] Thirdly, the present invention provides a computer storage medium, which is a computer-readable storage medium, wherein a computer program is stored in the computer storage medium, and when the computer program is executed by a computer, the computer is used to execute the aforementioned collaborative optimization method for charging infrastructure planning and charging scheduling.

[0083] This invention provides a collaborative optimization method, system, and storage medium for charging infrastructure planning and charging scheduling. By acquiring CPO information for a target area, parameters are initialized on a pre-constructed CPO-EVCI two-layer model, resulting in initialized CPO and EVCI layers. The CPO layer connects newly entering CPOs to the CPO relationship network according to pre-defined connection rules and updates the CPO relationship network with strategies, obtaining strategy update results. Under the constraints of the strategy update results, the EVCI layer predicts the total charging demand in each preset planning period of the target area and calculates the number of new fast-charging and slow-charging stations and investment costs for each CPO, obtaining the site size and capacity parameters for each CPO. Based on the site size and capacity parameters of each CPO, a typical day's charging scheduling optimization model is constructed. With the goal of maximizing CPO profits, the charging power and queue allocation of electric vehicles across different time periods and different types of charging stations are analyzed to obtain charging station control information and typical day operating revenue under different investment cost scenarios. The method and system provided by this invention construct a CPO-EVCI two-layer model over a longer time scale to characterize the strategy learning and expansion differences of CPOs in a competitive environment, avoiding single-entity decision-making. Furthermore, it performs typical daily operation scheduling over a short time scale so that changes in operational constraints, service levels, and revenues can be reflected in the long-term evolution process in a traceable manner. This provides quantifiable technical support for differentiated subsidy design and the efficient and orderly development of EVCI.

[0084] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0085] The embodiments described above are merely examples of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application.

Claims

1. A collaborative optimization method for charging infrastructure planning and charging scheduling, characterized in that, include: Obtain CPO information for the target region, initialize the parameters of the pre-constructed CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer; wherein, the two-layer model includes: CPO layer and EVCI layer; wherein, the CPO layer represents the competition or information interaction relationship between each CPO as a CPO relationship network; The CPO layer connects newly entered CPOs in each round of evolution to the CPO relationship network according to pre-connection rules, and updates the strategy of the CPO relationship network using an evolutionary game mechanism to obtain the strategy update result of each round of evolution; Under the constraint of the strategy update results in each round of evolution, the EVCI layer predicts the total charging demand in each preset planning period of the target area, and calculates the number of new fast charging stations and slow charging stations and investment costs for each CPO, so as to obtain the site scale and capacity parameters of each CPO in each round of evolution. Based on the scale and capacity parameters of each CPO site in each round of evolution, a typical day's charging scheduling optimization model is constructed. With the goal of maximizing CPO profits, the charging power and queue allocation of electric vehicles in different time periods and different types of charging stations are evaluated to obtain charging station control information and typical day operating revenue under different investment cost scenarios.

2. The collaborative optimization method for charging infrastructure planning and charging scheduling according to claim 1, characterized in that, The steps of obtaining CPO information of the target region and initializing the parameters of the pre-built CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer include: Obtain the number of CPOs in the target area, the market share of each CPO, the information transparency among CPOs, the initial FCS ratio strategy of each CPO, and the number of fast charging stations and slow charging stations in each CPO. Configure market share weights for each CPO in the CPO relationship network, and configure information transparency weights for the connections between CPOs.

3. The collaborative optimization method for charging infrastructure planning and charging scheduling according to claim 1, characterized in that, The steps of connecting newly entering CPOs in each round of evolution to the CPO relationship network according to pre-connection rules, and updating the policy of the CPO relationship network using an evolutionary game mechanism to obtain the policy update result for each round of evolution include: In each round of evolution, newly entering CPOs are connected to at least one CPO in the CPO relationship network according to the priority connection rule to obtain an updated CPO relationship network; After the structure of the CPO relationship network is updated, each CPO selects a strategy imitation object according to the neighbor selection mechanism; wherein, the neighbor selection mechanism takes the proximity of the connection relationship between two CPOs and the weight of information transparency as factors to consider when selecting a strategy imitation object; Each CPO calculates the strategy imitation probability using the Fermi function based on the difference in revenue between itself and the strategy imitation target, and determines its own fast charging station ratio based on the strategy imitation probability.

4. The collaborative optimization method for charging infrastructure planning and charging scheduling according to claim 1, characterized in that, Under the constraint of the strategy update results in each round of evolution, the EVCI layer predicts the total charging demand in the target area within each preset planning period, and calculates the number of new fast charging stations and slow charging stations and investment costs for each CPO, obtaining the site scale and capacity parameters of each CPO in each round of evolution. Under the constraints of the strategy update results, the traffic demand within the target area is calculated using a preset traffic demand prediction model. Based on the aforementioned transportation demand, the total charging demand within the target area is calculated. Based on the total charging demand, the number of new fast charging stations and slow charging stations for each CPO is calculated using a preset energy system model. The number of new fast charging stations and slow charging stations is then added to the existing stock of each CPO to obtain the site size and capacity parameters of each CPO.

5. The collaborative optimization method for charging infrastructure planning and charging scheduling according to claim 1, characterized in that, The step of calculating the traffic demand within the target area using a preset traffic demand prediction model under the constraints of the strategy update result includes: The traffic demand forecasting model is constructed based on regional socio-economic variables and traffic supply variables. The regional socio-economic variables include per capita GDP, population size, and per capita built-up area. The traffic supply variables include per capita road supply and generalized travel costs.

6. The collaborative optimization method for charging infrastructure planning and charging scheduling according to claim 5, characterized in that, The step of calculating the number of new fast charging stations and slow charging stations for each CPO based on the total charging demand and using a preset energy system model includes: The energy system model aims to minimize the total cost of new EVCIs over a long time scale, and outputs the number of new FCSs, the number of new slow charging stations, and the cost for each CPO. The constraints of the energy system model include one or more of the following: supply and demand balance constraints, strategy consistency constraints, budget constraints, and cost composition constraints. Among them, the strategy consistency constraints ensure that the proportion of FCSs corresponding to the new stations falls within the strategy range given by the CPO layer.

7. The collaborative optimization method for charging infrastructure planning and charging scheduling according to claim 1, characterized in that, The steps of constructing a typical day's charging scheduling optimization model based on the scale and capacity parameters of each CPO site in each round of evolution, with the goal of maximizing CPO profits, and obtaining charging station control information and typical day operating revenue under different investment cost scenarios by allocating charging power and queueing allocation for electric vehicles at different time periods and different types of charging stations, include: Based on the scale and capacity parameters of each CPO site, a set of typical days and the weights of each set of typical days are constructed. Based on the intraday time scale of a selected typical day, a charging scheduling optimization model for the typical day is constructed. With the goal of maximizing CPO profits, the optimized operating revenue and costs of each CPO are obtained based on the aforementioned charging scheduling optimization model. The annual profit or revenue of each CPO is calculated by weighting the operating revenue and costs of each CPO. The annual profit or revenue of each CPO is fed back to the CPO layer to update the market share weight of each CPO, and the updated market share weight of each CPO is used as the revenue input for the next round of strategy evolution. The strategy update process is repeated on the CPO relationship network based on the updated revenue input. The strategy update results are obtained, the number of new fast charging stations and slow charging stations and the investment cost of each CPO are calculated, and the site size and capacity parameters of each CPO are obtained. With the site size and capacity parameters of each CPO as constraints and the goal of maximizing CPO profits, the optimized operating revenue and cost of each CPO are obtained based on the charging scheduling optimization model until the preset evolution termination condition is met. The charging station control information and typical daily operating revenue under different investment cost scenarios are obtained.

8. The collaborative optimization method for charging infrastructure planning and charging scheduling according to claim 1, characterized in that, The strategy update results include: the FCS ratio and subsidy policy parameters for each charging station; the subsidy policy parameters include: construction-driven subsidies and demand / operation-driven subsidies, wherein construction-driven subsidies include investment subsidies and equipment subsidies, and demand / operation-driven subsidies include operation subsidies and user charging subsidies.

9. A collaborative optimization system for charging infrastructure planning and charging scheduling, characterized in that, include: An initialization module is used to acquire CPO information of the target area and initialize the parameters of a pre-constructed CPO-EVCI two-layer model to obtain the initialized CPO layer and EVCI layer; wherein, the two-layer model includes: a CPO layer and an EVCI layer; wherein, the CPO layer represents the competition or information interaction relationship between each CPO as a CPO relationship network; The strategy evolution module is used by the CPO layer to connect newly entered CPOs in each round of evolution to the CPO relationship network according to pre-connection rules, and to update the strategy of the CPO relationship network using an evolutionary game mechanism to obtain the strategy update result of each round of evolution; The site information calculation module is used by the EVCI layer to predict the total charging demand in each preset planning period of the target area under the constraint of the strategy update result in each round of evolution, and to calculate the number of new fast charging stations and slow charging stations and investment costs of each CPO, so as to obtain the site scale and capacity parameters of each CPO in each round of evolution. The charging scheduling optimization module is used to construct a typical day's charging scheduling optimization model based on the scale and capacity parameters of each CPO site in each round of evolution. With the goal of maximizing CPO profits, it obtains charging station control information and typical day operating revenue under different investment cost scenarios by allocating charging power and queuing for electric vehicles in different time periods and different types of charging stations.

10. A computer storage medium, wherein the storage medium is a computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a computer, enables the computer to perform the collaborative optimization method for charging infrastructure planning and charging scheduling as described in any one of claims 1-8.

Images