A two-layer optimization method and device based on mobile charging station leasing model

Abstract

The present invention discloses a two-layer optimization method and device based on a mobile charging station leasing model. The method constructs a two-layer game model based on the mobile charging station leasing model; the two-layer game model includes a TMCO model and a CFO two-stage DRCC model; a Wasserstein distance distributional robustness model based on opportunity constraints is introduced through a distributional robustness optimization strategy; the economy and robustness of the two-layer game model are balanced by the Wasserstein distance distributional robustness model; the TMCO model is solved by a genetic algorithm to obtain a configuration and leasing price that maximizes the benefits; the CFO two-stage DRCC model is solved by a nested column and constraint generation algorithm to obtain a leasing plan and scheduling strategy that maximizes the utility. The present invention can effectively balance the needs of various operators, while ensuring the benefits of TMCS operators, improving the economic efficiency of charging facility operators.

Description

Technical Field

[0001] The present invention relates to the technical field of charging facility optimization, and in particular to a double-layer optimization method and device based on a mobile charging station leasing model. Background Art

[0002] Emerging mobile charging stations (TMCS) integrate a number of charging piles and energy storage battery packs in a container loaded on a truck. As an effective supplement to traditional fixed charging facilities, TMCS are more flexible and scalable than fixed charging stations. However, the commercial operation of TMCS still faces problems such as high initial costs and low utilization rates, which hinder operators' investment enthusiasm.

[0003] Currently, charging infrastructure operators typically employ two main business models. The first is "self-build," whereby the operator invests in and manages the charging stations themselves, as exemplified by companies like Tesla in the United States and NIO and StarCharge in China. This model involves high investment costs, but the operator maintains full control over operations and revenue. The second model, known as the "owner-operator" model, involves the owner providing charging infrastructure and outsourcing services to the operator, as exemplified by companies like EVgo. This ensures revenue security for the owner while transferring operational risk, which faces uncertainty in charging demand and revenue, to the operator, allowing the operator to expand its charging services without incurring significant upfront investments. Despite the existence of these business models in the charging infrastructure sector, the rollout of TMCS presents unique challenges. Because TMCS primarily serves as a temporary supplement to fixed charging infrastructure, this auxiliary role complicates its deployment. TMCS operators typically face high initial costs, a long payback period, and uncertain demand. These challenges could impact operator profits and further TMCS adoption.

[0004] Therefore, how to invent an optimization method based on the mobile charging station leasing model that can effectively balance the needs of various operators and improve the economic efficiency of charging facility operators while ensuring the profits of TMCS operators has become an urgent problem to be solved. Summary of the Invention

[0005] To this end, the present invention provides a two-layer optimization method and device based on the mobile charging station leasing model, which can effectively balance the needs of various operators and improve the economic efficiency of the charging facility operator (CFO) while ensuring the revenue of the TMCS operator (TMCO) through adaptive pricing and resource allocation strategies.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a two-layer optimization method based on a mobile charging station leasing model, comprising:

[0007] Based on the mobile charging station leasing model, a two-tier game model is constructed; the two-tier game model includes the TMCO model and the CFO two-stage DRCC model;

[0008] A Wasserstein distance distributional robustness model based on chance constraints is introduced through a distributional robustness optimization strategy; the economy and robustness of the two-level game model are balanced through the Wasserstein distance distributional robustness model;

[0009] The TMCO model is solved by genetic algorithm to obtain the profit-maximizing configuration and leasing price; the CFO two-stage DRCC model is solved by nested column and constraint generation algorithm to obtain the utility-maximizing leasing plan and scheduling strategy.

[0010] As a preferred solution of the two-layer optimization method based on the mobile charging station leasing model, the optimization objective of the TMCO model is to maximize the net profit; the parameters of the TMCO model include:

[0011] TMCS rental revenue; TMCO's profit function, representing the difference between the total rental and energy arbitrage revenue and the investment, operation, maintenance, and loss costs; TMCS profit from energy arbitrage; TMCS investment and operation and maintenance costs; TMCS set, and subsets for short-term and long-term rentals; rental period and planning cycle; TMCS number and total number; TMCSω energy storage battery cost, energy storage battery replacement cost, charging pile and inverter cost, other accessory costs such as trucks and containers, garage cost, and operation and maintenance costs; TMCS system, energy storage battery, and maintenance cost conversion factor; operating life of energy storage batteries and TMCS system; discount rate based on the planning cycle; replacement sequence number and total number of energy storage battery replacements;

[0012] The operational constraint parameters of the TMCO model include:

[0013] The dispatch time and dispatch period; the maximum number of TMCSs to be deployed, determined by the TMCO's budget; the price cap determined by CFOj's self-purchase cost and leasing preference; the amortized self-purchase cost converted to the dispatch period; the adjustment coefficient; the electricity price per kWh when the TMCS returns to the warehouse for energy replenishment; the node electricity price at which the TMCS participates in the day-ahead energy market for energy arbitrage; the operating location of the TMCS, that is, the energy arbitrage node where it interacts with the grid; the charging and discharging power of TMCSω at node n at time th; the time when the TMCS ends its service and the corresponding garage; the total distance traveled during the dispatch period; the average driving speed of the TMCS; the energy consumption per kilometer of the TMCS; the labor cost of TMCSω; the marginal aging cost of the TMCS life cycle; the calendar aging parameters of the TMCS battery pack; the year corresponding to the commissioning of the TMCS; the maximum charging and discharging power; the charging and discharging efficiency of the TMCS; the capacity of TMCSω; the SOC value of the TMCS at the dispatch time; the maximum and minimum SOC values ​​of the TMCS.

[0014] As a preferred solution of the two-layer optimization method based on the mobile charging station leasing model, the optimization goal of the CFO two-stage DRCC model is to maximize utility; the parameters of the CFO two-stage DRCC model include:

[0015] The utility function of CFOj, i.e., maximizing the operator's profit while ensuring the charging service quality requirements; the set of TMCSs, short-term and long-term rentals of CFOj; the EV charging demand of TMCS charging at node m; the charging demand fuzzy set; the optimization variables; the EV charging service revenue; the TMCS rental cost; the TMCS operating cost; the EV charging electricity cost; the charging and discharging power of TMCSω at nodes n and m at time th; the corresponding EV charging service nodes in the road network; the TMCS travel distance obtained by the shortest path method;

[0016] The constraint parameters of the CFO two-stage DRCC model include:

[0017] The total number of leased units of CFOj at time t; the maximum output power of TMCSω when providing EV charging service is determined by the number and rated power of charging piles of TMCS; the EV charging demand response ratio of CFOj reflects the charging service quality preference of CFO;

[0018] As a preferred solution of the two-level optimization method based on the mobile charging station leasing model, in the process of solving the CFO two-stage DRCC model through the nested column and constraint generation algorithm, the CFO two-stage DRCC model solution problem is converted into a main problem MP and a sub-problem SP.

[0019] As a preferred solution of the two-layer optimization method based on the mobile charging station leasing model, in the process of solving the sub-problem SP, the sub-problem SP is decomposed into a main problem subset MPS and a sub-problem subset SPS.

[0020] The present invention also provides a two-layer optimization device based on the mobile charging station leasing model, which adopts the above-mentioned two-layer optimization method based on the mobile charging station leasing model, including:

[0021] A two-tier game model construction module is used to construct a two-tier game model based on the mobile charging station leasing model; the two-tier game model includes a TMCO model and a CFO two-stage DRCC model;

[0022] A Wasserstein distance distribution robustness model processing module is used to introduce a Wasserstein distance distribution robustness model based on chance constraints through a distribution robustness optimization strategy; the Wasserstein distance distribution robustness model is used to balance the economy and robustness of the two-layer game model;

[0023] A two-layer game model solving module is used to solve the TMCO model using a genetic algorithm to obtain a profit-maximizing configuration and rental price; and to solve the CFO two-stage DRCC model using a nested column and constraint generation algorithm to obtain a utility-maximizing rental solution and scheduling strategy.

[0024] As a preferred solution of a two-layer optimization device based on the mobile charging station leasing model, in the two-layer game model construction module, the optimization goal of the TMCO model is to maximize the net profit; the optimization goal of the TMCO model is to maximize the net profit; the parameters of the TMCO model include:

[0025] TMCS rental revenue; TMCO's profit function, representing the difference between the total rental and energy arbitrage revenue and the investment, operation, maintenance, and loss costs; TMCS profit from energy arbitrage; TMCS investment and operation and maintenance costs; TMCS set, and subsets for short-term and long-term rentals; rental period and planning cycle; TMCS number and total number; TMCSω energy storage battery cost, energy storage battery replacement cost, charging pile and inverter cost, other accessory costs such as trucks and containers, garage cost, and operation and maintenance costs; TMCS system, energy storage battery, and maintenance cost conversion factor; operating life of energy storage batteries and TMCS system; discount rate based on the planning cycle; replacement sequence number and total number of energy storage battery replacements;

[0026] The operational constraint parameters of the TMCO model include:

[0027] The dispatch time and dispatch period; the maximum number of TMCSs to be deployed, determined by the TMCO's budget; the price cap determined by CFOj's self-purchase cost and leasing preference; the amortized self-purchase cost converted to the dispatch period; the adjustment coefficient; the electricity price per kWh when the TMCS returns to the warehouse for energy replenishment; the node electricity price at which the TMCS participates in the day-ahead energy market for energy arbitrage; the operating location of the TMCS, that is, the energy arbitrage node where it interacts with the grid; the charging and discharging power of TMCSω at node n at time th; the time when the TMCS ends its service and the corresponding garage; the total distance traveled during the dispatch period; the average driving speed of the TMCS; the energy consumption per kilometer of the TMCS; the labor cost of TMCSω; the marginal aging cost of the TMCS life cycle; the calendar aging parameters of the TMCS battery pack; the year corresponding to the commissioning of the TMCS; the maximum charging and discharging power; the charging and discharging efficiency of the TMCS; the capacity of TMCSω; the SOC value of the TMCS at the dispatch time; the maximum and minimum SOC values ​​of the TMCS.

[0028] As a preferred solution of the two-layer optimization device based on the mobile charging station leasing model, in the two-layer game model construction module, the optimization goal of the CFO two-stage DRCC model is to maximize utility; the parameters of the CFO two-stage DRCC model include:

[0029] The utility function of CFOj, i.e., maximizing the operator's profit while ensuring the charging service quality requirements; the set of TMCSs, short-term and long-term rentals of CFOj; the EV charging demand of TMCS charging at node m; the charging demand fuzzy set; the optimization variables; the EV charging service revenue; the TMCS rental cost; the TMCS operating cost; the EV charging electricity cost; the charging and discharging power of TMCSω at nodes n and m at time th; the corresponding EV charging service nodes in the road network; the TMCS travel distance obtained by the shortest path method;

[0030] The constraint parameters of the CFO two-stage DRCC model include:

[0031] The total number of leased units of CFOj at time t; the maximum output power of TMCSω when providing EV charging service is determined by the number and rated power of charging piles of TMCS; the EV charging demand response ratio of CFOj reflects the charging service quality preference of CFO;

[0032] As an optimal solution for a two-layer optimization device based on the mobile charging station leasing model, in the two-layer game model solving module, in the process of solving the CFO two-stage DRCC model through the nested column and constraint generation algorithm, the CFO two-stage DRCC model solving problem is converted into a main problem MP and a sub-problem SP.

[0033] As a preferred solution of the two-layer optimization device based on the mobile charging station leasing model, in the two-layer game model solving module, in the process of solving the sub-problem SP, the sub-problem SP is decomposed into a main problem subset MPS and a sub-problem subset SPS.

[0034] The present invention has the following advantages: the present invention constructs a two-layer game model based on the mobile charging station leasing model; the two-layer game model includes a TMCO model and a CFO two-stage DRCC model; a Wasserstein distance distributional robustness model based on opportunity constraints is introduced through a distributional robustness optimization strategy; the Wasserstein distance distributional robustness model is used to balance the economy and robustness of the two-layer game model; the TMCO model is solved by a genetic algorithm to obtain a configuration and leasing price that maximizes revenue; the CFO two-stage DRCC model is solved by a nested column and constraint generation algorithm to obtain a leasing plan and scheduling strategy that maximizes utility. The present invention proposes a two-layer optimization method framework based on the TMCS leasing model. The upper layer is to maximize the revenue of the TMCS operator (TMCO). The TMCO sets long-term and short-term rental packages based on the difference in leasing demand, and weighs the leasing prices and TMCS configuration quantity of different participants, and schedules them to participate in grid energy arbitrage during the idle period of the TMCS; the lower layer is aimed at maximizing the utility of the charging facility operator (CFO), responds to the leasing package and returns the lease quantity and lease time. This paper introduces the Wasserstein distance distribution robustness (DRCC) model based on chance constraints, using the distributionally robust optimization (DRO) approach. This model fully accounts for the impact of charging forecast information errors on the optimization results. Based on the conditional value at risk (CVaR) approximation method, the present invention employs a genetic algorithm and a nested column constraint generation (NC&CG) algorithm for solution at the upper and lower levels of the model, respectively. Through adaptive pricing and resource allocation strategies, the present invention effectively balances the needs of TMCO and CFO, ensuring TMCO profitability while improving the economic efficiency of CFO. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can, without inventive effort, derive other implementation drawings based on the provided drawings.

[0036] The structures, proportions, sizes, etc. illustrated in this specification are intended solely to complement the contents disclosed herein and to facilitate understanding and reading by persons skilled in the art. They are not intended to limit the conditions under which the present invention may be implemented and therefore have no substantive technical significance. Any structural modifications, changes in proportions, or adjustments in sizes, without affecting the efficacy and objectives of the present invention, shall remain within the scope of the technical contents disclosed herein.

[0037] Figure 1 This is a flow chart of a two-layer optimization method based on a mobile charging station leasing model provided in Example 1 of the present invention;

[0038] Figure 2 This is a schematic diagram of a specific implementation framework of a two-layer optimization method based on a mobile charging station leasing model provided in Example 1 of the present invention;

[0039] Figure 3 This is a schematic diagram of a model solution process in a two-layer optimization method based on a mobile charging station leasing model provided in Example 1 of the present invention;

[0040] Figure 4 This is a schematic diagram of the distribution of the operating locations of the ring highway network and fixed charging stations of various operators in a possible embodiment provided in Example 1 of the present invention;

[0041] Figure 5 This is a schematic diagram of economic indicators of each operator under different working conditions in a possible embodiment provided in Example 1 of the present invention; wherein (a) represents the number of TMCS and the rental price; (b) represents the cost and profit;

[0042] Figure 6 Schematic diagram of charging demand and TMCS operating status in a possible embodiment provided in Example 1 of the present invention; wherein (a) is the charging demand on typical day 1; (b) is the TMCS operating status on typical day 1; (c) is the charging demand on typical day 2; and (d) is the TMCS operating status on typical day 2.

[0043] Figure 7 This is a schematic diagram of a two-layer optimization device architecture based on a mobile charging station leasing model provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0044] The following describes the implementation of the present invention using specific embodiments. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. Obviously, the embodiments described are only a portion of the present invention, not all of it. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0045] Example 1

[0046] See also Figure 1 Embodiment 1 of the present invention provides a two-layer optimization method based on a mobile charging station leasing model, comprising the following steps:

[0047] S1. Based on the mobile charging station leasing model, a two-tier game model is constructed; the two-tier game model includes a TMCO model and a CFO two-stage DRCC model;

[0048] S2. Introducing a Wasserstein distance distributional robustness model based on chance constraints through a distributional robustness optimization strategy; balancing the economy and robustness of the two-level game model through the Wasserstein distance distributional robustness model;

[0049] S3. Solve the TMCO model using a genetic algorithm to obtain a configuration and rental price that maximizes revenue; solve the CFO two-stage DRCC model using a nested column and constraint generation algorithm to obtain a leasing plan and scheduling strategy that maximizes utility.

[0050] In this embodiment, in step S1, a two-layer game model is constructed based on the mobile charging station leasing model; the two-layer game model includes a TMCO model and a CFO two-stage DRCC model;

[0051] Specifically, the framework of the two-layer game model is as follows Figure 2 This is called the "supplier-tenant" model. As the lessee, the CFO typically owns a certain number of fixed charging stations. Due to the rapid growth and tidal nature of electric vehicle (EV) charging demand, the CFO may face a shortage of charging facilities in certain scenarios. As the supplier, the TMCO invests in and owns the TMCS, leasing it to the CFO to provide EV charging services. This allows the TMCO to focus on asset ownership and equipment maintenance, while the CFO can scale up operations without significant initial costs.

[0052] To meet the differentiated needs of CFOs, TMCO offers long-term and short-term leasing packages and adjusts package prices and TMCS configuration based on CFO feedback. Idle TMCS can act as price takers in the electricity spot market, participating in energy arbitrage by connecting to the distribution grid to generate additional profits. Through an embedded two-stage distributed robust chance-constrained optimization (DRCC), the model can adapt to uncertain EV charging demand and improve TMCS utilization. Based on Stackelberg game theory, the above process can be expressed as follows:

[0053]

[0054] Where, set J is the CFO set, where CFOs act as followers and choose the optimal package based on the rental price set by the game leader TMCO; X j is the strategy set of CFOj, including the number of TMCSs selected for long-term and short-term leases and the lease time j is the CFO number; X T The set of policies set for TMCO, including the number of TMCS configurations (W tmc ) and the long-term and short-term rental package prices you set is the utility function of CFOj, which is to maximize the operator's profit while ensuring the charging service quality requirements; is the revenue function of TMCO, which represents the difference between the total revenue from leasing and energy arbitrage and the costs of investment, operation and maintenance, and losses.

[0055] Through the strategies chosen by each participant, the CFO and TMCO aim to maximize their utility and profit, respectively. Therefore, a feasible solution to the game is the Stackelberg equilibrium (SE), where the leader obtains the optimal price through the followers' optimal strategies, and the followers determine their optimal lease portfolio. In this equilibrium state, no participant can achieve a better outcome by unilaterally changing their strategy.

[0056] In this embodiment, the two-tier game model includes the TMCO model and the CFO two-stage DRCC model;

[0057] The optimization objective of the TMCO model is to maximize the net profit; the expression of the TMCO model is:

[0058]

[0059] Where, is the rental income of TMCS; is the revenue function of TMCO, which represents the difference between the total revenue from leasing and energy arbitrage and the costs of investment, operation and maintenance, and losses; The profit of TMCS when conducting energy arbitrage; C IOM Investment and operation and maintenance costs of TMCS; are the TMCS set and the subsets for short-term and long-term rentals respectively, ∪ represents the union operation; t, T are the rental period and planning period respectively; ω, W tmc The number and total number of TMCS respectively; is a Boolean variable, which is 1 if TMCSω is in the short-term lease, long-term lease or grid energy arbitrage state during period t, and 0 otherwise; These are the prices for short-term and long-term rental packages respectively; c dp , are the energy storage battery cost of TMCSω, the energy storage battery cost when replacing, the cost of charging piles and converters, the cost of other accessories such as trucks and containers, the garage cost, and the operation and maintenance cost; η s ,η b ,η b,f ,η mt are the conversion coefficients of TMCS system, energy storage battery and maintenance costs respectively; K b ,K s are the operating life of the energy storage battery and TMCS system respectively; is the discount rate converted to the planning period; r c ,N rc They are the replacement serial number and total number of replacements of the energy storage battery;

[0060] Among them, the first term in formula (2a) is the total revenue of TMCO, and the second term is the discounted value of investment and operation and maintenance costs; the first to third terms in (2c) are the discounted values ​​of the charger, converter, truck and container component costs, battery investment and replacement costs, and maintenance costs, respectively.

[0061] Defining a Collection are all (ω,t) pairs that satisfy the conditions, where represents the Cartesian product of ω and t, that is, all possible (ω, t) combinations.

[0062] The operating constraints of the TMCO model are:

[0063]

[0064] Where th and H are the scheduling time and scheduling period respectively; is the maximum number of TMCSs that can be deployed, determined by the TMCO budget; is the price cap determined by CFOj’s acquisition cost and leasing preference; is the amortized cost of self-purchase converted to the scheduling period; η j is the adjustment coefficient; The electricity price when TMCS returns to the warehouse for recharging; is the node electricity price at which TMCS participates in the day-ahead electricity market for energy arbitrage; n, v are the operating locations of TMCS, i.e., the energy arbitrage nodes where it interacts with the grid; are the charging and discharging powers of TMCSω on node n at time th; is a Boolean variable. If ω is at node n or moving on the path (n, v) at the th moment, it is 1, otherwise it is 0. e ,n e The time when TMCS service ends and the corresponding garage; is the total travel distance during the scheduling period; v a is the average travel speed of TMCS; is the energy consumption per kilometer of TMCS; is the labor cost of TMCSω; c MDC is the marginal aging cost of TMCS life cycle; q th is the calendar aging parameter of the TMCS battery pack; κ is the year number corresponding to the commissioning of the TMCS; is its maximum charge and discharge power; is a Boolean variable, which is 1 if ω is charged or discharged at time th, otherwise it is 0; η ch,ω ,η dch,ω are the charge and discharge efficiencies of TMCS, respectively; is the capacity of TMCSω; is the SOC value of TMCSω at the scheduling time th; SOC max ,SOC min are the maximum and minimum SOC values ​​of TMCS.

[0065] Among them, Equation (3a) is the spatiotemporal operation constraint of TMCS between short-term rental and arbitrage; Equation (3b) is the rental quantity constraint; Equations (3c)-(3d) are the price constraints; Equations (3f)-(3g) are the TMCS transfer constraints; Equations (3j)-(3k) define the charge and discharge constraints related to the arbitrage model; and Equations (3l) and (3m) define the SOC constraints.

[0066] In this example, EV charging demand, a typical exogenous variable, significantly influences the CFO's leasing decisions and scheduling arrangements. Because historical charging load data can provide probabilistic information, this invention employs the DRCC method to make decisions based on the worst-case uncertainty probability distribution, effectively mitigating decision-making risks and ensuring the quality of the CFO's charging services. The introduction of opportunity constraints helps mitigate the impact of extreme factors on the CFO's profitability, thereby enabling the CFO to balance decision-making risk with economic efficiency.

[0067] The optimization goal of the CFO two-stage DRCC model is to maximize utility. The expression of the CFO two-stage DRCC model is:

[0068]

[0069]

[0070] Where, is the utility function of CFOj, which is to maximize the operator's profit while ensuring the charging service quality requirements; are CFOj’s TMCS set, short-term lease set, and long-term lease set respectively; is the charging demand of EVs that choose TMCS charging at node m; Δ is the charging demand fuzzy set; y, d are optimization variables; Revenue from EV charging services; is the TMCS leasing cost; Cost of running TMCS; Electricity costs for charging EVs; are the charging and discharging powers of TMCSω at node n and m at time th, respectively; m and u are the corresponding EV charging service nodes in the road network; is the TMCS travel distance obtained by the shortest path method;

[0071] Wherein, Equation (4a) represents the revenue of TMCSs providing EV charging services minus its leasing and operating costs.

[0072] The constraints of the CFO two-stage DRCC model are:

[0073]

[0074]

[0075] Where, is the total number of leased units of CFOj at time t; The maximum output power when providing EV charging services to TMCSω is determined by the number and rated power of charging piles of TMCS; is the EV charging demand response ratio of CFOj, reflecting the charging service quality preference of CFO; is a Boolean variable. If ω moves on the path (m,u) or (m,n) at time th, then or Otherwise, they are 0.

[0076] Among them, (5a)-(5c) are the leasing time and space operation constraints, (5d)-(5e) are the leasing quantity constraints; (5f)-(5h) are the TMCS charging service and transfer constraints, and (5k)-(5m) are the TMCS power and charging service quality constraints.

[0077] The outer layer max represents the first-stage optimization problem, which is to find the leasing scheme that maximizes the objective function; the inner layer min represents the second-stage optimization problem, which is to find the worst scenario in the charging demand uncertainty set under a given leasing scheme. The inner layer max adjusts the operating variables (mobility and charging service strategies of each TMCS) under the given leasing scheme and the worst scenario to make the leasing scheme meet all constraints and maximize the objective function.

[0078] In this embodiment, in step S2, a Wasserstein distance distribution robustness model based on chance constraints is introduced through a distribution robustness optimization strategy; the economy and robustness of the two-layer game model are balanced by the Wasserstein distance distribution robustness model;

[0079] Although it is not possible to obtain While limited historical data may not provide a precise probability distribution, some reliable probability information can be used to construct fuzzy sets. A probability distribution fuzzy set is a set of distributions that approximate a benchmark distribution within a certain statistical distance, providing a margin for variations in scenario probability distributions. This type of method employs metrics such as the 1-norm and ∞-norm, the χ2 distance, and the Wasserstein distance to construct fuzzy sets, aiming to better balance the model's economy and robustness.

[0080] Assume there is a historical data set Constructing sample distribution using Dirac function As an estimate of the true distribution:

[0081]

[0082] Where N is the number of sample groups; for The step response at

[0083] When N→∞ Converge to That is, when more data is available, and The "distance" between them becomes smaller.

[0084] Establish arrive One way to measure the "distance" of convergence is to use the Wasserstein distance, which is defined as follows:

[0085]

[0086] Where W(·) represents the Wasserstein distance; Π is the joint distribution of random variables ξ1 and ξ2, and their marginal distributions are and Ξ 2 is the support set of the random variable; ||·||1 means finding the norm, generally the first-order norm.

[0087] Therefore, there is Where ε(·) is a monotonic function related to a certain sample, which decreases to 0 as N tends to infinity. Therefore, given a historical data set with N samples, the true distribution Belongs to the following fuzzy sets:

[0088]

[0089] Where, P N The sample distribution The Wasserstein sphere with the center and radius ε(N) is P(Ξ). P(Ξ) represents the space of all possible values ​​of P.

[0090] The radius ε is related to the sample size (N) and the confidence level (1-ρ ε )related:

[0091]

[0092] Where D ε is an auxiliary variable;

[0093]

[0094] Where, is the sample mean; α is the auxiliary variable; ρ ε Indicates confidence.

[0095] Solve equation (10) by binary search method to get auxiliary variable α, and then get D ε , and then substitute into formula (9) to get the radius ε. Then, rewrite the constraint (5m) into the form of distributed robust chance constraint:

[0096]

[0097] Where, P ev is a possible distribution in the fuzzy set; 1-ρ ev Represents the confidence level for the chance constraint.

[0098] Without loss of generality, we further express this expression in a compact form:

[0099]

[0100] Where E(·) represents the expected calculation; a and b are coefficient vectors.

[0101] However, such probabilistic nonlinear constraints are difficult to solve directly. and s l To derive the approximate value of the conditional value at risk (CVaR) under the above probability constraints:

[0102]

[0103] The above process shows that the CVaR approximation transforms the chance constraints into a set of tractable linear constraints.

[0104] In this embodiment, in step S3, the TMCO model is solved by a genetic algorithm to obtain a configuration and rental price that maximizes revenue; and the CFO two-stage DRCC model is solved by a nested column and constraint generation algorithm to obtain a leasing plan and scheduling strategy that maximizes utility.

[0105] Specifically, the two-layer game model belongs to a two-layer optimization problem, and the objective function and constraints are all linear. Model includes continuous variables and integer variables, belongs to mixed integer linear programming problem, and nested two-stage distributed robust optimization problem in the model, is difficult to solve by a unified optimization method, and processing this type of optimization problem is usually solved using multiple algorithms in combination. Therefore, the present invention adopts a hybrid method combined with a genetic algorithm (GA) and a nested column and constraint generation algorithm (NC&CG) to solve the model.

[0106] The upper-level problem is to determine the configuration quantity and rental price of TMCS by TMCO, and the GA algorithm has good robustness in searching for the global approximate optimal solution.

[0107] The lower-level problem is solved using NC&CG, which includes both inner and outer C&CG loops. Similar to the traditional Benders decomposition method, the proposed method decomposes the original problem into a main problem and subproblems, which are solved through alternating iterations. Because the iterative process continuously adds variables and constraints related to the subproblems to the main problem, C&CG can obtain a more compact lower bound on the original objective function value, effectively reducing the number of iterations. For ease of explanation, the CFO model from the previous section is rewritten in the following compact form:

[0108]

[0109] Where, F T , G T are all constant coefficient matrices; y, d1 are 0-1 decision variables optimized in the first and second stages of the original problem respectively; d2 is a continuous decision variable optimized in the second stage; β is the uncertain parameter of EV charging demand; F, G, L1~L7, J1~J2, S1~S3, V1~V3 are corresponding coefficient matrices; Q1~Q7 are constant column vectors;

[0110] Among them, the first constraint of formula (14) represents formulas (5a)-(5c) of the original problem; the second constraint of formula (14) represents formulas (5d)-(5e) of the original problem; the third constraint of formula (14) represents formulas (5f), (5i)-(5j) of the original problem; the fourth constraint of formula (14) represents formulas (5g)-(5h), (5n) of the original problem; the fifth constraint of formula (14) represents formulas (5k), (5l) of the original problem; the sixth constraint of formula (14) represents formula (5m) of the original problem; the seventh constraint of formula (14) represents formula (5o) of the original problem.

[0111] Decomposing Equation (14), we get the following main problem MP and sub-problem SP:

[0112] The expression of the main problem MP is:

[0113]

[0114] Where * represents a known quantity; s1 and s2 are the history and current iteration times of the outer loop respectively; χ1 is an auxiliary variable, which represents the optimal value of the objective function in the second stage;

[0115] The expression of the subproblem SP is:

[0116]

[0117] Where τ1, τ2, τ3 are the dual variables corresponding to each constraint.

[0118] The outer C&CG algorithm is used to iteratively solve MP and determine the upper bound U in combination with the scene variable β out Substitute the first-stage decision variable y into the solution of SP and obtain the lower bound L of the original problem out The feedback from the SP is used to update the MP, iteratively introducing new constraints and variables until the convergence condition described in (17) is met and the optimal solution is obtained.

[0119] |U out -L out | / L out ≤ψ (17)

[0120] Where ψ is a very small positive real number, representing the convergence gap.

[0121] SP is a two-level MILP with Boolean variables, including a binary variable d1 that does not satisfy the KKT condition. To this end, SP is decomposed into a main problem subset MPS and a subproblem subset SPS.

[0122] The expression of the main problem subset MPS is:

[0123]

[0124] Where r1 and r2 are the history and current iteration times of the inner loop respectively; χ2 is an auxiliary variable, which represents the optimal value of the inner objective function; M is a large positive real number; are the 0-1 variables introduced in the linearization process of KKT complementary relaxation conditions;

[0125] The expression of the subproblem subset SPS is:

[0126]

[0127] Where, β * is the realized value of uncertainty.

[0128] MPS reformulates the inner-level maximum problem into a single-level minimum problem and generates L in , and calculate U using the β obtained in MPS in SPS in . Use d1 and the new constraints to iteratively update until the inner loop converges, as shown in Equation (20), and set L in and L out Return to MP as final scene feedback.

[0129] |U in -L in | / L in ≤ψ (20)

[0130] The solution process of the two-layer game model is as follows Figure 3 As shown, when the operator utility function between two iterations (i.e. ) is less than the convergence gap, the iteration ends and the maximization scheduling strategy is obtained.

[0131] In a possible embodiment, a specific optimization example is provided as follows:

[0132] In this example, a ring highway network such as Figure 4 As shown in Figure 1, the road network has a total of 5 entrances and exits with a total mileage of 465km, of which 1, 2, and 4 are entrances and exits for large cities, and 3 and 5 are entrances and exits for small cities. This topology reflects the current application of TMCS in practice, that is, mobile charging stations are mainly used as temporary supplements to FCS along highways. The planning scope is set as H for one day and T for one year. The sample set is based on the traffic statistics of the road network in the Pearl River Delta region of China and the holiday schedule for 2023, and a planning year is divided into three typical days: weekdays, weekends, and holidays. Assume that the 18 fixed charging stations along the road network belong to four independent CFO operations, and their locations are as follows: Figure 4 Grid arbitrage profits are calculated based on time-of-use electricity prices in a certain province in China. The genetic algorithm population size is set to 30, the maximum number of iterations is 50, and the mutation rate and crossover rate are 0.2 and 0.6, respectively. Other parameters are shown in Table 1.

[0133]

[0134]

[0135] Table 1 Other simulation parameters

[0136] To verify the effectiveness of this invention, we consider the following three scenarios: Case 1, where the CFO adopts a self-purchase model; Case 2, where a uniform pricing scheme is adopted within the proposed TMCO leasing framework; and Case 3, where the proposed method employs a differentiated pricing scheme within the proposed TMCO leasing framework. Furthermore, during idle periods when there is no charging demand, the TMCS owner (the CFO in Case 1 and the TMCO in Cases 2 and 3) increases revenue by participating in grid energy arbitrage.

[0137] Figure 5 are the operating conditions and economic indicators of each operator, where the capacity of a single TMCS is 2MW and the rental price is the converted value according to the dispatch cycle. are the costs and profits of each operator during the planning period, ROI is the operator's return on investment, and CFOs represents the total value of each CFO. Figure 5 (a) As can be seen, CFO2 and CFO4 all choose short-term leases, while CFO1 has 67% of its lease needs for short-term leases, and only CFO3 chooses long-term leases. This also reflects that most CFOs' demand for TMCS is mainly to meet short-term EV charging needs, which will lead to low utilization within the planning period. Under the current electricity market policy, the profit obtained through grid energy arbitrage is limited. Therefore, Case 1 is less economical. Figure 5 The ROIs in (b) are all negative, indicating that the operators are not profitable.

[0138] In Case 2 and Case 3, CFO's costs were significantly reduced, and TMCO optimized the number of TMCS configurations by balancing differentiated short-term rental demands, thereby improving its profitability. Figure 6 The charging demand and TMCS operation status of some stations, including Figure 6 (b) Figure 6 Positive power in (d) indicates TMCS charging, while negative power indicates discharging. This shows that CFO2 requires short-term rental of two TMCSs on Typical Day 1, but only one TMCS is needed to meet charging needs on Typical Day 2. CFO4, on the other hand, requires zero and one TMCS rentals in these two scenarios, respectively. Therefore, TMCO only needs to deploy two TMCSs to meet these demands. However, because the rental price in Case 2 is constrained by the lower rental demand (CFO3 for long-term rentals and CFO4 for short-term rentals), TMCO's profit margin is low. In Case 3, TMCO negotiates with each CFO to determine its rental price. Due to the relatively high rental demand from CFO1 and CFO2, TMCO enhances its profitability by raising its rental price.

[0139] This shows that the traditional self-purchase operation model is more suitable for long-term, continuous charging loads with high EV charging needs (such as CFO3), while the leasing business model proposed in this invention is applicable to both long-term and short-term charging loads. Short-term charging needs are more beneficial to CFOs, ensuring both charging service quality and profitability. For TMCOs, their profits primarily come from long-term leases, and they can formulate different pricing strategies for CFOs to ensure their overall profit margin. However, it may be necessary to balance the interests of various CFOs to prevent some from exiting the market due to excessive pricing. It is also important to note that when the short-term leasing needs of different CFOs are complementary (i.e., occurring at different times), the TMCO can effectively maximize the utilization of the TMCS. This reduces the required fleet size and related investment costs, similar to time-sharing scheduling. By responding to different short-term leasing needs, the proposed business model improves the utilization of the TMCS, achieving a win-win situation for both CFOs and TMCOs.

[0140] In summary, the present invention has the following advantages: the present invention constructs a two-layer game model based on the mobile charging station leasing model; the two-layer game model includes a TMCO model and a CFO two-stage DRCC model; a Wasserstein distance distributional robustness model based on opportunity constraints is introduced through a distributional robustness optimization strategy; the economy and robustness of the two-layer game model are balanced through the Wasserstein distance distributional robustness model; the TMCO model is solved by a genetic algorithm to obtain a configuration and leasing price that maximizes the profit; the CFO two-stage DRCC model is solved by a nested column and constraint generation algorithm to obtain a leasing plan and scheduling strategy that maximizes the utility. This paper proposes a two-layer optimization framework based on the TMCS leasing model. The upper layer aims to maximize the revenue of the TMCS operator (TMCO). The TMCO sets long-term and short-term rental packages based on differences in leasing demand, balances the leasing prices and TMCS configuration quantity of different participants, and dispatches them to participate in grid energy arbitrage during the TMCS's idle periods. The lower layer aims to maximize the utility of the charging facility operator (CFO), responding to the leasing packages and returning the lease quantity and lease time. This paper adopts the distributionally robust optimization (DRO) method and introduces the Wasserstein distance distributionally robust (DRCC) model based on chance constraints to fully account for the impact of charging prediction information errors on the optimization results. Based on the conditional value at risk (CVaR) approximation method, this paper uses a genetic algorithm and a nested column constraint generation (NC&CG) algorithm to solve the upper and lower layers of the model, respectively. Through adaptive pricing and resource allocation strategies, this paper can effectively balance the needs of the TMCO and the CFO, ensuring the profitability of the TMCO while improving the economic efficiency of the CFO.

[0141] It should be noted that the method of the embodiments of the present disclosure can be performed by a single device, such as a computer or server. The method of the embodiments of the present disclosure can also be applied in a distributed scenario, where multiple devices cooperate to perform the method. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiments of the present disclosure, and the multiple devices will interact with each other to complete the method.

[0142] It should be noted that the above description is limited to some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0143] Example 2

[0144] See also Figure 7 , Embodiment 2 of the present invention further provides a two-layer optimization device based on a mobile charging station leasing model, comprising:

[0145] A two-tier game model construction module 001 is used to construct a two-tier game model based on the mobile charging station leasing model; the two-tier game model includes a TMCO model and a CFO two-stage DRCC model;

[0146] Wasserstein distance distribution robustness model processing module 002, used to introduce the Wasserstein distance distribution robustness model based on chance constraints through a distribution robustness optimization strategy; the Wasserstein distance distribution robustness model is used to balance the economy and robustness of the two-layer game model;

[0147] The two-layer game model solving module 003 is used to solve the TMCO model through a genetic algorithm to obtain the configuration and rental price that maximizes the benefits; and solve the CFO two-stage DRCC model through a nested column and constraint generation algorithm to obtain the leasing plan and scheduling strategy that maximizes the utility.

[0148] In this embodiment, in the two-layer game model construction module 001, the optimization goal of the TMCO model is to maximize the net profit; the expression of the TMCO model is:

[0149]

[0150]

[0151] Where, is the rental income of TMCS; is the revenue function of TMCO, which represents the difference between the total revenue from leasing and energy arbitrage and the costs of investment, operation and maintenance, and losses; The profit of TMCS when conducting energy arbitrage; C IOM Investment and operation and maintenance costs of TMCS; are the TMCS set and the subsets for short-term and long-term rentals respectively, ∪ represents the union operation; t, T are the rental period and planning period respectively; ω, W tmc The number and total number of TMCS respectively; is a Boolean variable, which is 1 if TMCSω is in the short-term lease, long-term lease or grid energy arbitrage state during period t, and 0 otherwise; These are the prices for short-term and long-term rental packages respectively; c dp , are the energy storage battery cost of TMCSω, the energy storage battery cost when replacing, the cost of charging piles and converters, the cost of other accessories such as trucks and containers, the garage cost, and the operation and maintenance cost; η s ,η b ,η b,f ,η mt are the conversion coefficients of TMCS system, energy storage battery and maintenance costs respectively; K b ,K s are the operating life of the energy storage battery and TMCS system respectively; is the discount rate converted to the planning period; r c ,N rc They are the replacement serial number and total number of replacements of the energy storage battery;

[0152] The operating constraints of the TMCO model are:

[0153]

[0154]

[0155] Where th and H are the scheduling time and scheduling period respectively; is the maximum number of TMCSs that can be deployed, determined by the TMCO budget; is the price cap determined by CFOj’s acquisition cost and leasing preference; is the amortized cost of self-purchase converted to the scheduling period; η j is the adjustment coefficient; The electricity price when TMCS returns to the warehouse for recharging; is the node electricity price at which TMCS participates in the day-ahead electricity market for energy arbitrage; n, v are the operating locations of TMCS, i.e., the energy arbitrage nodes where it interacts with the grid; are the charging and discharging powers of TMCSω on node n at time th; is a Boolean variable. If ω is at node n or moving on the path (n, v) at the th moment, it is 1, otherwise it is 0. e ,n e The time when TMCS service ends and the corresponding garage; is the total travel distance during the scheduling period; v a is the average travel speed of TMCS; is the energy consumption per kilometer of TMCS; is the labor cost of TMCSω; c MDC is the marginal aging cost of TMCS life cycle; q th is the calendar aging parameter of the TMCS battery pack; κ is the year number corresponding to the commissioning of the TMCS; is its maximum charge and discharge power; is a Boolean variable, which is 1 if ω is charged or discharged at time th, otherwise it is 0; η ch,ω ,η dch,ω are the charge and discharge efficiencies of TMCS, respectively; is the capacity of TMCSω; is the SOC value of TMCSω at the scheduling time th; SOC max ,SOC min are the maximum and minimum SOC values ​​of TMCS.

[0156] In this embodiment, in the two-layer game model construction module 001, the optimization goal of the CFO two-stage DRCC model is to maximize utility; the expression of the CFO two-stage DRCC model is:

[0157]

[0158]

[0159] Where, is the utility function of CFOj, which is to maximize the operator's profit while ensuring the charging service quality requirements; are CFOj’s TMCS set, short-term lease set, and long-term lease set respectively; is the charging demand of EVs that choose TMCS charging at node m; Δ is the charging demand fuzzy set; y, d are optimization variables; Revenue from EV charging services; is the TMCS leasing cost; Cost of running TMCS; Electricity costs for charging EVs; are the charging and discharging powers of TMCSω at node n and m at time th, respectively; m and u are the corresponding EV charging service nodes in the road network; is the TMCS travel distance obtained by the shortest path method;

[0160] The constraints of the CFO two-stage DRCC model are:

[0161]

[0162]

[0163] Where, is the total number of leased units of CFOj at time t; The maximum output power when providing EV charging services to TMCSω is determined by the number and rated power of charging piles of TMCS; is the EV charging demand response ratio of CFOj, reflecting the charging service quality preference of CFO; is a Boolean variable. If ω moves on the path (m,u) or (m,n) at time th, then or Otherwise, they are 0.

[0164] In this embodiment, in the two-level game model solving module 003, in the process of solving the CFO two-stage DRCC model by the nested column and constraint generation algorithm, the CFO two-stage DRCC model solution problem is converted into a main problem MP and a sub-problem SP; the expression of the main problem MP is:

[0165]

[0166] Where, F T , G T are all constant coefficient matrices; y and d1 are 0-1 decision variables optimized in the first and second stages of the original problem respectively; d2 is a continuous decision variable optimized in the second stage; β is the uncertain parameter of EV charging demand; F, G, L1~L7, J1~J2, S1~S3, V1~V3 are corresponding coefficient matrices; Q1~Q7 are constant column vectors; * indicates a known quantity; s1 and s2 are the history and current iteration numbers of the outer loop respectively; χ1 is an auxiliary variable, representing the optimal value of the objective function in the second stage;

[0167] The expression of the subproblem SP is:

[0168]

[0169] Where τ1, τ2, τ3 are the dual variables corresponding to each constraint.

[0170] In this embodiment, in the two-layer game model solving module 003, in the process of solving the subproblem SP, the subproblem SP is decomposed into a main problem subset MPS and a subproblem subset SPS; the expression of the main problem subset MPS is:

[0171]

[0172] Where r1 and r2 are the history and current iteration times of the inner loop respectively; χ2 is an auxiliary variable, which represents the optimal value of the inner objective function; M is a large positive real number; are the 0-1 variables introduced in the linearization process of KKT complementary relaxation conditions;

[0173] The expression of the subproblem subset SPS is:

[0174]

[0175] Where, β * is the realized value of uncertainty.

[0176] It should be noted that the information interaction, execution process, etc. between the modules of the above-mentioned system are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown above in the present application, and no further details will be given here.

[0177] Example 3

[0178] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which a program code for a two-layer optimization method based on a mobile charging station rental model is stored. The program code includes instructions for executing embodiment 1 or any possible implementation thereof, a two-layer optimization method based on a mobile charging station rental model.

[0179] Computer-readable storage media can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0180] Example 4

[0181] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0182] The processor and the memory communicate with each other through a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute a two-layer optimization method based on a mobile charging station rental model according to Example 1 or any possible implementation thereof.

[0183] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor, which is implemented by reading software code stored in a memory. The memory can be integrated into the processor or located outside the processor and exist independently.

[0184] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode.

[0185] Obviously, those skilled in the art will appreciate that the various modules or steps of the present invention described above can be implemented using a general-purpose computing system. They can be centralized on a single computing system or distributed across a network of multiple computing systems. Alternatively, they can be implemented using program code executable by a computing system, and thus, they can be stored in a storage system and executed by the computing system. In some cases, the steps shown or described herein can be performed in a different order than that shown, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0186] Although the present invention has been described in detail above using general descriptions and specific embodiments, it will be apparent to those skilled in the art that modifications and improvements may be made thereto. Therefore, such modifications and improvements, without departing from the spirit of the present invention, are intended to be within the scope of protection claimed herein.

Claims

1. A two-layer optimization method based on the mobile charging station leasing model, characterized in that: include: Based on the mobile charging station leasing model, a two-tier game model is constructed; the two-tier game model includes a mobile charging station operator TMCO model and a charging facility operator CFO two-stage distributed robust opportunity constrained optimization DRCC model; the optimization objective of the TMCO model is to maximize net profit; the optimization objective of the CFO two-stage distributed robust opportunity constrained optimization DRCC model is to maximize utility; A Wasserstein distance distributional robustness model based on chance constraints is introduced through a distributional robustness optimization strategy; the economy and robustness of the two-level game model are balanced through the Wasserstein distance distributional robustness model; The TMCO model is solved by genetic algorithm to obtain the profit-maximizing configuration and leasing price; the CFO two-stage DRCC model is solved by nested column and constraint generation algorithm to obtain the utility-maximizing leasing plan and scheduling strategy.

2. A two-layer optimization method based on the mobile charging station leasing model according to claim 1, characterized in that: The parameters of the TMCO model include: Rental revenue of the mobile charging station TMCS; the profit function of TMCO, which represents the difference between the total rental and energy arbitrage revenue and the investment, operation and maintenance, and loss costs; the profit of TMCS when performing energy arbitrage; the investment and operation and maintenance costs of TMCS; the set of TMCSs, and the subsets for short-term and long-term rentals respectively; the rental period and planning period; the number of TMCSs ω and the total number of units; the energy storage battery cost of TMCSω, the energy storage battery cost when replaced, the cost of charging piles and inverters, the cost of other truck and container accessories, the garage cost, and the operation and maintenance costs; the conversion factor of the TMCS system, energy storage battery, and maintenance costs; the operating life of the energy storage battery and TMCS system; the discount rate converted to the planning period; the replacement sequence number and total number of energy storage battery replacements; The operational constraint parameters of the TMCO model include: The dispatch time and dispatch period; the maximum number of TMCSs to be deployed, as determined by the TMCO's budget; the price ceiling determined by the self-purchase cost and leasing preference of CFO j, where j is the CFO number; the amortized self-purchase cost converted to the dispatch period; the adjustment coefficient; the per-kilowatt-hour electricity price when the TMCS returns to the warehouse for refueling; the node electricity price at which the TMCS participates in the day-ahead electricity market for energy arbitrage; the operating location of the TMCS, that is, the energy arbitrage node where it interacts with the grid; the charge and discharge power of TMCSω at node n at time th; the time when the TMCS ends its service and the corresponding garage; the total distance traveled during the dispatch period; the average driving speed of the TMCS; the energy consumption per kilometer of the TMCS; the labor cost of TMCSω; the marginal aging cost of the TMCS life cycle; the calendar aging parameters of the TMCS battery pack; the serial number corresponding to the year from the commissioning of the TMCS; the maximum charge and discharge power; the charge and discharge efficiency of the TMCS; the capacity of TMCSω; the state of charge (SOC) value of the TMCS at the dispatch time; the maximum and minimum SOC values ​​of the TMCS.

3. A two-layer optimization method based on the mobile charging station leasing model according to claim 2, characterized in that: The parameters of the CFO two-stage distributed robust opportunity constrained optimization DRCC model include: CFO j's utility function, which maximizes the operator's profit while ensuring the required charging service quality; CFO j's set of TMCSs, including short-term and long-term rentals; EV charging demand for TMCS charging at node m; charging demand fuzzy set; optimization variables; EV charging service revenue; TMCS rental cost; TMCS operating cost; EV charging electricity cost; charging and discharging power of TMCS ω at nodes n and m at time th; corresponding EV charging service nodes in the road network; TMCS travel distance obtained by the shortest path method; The constraint parameters of the CFO two-stage DRCC model include: The total number of leased units of CFO j at time t; the maximum output power of TMCSω when providing EV charging service, which is determined by the number and rated power of charging piles of TMCS; and the EV charging demand response ratio of CFO j.

4. A two-layer optimization method based on the mobile charging station leasing model according to claim 3, characterized in that: In the process of solving the CFO two-stage DRCC model by using the nested column and constraint generation algorithm, the CFO two-stage DRCC model solution problem is converted into a main problem MP and a subproblem SP.

5. A two-layer optimization method based on the mobile charging station leasing model according to claim 4, characterized in that: In the process of solving the sub-problem SP, the sub-problem SP is decomposed into a main problem subset MPS and a sub-problem subset SPS.

6. A two-layer optimization device based on a mobile charging station leasing model, adopting a two-layer optimization method based on a mobile charging station leasing model according to any one of claims 1 to 5, characterized in that: include: A two-tier game model construction module is used to construct a two-tier game model based on the mobile charging station leasing model; the two-tier game model includes a mobile charging station operator TMCO model and a charging facility operator CFO two-stage distributed robust opportunity constrained optimization DRCC model; the optimization objective of the TMCO model is to maximize net profit; the optimization objective of the CFO two-stage distributed robust opportunity constrained optimization DRCC model is to maximize utility; A Wasserstein distance distribution robustness model processing module is used to introduce a Wasserstein distance distribution robustness model based on chance constraints through a distribution robustness optimization strategy; the Wasserstein distance distribution robustness model is used to balance the economy and robustness of the two-layer game model; A two-layer game model solving module is used to solve the TMCO model using a genetic algorithm to obtain a profit-maximizing configuration and rental price; and to solve the CFO two-stage DRCC model using a nested column and constraint generation algorithm to obtain a utility-maximizing rental solution and scheduling strategy.

7. A double-layer optimization device based on the mobile charging station leasing model according to claim 6, characterized in that: In the two-layer game model construction module, the optimization goal of the TMCO model is to maximize the net profit; The parameters of the TMCO model include: Rental revenue of the mobile charging station TMCS; the profit function of TMCO, which represents the difference between the total rental and energy arbitrage revenue and the investment, operation and maintenance, and loss costs; the profit of TMCS when performing energy arbitrage; the investment and operation and maintenance costs of TMCS; the set of TMCSs, and the subsets for short-term and long-term rentals respectively; the rental period and planning period; the number of TMCSs ω and the total number of units; the energy storage battery cost of TMCSω, the energy storage battery cost when replaced, the cost of charging piles and inverters, the cost of other truck and container accessories, the garage cost, and the operation and maintenance costs; the conversion factor of the TMCS system, energy storage battery, and maintenance costs; the operating life of the energy storage battery and TMCS system; the discount rate converted to the planning period; the replacement sequence number and total number of energy storage battery replacements; The operational constraint parameters of the TMCO model include: The dispatch time and dispatch period; the maximum number of TMCSs to be deployed, as determined by the TMCO's budget; the price ceiling determined by the self-purchase cost and leasing preference of CFO j, where j is the CFO number; the amortized self-purchase cost converted to the dispatch period; the adjustment coefficient; the per-kilowatt-hour electricity price when the TMCS returns to the warehouse for refueling; the node electricity price at which the TMCS participates in the day-ahead electricity market for energy arbitrage; the operating location of the TMCS, that is, the energy arbitrage node where it interacts with the grid; the charge and discharge power of TMCSω at node n at time th; the time when the TMCS ends its service and the corresponding garage; the total distance traveled during the dispatch period; the average driving speed of the TMCS; the energy consumption per kilometer of the TMCS; the labor cost of TMCSω; the marginal aging cost of the TMCS life cycle; the calendar aging parameters of the TMCS battery pack; the serial number corresponding to the year from the commissioning of the TMCS; the maximum charge and discharge power; the charge and discharge efficiency of the TMCS; the capacity of TMCSω; the state of charge (SOC) value of the TMCS at the dispatch time; the maximum and minimum SOC values ​​of the TMCS.

8. A double-layer optimization device based on the mobile charging station leasing model according to claim 7, characterized in that: In the two-layer game model construction module, the parameters of the CFO two-stage distributed robust opportunity constraint optimization DRCC model include: CFO j's utility function, which maximizes the operator's profit while ensuring the required charging service quality; CFO j's set of TMCSs, including short-term and long-term rentals; EV charging demand for TMCS charging at node m; charging demand fuzzy set; optimization variables; EV charging service revenue; TMCS rental cost; TMCS operating cost; EV charging electricity cost; charging and discharging power of TMCS ω at nodes n and m at time th; corresponding EV charging service nodes in the road network; TMCS travel distance obtained by the shortest path method; The constraint parameters of the CFO two-stage DRCC model include: The total number of leased units of CFO j at time t; the maximum output power of TMCSω when providing EV charging service is determined by the number and rated power of charging piles of TMCS; the EV charging demand response ratio of CFO j reflects the CFO's preference for charging service quality.

9. A double-layer optimization device based on the mobile charging station leasing model according to claim 8, characterized in that: In the two-layer game model solving module, in the process of solving the CFO two-stage DRCC model by using the nested column and constraint generation algorithm, the CFO two-stage DRCC model solving problem is converted into a main problem MP and a subproblem SP.

10. A double-layer optimization device based on the mobile charging station leasing model according to claim 9, characterized in that: In the two-layer game model solving module, in the process of solving the sub-problem SP, the sub-problem SP is decomposed into a main problem subset MPS and a sub-problem subset SPS.

Images