Cost and risk multi-objective based method and system for configuring capacity of optical storage charging station

Abstract

The application belongs to the technical field of photovoltaic storage charging station planning, and discloses a photovoltaic storage charging station capacity configuration method and system based on cost and risk multi-objective, a photovoltaic storage charging station multi-objective capacity optimization configuration model containing cost and risk is established, in the model, a risk function is an investment maintenance and operation cost risk quantified by a conditional risk value, and a cost function is the sum of expected investment maintenance and operation costs under typical scenes considering charging demand and photovoltaic output; in combination with a augmented epsilon-constraint method, the cost is taken as a main target, and a risk secondary target is taken as a constraint; a model is solved to obtain a pareto frontier of cost and risk under different risk preferences and corresponding configuration capacities, and an entropy weight-TOPSIS method is used to screen out an objective decision scheme. Compared with a linear weighting method in traditional risk management, the augmented epsilon-constraint method in the application can obtain a frontier with better distribution and boundary optimality after processing the target, and provides an investment scheme with more detailed cost and risk division.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic-storage-charging station planning technology, and particularly relates to a method and system for configuring the capacity of photovoltaic-storage-charging stations based on multiple objectives of cost and risk. Background Technology

[0002] Currently, driven by dual carbon targets, the electric vehicle (EV) and charging station industries are developing rapidly. Photovoltaic and energy storage charging stations, which combine photovoltaic and energy storage, have received widespread attention due to their advantages such as local photovoltaic absorption and direct reduction of carbon emissions.

[0003] When planning and constructing photovoltaic (PV) and energy storage (ESS) charging stations, investors focus on the costs during commissioning. A reasonable combination of PV and ESS capacity can not only reduce equipment investment and operating costs but also lower operating costs by increasing PV grid integration and reducing the peak-to-valley difference in the power purchase curve. However, the uncertainty of PV output and EV charging demand affects the capacity configuration of PV-ESS charging stations, further impacting the commissioning costs for investors. Therefore, opportunity-constrained planning transforms uncertain planning into deterministic planning, reducing investment costs and shortening the payback period, thus achieving economic efficiency in the investment and operation of PV-ESS charging stations. However, in most planned operations, investors still face high cost risks due to uncertainty, i.e., the risk of costs exceeding expectations. In the planning of PV-ESS charging stations, this risk stems from the uncertainty of PV output and EV charging demand. Ignoring this risk will affect investors' estimates of commissioning costs; therefore, it is necessary to quantify the risk and measure its impact on costs.

[0004] The main methods for managing risk include variance analysis, value at risk (VaR), and conditional value at risk (CVaR). Among these, CVaR possesses favorable mathematical properties and, under the conditions of subadditivity and consistency axioms, can accurately quantify the tail risk of uncertainties under a given probability distribution. It has been widely applied in power system planning and optimal dispatching. For example, in integrated energy system planning and virtual power plant capacity configuration, CVaR is used to measure the operating cost risk caused by uncertainties in renewable energy output and load variations, as well as electricity price fluctuations; in the optimal dispatching of spinning reserve in wind power systems, CVaR is used to measure the risk to the safe operation of the system posed by uncertainties.

[0005] In research involving risk management, optimization of a composite objective function is typically employed. A traditional approach to handling the risk term involves linearly weighting it into the objective function using a weighting factor, providing solution sets and solutions under different subjectively set weighting factors. However, linear weighting methods for multi-objective problems do not guarantee an optimally distributed Pareto solution set. The augmented ε-constraint method, on the other hand, ensures a better distribution and boundary optimality in the Pareto solution set obtained when solving multi-objective optimization problems. It can map the actual Pareto front of the multi-objective problem, providing investors with investment solutions that allow for easier control of cost and risk gradients. Therefore, a novel method for configuring the capacity of photovoltaic-storage-charging stations is urgently needed.

[0006] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0007] (1) In the prior art, the uncertainty of photovoltaic power output and EV charging demand will affect the capacity configuration of photovoltaic and energy storage charging stations, and further affect the cost of the commissioning process of photovoltaic and energy storage charging station investors.

[0008] (2) In most planned operations, investors will also face high cost risks due to uncertainty, namely the risk of costs exceeding expectations. Ignoring this risk will affect investors’ estimates of the costs of the commissioning process.

[0009] (3) In the traditional methods for dealing with risk terms, the linear weighting method cannot guarantee the obtaining of the optimal distribution of the Pareto solution set when dealing with multi-objective problems. Summary of the Invention

[0010] To address the problems existing in the prior art, this invention provides a method and system for configuring the capacity of photovoltaic-storage-charging stations based on multiple objectives of cost and risk, and particularly relates to a method, system, medium, equipment, and terminal for optimizing the capacity configuration of photovoltaic-storage-charging stations using a cost and risk augmented ε-constraint method.

[0011] This invention is implemented as follows: a capacity configuration method for photovoltaic-storage-charging stations based on multiple objectives of cost and risk. The method includes: using CVaR to quantify the risks arising from the uncertainty of photovoltaic power output and electric vehicle charging demand, converting the risks into economic indicators, i.e., risk values; combining the augmented ε-constraint method with cost as the primary objective and risk as a secondary objective, establishing a multi-objective planning and operation model for risk and cost; solving the model to obtain the Pareto frontier of cost and risk under different risk preferences and the corresponding configuration capacity, providing investors with subjective decision-making basis while using the entropy weight-TOPSIS method to screen out objective decision-making schemes.

[0012] Furthermore, the capacity configuration method for photovoltaic-storage-charging stations based on multiple objectives of cost and risk includes the following steps:

[0013] Step 1: Use Monte Carlo sampling to obtain electric vehicle charging demand scenarios and use typical photovoltaic output in four seasons to obtain photovoltaic power generation scenarios; combine scenario-based methods to construct uncertainty functions for electric vehicle charging demand and photovoltaic output.

[0014] Step 2: Establish the investment and maintenance cost function, operating cost function, and CVaR risk measurement function for the photovoltaic and energy storage system. The investment and maintenance cost of the photovoltaic and energy storage system includes the initial investment cost at equal annual values ​​and the annual operating and maintenance cost. The cost of the entire planning and operation phase is the sum of the investment and maintenance cost and the operating cost of the photovoltaic and energy storage system. The risk value represents the high cost risk caused by uncertainty, and this value is quantified using CVaR theory.

[0015] Step 3: Based on the objectives of minimizing costs and risks, and combined with the augmented ε-constraint method to handle multiple objectives, establish a multi-objective capacity optimization configuration model for photovoltaic-storage-charging stations that addresses costs and risks.

[0016] Step four: Use the Gurobi solver to solve the transformed mixed-integer linear programming model. After the solution is completed, the capacity planning results of the photovoltaic-storage-charging station and the corresponding optimized operation strategy are obtained under different risk values.

[0017] Furthermore, in step one, the process of transforming the uncertainty of electric vehicle charging demand and photovoltaic output into the required research scenario is as follows:

[0018] The charging demand of electric vehicles is determined by the initial charging time and the initial state of charge (SOC). The initial SOC of electric vehicles approximately follows a log-normal distribution, while the initial charging time approximately follows a normal distribution.

[0019]

[0020]

[0021] In the formula, S OC1 Initial SOC for charging electric vehicles; t1 is the initial charging time; and The mean and standard deviation of the logarithm of the initial SOC variable for charging electric vehicles; and The mean and standard deviation of the logarithms of the initial SOC variable for electric vehicle charging are calculated. The initial charging time and initial SOC state of the electric vehicle are sampled using the Monte Carlo method to obtain the charging demand of the electric vehicle. The number of charging scenarios required for the electric vehicle is then obtained using k-means clustering.

[0022] P PV (t)=per (t)P PV ;

[0023] In the formula, P PV (t), P PV p er (t) represents the photovoltaic output power at time t, the percentage of photovoltaic power output at time t, and the photovoltaic configuration capacity, respectively. A scenario-based approach is used to address the uncertainties in photovoltaic power output and electric vehicle charging demand. By simulating various possible scenarios through numerous scenarios, the stochastic programming problem is transformed into a deterministic programming problem. The photovoltaic power output scenario set is defined as s = {s...} i i = 1, 2, ..., n s}, Electric vehicle charging demand scenario set e={e j i = 1, 2, ..., n e}; where n s and e j These represent the total number of scenarios for photovoltaic power output and electric vehicle charging demand, respectively; the superscript 'se' indicates that the value is at the 's'th position. i One photovoltaic power output scenario, the eth j In various scenarios involving electric vehicle charging needs.

[0024] Furthermore, in step two, the process of describing the cost function and risk measurement function when configuring the capacity of the photovoltaic-storage system in the photovoltaic-storage charging station is as follows:

[0025] The investment and maintenance costs of a photovoltaic and energy storage system include both the initial investment cost (equivalent annual value) and the annual operation and maintenance costs.

[0026] C inv =(C PV P PV +C ESS,W W ESS +C ESS,P P ESS C RF ;

[0027]

[0028]

[0029] C cost =C inv +C OM ;

[0030]

[0031] In the formula, Cinv, CPV, and C ESS,W C ESS,P C RFRepresent the equivalent annual investment cost of the photovoltaic and energy storage system, the unit capacity investment cost of the photovoltaic and energy storage systems, the unit power investment cost of the energy storage system, and the equivalent annual investment factor, respectively; r and m represent the discount rate and the corresponding system's service life, respectively; C OM , C cost Represents the annual maintenance cost of a photovoltaic and energy storage system, the annual maintenance cost per unit capacity of a photovoltaic and energy storage system, and the investment and maintenance cost of a photovoltaic and energy storage system, respectively; Rse, P s se , Let be the annual operating cost of scenario se, the electricity purchased and sold from the grid by the photovoltaic-storage charging station, and the charging power of the electric vehicle, respectively; a, b, and c be the electricity purchase and sale price from the grid by the photovoltaic-storage charging station and the charging price of the electric vehicle, respectively; T be the number of operating hours, taken as 24 hours; π(s) i ), π(e j ) are respectively the sth i One photovoltaic power output scenario, the eth j The probability of a specific electric vehicle charging demand scenario;

[0032]

[0033] In the formula, α is the confidence level, and the maximum potential loss risk at confidence level α is C. VaR C CVaR To exceed C VaR The average loss of a portion represents cost risk; se CVaR is introduced as a dummy variable to quantify the cost risk value, representing the risk associated with expected investment, maintenance, and operating costs.

[0034] Furthermore, in step three, the process of constructing an augmented ε-constraint method to handle cost and risk multi-objective models and solve the multi-objective planning problem of photovoltaic-storage charging stations is as follows:

[0035] The objective function that minimizes cost is:

[0036]

[0037] The objective function for minimizing risk alone is:

[0038] min{F2(x)=C CVaR};

[0039] The objective function of the CVaR programming problem introduced by the traditional linear weighting method is:

[0040] min{(1-β)F1(x)+βF2(x)};

[0041] Different planning schemes are obtained by artificially adjusting risk weights through weighted factors. β is a weighted factor ranging from [0,1], used to balance cost and risk, representing a risk preference coefficient. Different investment schemes are obtained by changing the parameter β, constructing an efficient frontier of cost and risk; a larger β indicates a greater emphasis on risk, suggesting a risk-averse investor; a smaller β indicates a greater disregard for risk, suggesting a risk-seeking investor.

[0042] The augmented ε-constraint method handles multiple objectives related to cost and risk as follows:

[0043] The augmented ε-constraint method optimizes another primary objective by using a secondary objective as a constraint. It solves the problem by adjusting the value of the auxiliary variable ε within a certain range and calculating the range of values ​​for each objective.

[0044] F 11 =min{F1(x):x∈S};

[0045] F 22 =min{F2(x):x∈S};

[0046] F 12 =min{F2(x):F1(x)=F 11 , x∈S};

[0047] F 21 =min{F1(x):F2(x)=F 22 , x∈S};

[0048] In the formula, F 11 and F 22 Both are minimum values ​​under a single objective, and are minimum values ​​when only F1(x) or F2(x) are considered; F 12 To minimize risk while minimizing cost and single objective, F 21 To minimize costs while minimizing risk and single objective.

[0049] Choosing cost as the primary objective and risk as a secondary objective and constraint, the scope is divided into p equal intervals. By combining auxiliary variable ε and slack variable s, the multi-objective optimization problem is transformed into a single-objective optimization problem.

[0050] ε=lb+(k+r) / p, k=0,1,...,p;

[0051]

[0052] stF2(x)+s=ε,s∈R + ;

[0053] In the formula, lb is the minimum value of the risk objective, p is the number of intervals into which the risk objective is divided, r is the range of the risk objective, α is a sufficiently small number, and s is the non-negative slack variable corresponding to the risk objective.

[0054] Energy storage battery capacity-power function:

[0055] 0.2W ESS ≤P ESS ≤W ESS ;

[0056] In the formula, W ESS P ESS This indicates the capacity and rated power of the energy storage battery.

[0057] Energy storage system capacity and charging / discharging power constraints:

[0058] The relationship between energy storage capacity and charging / discharging power is as follows:

[0059]

[0060] The range of stored energy capacity at time t is:

[0061]

[0062] The initial and final charge levels are equal throughout the operating cycle:

[0063]

[0064] In the formula, η、 D max D min These represent the energy storage system's charge at times t and t-1, its charge / discharge efficiency, charge / discharge power, and maximum depth of charge / discharge, respectively.

[0065]

[0066]

[0067] In the formula, PESS is the rated power configured for the energy storage system, and ut is a variable from 0 to 1. When the value is 1, it can only be charged, and when the value is 0, it can only be discharged.

[0068] Decoupling using the big-M method:

[0069]

[0070]

[0071]

[0072]

[0073] In the formula, M is a sufficiently large positive number to achieve decoupling of nonlinear constraints.

[0074] Power balance constraints:

[0075]

[0076] Power exchange constraints with the grid:

[0077]

[0078] 0≤P s se (t)≤(1-u e )P max ;

[0079] In the formula, P max This represents the maximum power exchanged between the photovoltaic-storage charging station and the power grid. ue is a variable from 0 to 1. When the value is 1, the station can only purchase electricity from the power grid, and when the value is 0, the station can only sell electricity to the power grid.

[0080] CVaR risk constraints:

[0081] z se ≥0;

[0082]

[0083] The CVaR risk constraint metric considers the expected costs for each scenario, describing the high cost risk associated with the investment, maintenance, and operation costs of photovoltaic storage capacity configuration when facing deterministic scenarios composed of uncertainties. After steps one through three, the planning model is transformed into a mixed-integer linear programming model, which is then solved using the Gurobi solver.

[0084] Furthermore, in step four, in the traditional linear weighted method, the confidence level α = 0.9, and the risk preference coefficient increases by 0.05; while in the augmented ε-constraint method, the interval p = 20, and the number of solution sets is 21.

[0085] Another object of the present invention is to provide a photovoltaic-storage-charging station capacity configuration system based on cost and risk multi-objectives, which applies the aforementioned cost and risk multi-objective photovoltaic-storage-charging station capacity configuration method. The cost and risk multi-objective photovoltaic-storage-charging station capacity configuration system includes:

[0086] The optimized configuration model building module is used to establish a multi-objective capacity optimization configuration model for photovoltaic-storage charging stations that includes costs and risks. The risk function is the investment, maintenance and operation cost risk quantified by conditional risk value, and the cost function is the sum of expected investment, maintenance and operation costs considering charging demand and typical photovoltaic output scenarios.

[0087] The augmented ε-constraint module is used to combine the augmented ε-constraint method with cost as the primary objective and risk as a secondary objective as the constraint.

[0088] The objective decision-making scheme selection module is used to solve the model to obtain the Pareto frontier of cost and risk and the corresponding configuration capacity under different risk preferences, and uses the entropy weight-TOPSIS method to select objective decision-making schemes.

[0089] Another object of the present invention is to provide a computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the steps of the method for configuring the capacity of a photovoltaic-storage-charging station based on multiple objectives of cost and risk.

[0090] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method for configuring the capacity of a photovoltaic-storage-charging station based on multiple objectives of cost and risk.

[0091] Another objective of the present invention is to provide an information data processing terminal for implementing the aforementioned photovoltaic energy storage charging station capacity configuration system based on multiple objectives of cost and risk.

[0092] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0093] The uncertainties in electric vehicle charging demand and photovoltaic output introduce uncertainty into the investment, maintenance, and operating costs of photovoltaic-storage charging stations. To quantify the risks posed by this uncertainty—specifically, the risk of costs exceeding expectations—this invention establishes a multi-objective capacity optimization configuration model for photovoltaic-storage charging stations, incorporating both cost and risk considerations. In this model, the risk function represents the investment, maintenance, and operating cost risk quantified by conditional risk value, while the cost function represents the sum of expected investment, maintenance, and operating costs under typical scenarios considering charging demand and photovoltaic output. The model combines an augmented ε-constraint method, prioritizing cost and using risk as a secondary constraint. Solving the model yields the Pareto fronts of cost and risk under different risk preferences, along with the corresponding configuration capacities. An entropy-weighted TOPSIS method is then used to select objective decision-making schemes. Compared to the linear weighting method in traditional risk management, the augmented ε-constraint method of this invention, after processing the objectives, yields a more distributed and boundary-optimal front, capable of mapping the actual Pareto front of the multi-objective problem and providing investment schemes with more detailed cost and risk classifications.

[0094] This invention provides a capacity configuration method for photovoltaic-storage-charging stations based on multiple objectives of cost and risk. This method quantifies risks for investors and presents the relationship between costs and risks. It employs an augmented ε-constraint method to handle the multi-objective problem, improving the uniformity of the solution set and providing richer planning schemes. This invention uses the augmented ε-constraint method to handle multi-objective optimization problems, providing investors with investment schemes with different cost and risk combinations, while simultaneously using the entropy weight-TOPSIS method to screen objective decision-making schemes for investors. To quantify cost and risk, this invention establishes a multi-objective capacity configuration model for assessing cost and risk using CVaR, and proposes an augmented ε-constraint method to handle multi-objective models containing costs and risks. This optimizes the capacity configuration of photovoltaic-storage-charging stations, quantifies the risk of investors facing costs exceeding expectations, and provides investors with objective planning schemes and operational strategies.

[0095] This invention quantifies the cost risks to photovoltaic (PV) power output and EV charging demand that bring to PV-storage charging stations, including investment, maintenance, and operating costs. It visually demonstrates the relationship between risk and cost, showing that higher costs correspond to lower risks, and provides investment and operational solutions for investors with different risk appetites. The invention quantifies the cost risks of uncertainty to PV-storage charging stations, establishes a multi-objective capacity optimization configuration model for cost and risk, and constructs an augmented ε-constraint method to handle multiple objectives. Compared to the traditional linear weighted method for handling risk terms, the Pareto efficient frontier obtained by the augmented ε-constraint method is more uniform and has better boundary points. It also more finely divides investment and operational schemes for different risk appetites, facilitating investors' control over risk and cost. Among different investment and operational schemes, investors can subjectively choose a decision scheme based on their risk appetite; or use the entropy weight-TOPSIS method to comprehensively evaluate each scheme, select an objective decision scheme, and obtain the optimal trade-off between cost and risk.

[0096] The expected benefits and commercial value of the technical solution of this invention after transformation are as follows: considering the impact of the uncertainty of photovoltaic output and electric vehicle charging demand on the capacity configuration and operation of photovoltaic-storage system, the cost risk value caused by this impact is quantified, and investment and operation schemes with different risks and costs are presented to investors of photovoltaic-storage charging stations, so as to facilitate different types of investors to choose investment schemes.

[0097] The technical solution of this invention fills a technological gap in the domestic and international industry: it uses CVaR to quantify risk and establish a multi-objective model of cost and risk. Compared with the traditional CVaR risk planning and operation model, the solution set obtained by combining the augmented ε-constraint method to handle multi-objective problems has a more selective gradient, that is, the control of the obtained planning and operation scheme is more detailed; and the solution set boundary is optimal, that is, the cost of the planning and operation scheme is smaller when only the single objective of risk is considered, and the risk of the planning and operation scheme is smaller when only the single objective of cost is considered.

[0098] Does the technical solution of this invention solve a long-standing technical problem that people have long desired to solve but have yet to succeed in? First, it quantifies the specific risks brought about by uncertainty for investors, converting risk into economic indicators. Second, it provides different investment operation plans for investors of different investment types, enabling investors to manage their own combinations of costs and risks. Third, it optimizes the model with risk objectives, solving the problems of poor solution set distribution and excessively high boundary point risk in traditional CVaR risk planning operation models. Specifically, the obtained planning operation plan remains the same under different risk preferences, or the cost is too high when only the single objective of risk is considered, or the risk is too high when only the single objective of cost is considered. Attached Figure Description

[0099] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0100] Figure 1 This is a flowchart of the photovoltaic-storage-charging station capacity configuration method based on multiple objectives of cost and risk provided in the embodiments of the present invention;

[0101] Figure 2 This is a schematic diagram of the photovoltaic-storage-charging station capacity configuration method based on multiple objectives of cost and risk provided in an embodiment of the present invention;

[0102] Figure 3 This is a schematic diagram of typical photovoltaic power output percentages throughout the four seasons provided in an embodiment of the present invention;

[0103] Figure 4 This is a flowchart of the augmented ε-constraint method for handling multiple objectives related to cost and risk, provided in an embodiment of the present invention.

[0104] Figure 5 This is a comparison diagram of the effective frontiers of the linear weighted method and the augmented ε-constraint method provided in the embodiments of the present invention;

[0105] Figure 6 This is a schematic diagram of the photovoltaic capacity and energy storage capacity power configuration results provided in the embodiments of the present invention;

[0106] Figure 7 This is a schematic diagram of the total EV charging power after clustering, provided in an embodiment of the present invention;

[0107] Figure 8 This is a schematic diagram of EV charging electricity price and electricity purchase and sale price provided in an embodiment of the present invention;

[0108] Figure 9 This is a schematic diagram of the comprehensive evaluation value of the Pareto solution set provided in an embodiment of the present invention;

[0109] Figure 10 This is a schematic diagram of photovoltaic output and EV charging demand in the 13th scenario provided in this embodiment of the invention;

[0110] Figure 11 This is a diagram illustrating the operation of the 13th scenario in the objective decision-making scheme provided in this embodiment of the invention.

[0111] Figure 12 This is a schematic diagram illustrating the operation of the 13th scenario in the cost-only investment scheme provided in this embodiment of the invention;

[0112] Figure 13 This is a schematic diagram illustrating the operation of the 13th scenario in the investment scheme that only considers risk provided in the embodiments of the present invention. Detailed Implementation

[0113] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0114] To address the problems existing in the prior art, this invention provides a method and system for configuring the capacity of photovoltaic-storage-charging stations based on multiple objectives of cost and risk. The invention will now be described in detail with reference to the accompanying drawings.

[0115] To enable those skilled in the art to fully understand how the present invention is specifically implemented, this section provides an explanatory description of the embodiments that expand upon the technical solutions of the claims.

[0116] like Figure 1 As shown, the photovoltaic-storage-charging station capacity configuration method based on multiple objectives of cost and risk provided by this embodiment of the invention includes the following steps:

[0117] S101, Establish a multi-objective capacity optimization configuration model for photovoltaic-storage-charging stations that includes costs and risks;

[0118] S102, combining the augmented ε-constraint method with cost as the primary objective and risk as a secondary objective as the constraint;

[0119] S103, solve the model to obtain the Pareto frontier of cost and risk and the corresponding configuration capacity under different risk preferences, and use the entropy weight-TOPSIS method to screen out objective decision schemes.

[0120] In the multi-objective capacity optimization configuration model for photovoltaic-storage charging stations provided in this embodiment of the invention, which includes cost and risk, the risk function is the investment, maintenance and operation cost risk quantified by conditional risk value, and the cost function is the sum of expected investment, maintenance and operation costs considering charging demand and typical photovoltaic output scenarios.

[0121] As a preferred embodiment, such as Figure 2 As shown, the photovoltaic-storage-charging station capacity configuration method based on multiple objectives of cost and risk provided in this embodiment of the invention specifically includes the following steps:

[0122] Step S1: Use Monte Carlo sampling to obtain electric vehicle charging demand scenarios and use typical photovoltaic output in four seasons to obtain photovoltaic power generation scenarios; combine scenario methods to construct uncertainty functions for electric vehicle charging demand and photovoltaic output.

[0123] Step S2 involves establishing the investment and maintenance cost function, operating cost function, and CVaR risk measurement function for the photovoltaic and energy storage system. The investment and maintenance cost includes the initial investment cost at an equal annual value and the annual operating and maintenance cost. The total cost during the entire planning and operation phase is the sum of the investment and maintenance cost and the operating cost of the photovoltaic and energy storage system. The risk value represents the high cost risk caused by uncertainty, and this value is quantified using CVaR theory.

[0124] Step S3: Based on the objectives of minimizing cost and risk, and combined with the augmented ε-constraint method to handle multiple objectives, establish a multi-objective capacity optimization configuration model for photovoltaic-storage-charging stations that addresses cost and risk.

[0125] Step S4: Use the Gurobi solver to solve the transformed mixed integer linear programming model. After the solution is completed, the capacity planning results of the photovoltaic-storage-charging station and the corresponding optimized operation strategy are obtained under different risk values.

[0126] Step S1 is used to transform an uncertain problem into a deterministic problem.

[0127] Step S2 aims to quantify the cost risks that the uncertainty of electric vehicle charging demand and photovoltaic output poses to the planning and operation of photovoltaic-storage charging stations.

[0128] Step S3 serves to establish an augmented ε-constraint method for handling multi-objective planning and operation models of cost and risk, providing investors with planning and operation schemes that have more detailed risk control gradients.

[0129] Step S4 is to use a solver to solve the transformed mixed-integer linear programming model and obtain the planning operation scheme.

[0130] In step S1 provided by this embodiment of the invention, the process of transforming the uncertainty of electric vehicle charging demand and photovoltaic output into the required research scenario is as follows:

[0131] The charging demand of electric vehicles is determined by the initial charging time and the initial state of charge (SOC). The initial SOC of electric vehicles approximately follows a log-normal distribution, while the initial charging time approximately follows a normal distribution.

[0132]

[0133]

[0134] In the formula, S OC1 Initial SOC for charging electric vehicles; t1 is the initial charging time; and The mean and standard deviation of the logarithm of the initial SOC variable for charging electric vehicles; and The mean and standard deviation of the logarithms of the initial SOC variable for electric vehicle charging are calculated. The initial charging time and initial SOC state of the electric vehicle are sampled using the Monte Carlo method to obtain the charging demand. The number of charging scenarios required for the electric vehicle is then obtained using k-means clustering.

[0135] P PV (t)=p er (t)P PV (3)

[0136] In the formula, P PV (t), P PV p er (t) represents the photovoltaic output power at time t, the percentage of photovoltaic power output at time t, and the photovoltaic configuration capacity, respectively. The typical percentage of photovoltaic power output across all four seasons is shown below. Figure 3 As shown. A scenario-based approach is used to address the uncertainties in photovoltaic power output and electric vehicle charging demand. Through numerous scenario simulations of various possible situations, the stochastic programming problem is ultimately transformed into a deterministic programming problem. Let the photovoltaic power output scenario set be s = {s...} i i = 1, 2, ..., n s}, Electric vehicle charging demand scenario set e={e j i = 1, 2, ..., n e}, where n s and e j These represent the total number of scenarios for photovoltaic power output and electric vehicle charging demand, respectively. In the following analysis, "se" in all superscript letters indicates that the value is at the s-th position. i One photovoltaic power output scenario, the eth j In various scenarios involving electric vehicle charging needs.

[0137] In step S2 of this embodiment of the invention, the process of describing the cost function and risk measurement function when configuring the capacity of the photovoltaic-storage system in a photovoltaic-storage charging station is as follows:

[0138] The investment and maintenance costs of a photovoltaic and energy storage system include two aspects: the initial investment cost at equivalent annual value and the annual operation and maintenance costs.

[0139] C inv=(C PV P PV +C ESS,W W ESS +C ESS,P P ESS C RF (4)

[0140]

[0141]

[0142] C cost =C inv +C OM (7)

[0143]

[0144] In the formula, C inv C PV C ESS,W C ESS,P C RF represents the equivalent annual investment cost of the photovoltaic and energy storage system, the unit capacity investment cost of the photovoltaic and energy storage systems, the unit power investment cost of the energy storage system, and the equivalent annual investment factor, respectively; r and m represent the discount rate and the corresponding system's service life, respectively. C OM , C cost These represent the annual maintenance cost of the photovoltaic and energy storage system, the annual maintenance cost per unit capacity of the photovoltaic and energy storage systems, and the investment and maintenance cost of the photovoltaic and energy storage system, respectively. se , P s se , π(si), π(e) represent the annual operating cost of scenario se, the electricity purchased and sold from the grid by the photovoltaic-storage charging station, and the charging power of the electric vehicle, respectively; a, b, and c represent the electricity purchase and sale price from the grid by the photovoltaic-storage charging station and the charging price of the electric vehicle, respectively; T represents the number of operating hours (24 hours in this invention); π(si) and π(e) represent the annual operating cost of scenario se, the electricity purchased and sold from the grid by the photovoltaic-storage charging station, ... power of the electric vehicle, respectively; a, b, and c represent the annual operating cost of scenario se, the electricity purchased and sold from the grid by the photovoltaic-storage charging station, and the charging price of the electric vehicle, respectively j ) represent the si-th photovoltaic power output scenario and the e-th photovoltaic power output scenario, respectively. j The probability of electric vehicle charging demand scenarios.

[0145]

[0146] In the formula, α is the confidence level, and the maximum potential loss risk at confidence level α is C. VaR And C CVaR To exceed C VaR The average loss of a portion, i.e., the cost risk in this invention, z seAs a dummy variable. Due to the uncertainty of photovoltaic output and EV charging demand, while minimizing the expected cost, there may be a maximum cost, that is, the risk of high cost. In order to overcome this ambiguity, CVaR is introduced to quantify the cost risk value. Equation (9) represents the risk related to expected investment, maintenance and operation costs. The specific risk items are shown in the CVaR risk constraint in step S3.

[0147] In step S3 of this embodiment of the invention, the process of constructing an augmented ε-constraint method to handle the multi-objective model of cost and risk and solving the multi-objective planning problem of photovoltaic-storage charging stations is as follows:

[0148]

[0149] min{F2(x)=C CVaR} (11)

[0150] Equations (10) and (11) represent the objective functions for minimizing cost and minimizing risk individually, respectively. We will first introduce the objective function of the CVaR programming problem using the traditional linear weighted method:

[0151] min{(1-β)F1(x)+βF2(x)} (12)

[0152] Equation (12) is the objective function form of the CVaR planning problem with risk terms. Different planning schemes are obtained by artificially adjusting the risk weights through weighting factors. Among them, β is a weighting factor in the range of [0,1], used to achieve the trade-off between cost and risk, i.e., the risk preference coefficient. Different investment schemes can be obtained by changing the parameter β, and an efficient frontier of cost and risk can be constructed. The larger β is, the more risk is valued. Such investors are risk-averse, i.e., they want to minimize risk as much as possible; the smaller β is, the more risk is ignored. Such investors are risk-seeking, i.e., they want to minimize cost as much as possible. However, the Pareto frontier distribution and boundary optimality obtained by the multi-objective model constructed by the traditional linear weighting method are poor. That is, when investors make a choice based on the cost and risk values ​​of each investment scheme, it is not easy to adjust the gradient of the investment scheme, and the risk is too high when only cost is considered, and the cost is too high when only risk is considered.

[0153] The augmented ε-constraint method proposed in this invention handles multiple objectives related to cost and risk as follows:

[0154] The augmented ε-constraint method uses a secondary objective as a constraint to optimize another primary objective, adjusting the value of the auxiliary variable ε within a certain range to solve the problem. The flowchart is as follows: Figure 4 As shown. First, calculate the value range for each target.

[0155] F 11=min{F1(x):x∈S} (13)

[0156] F 22 =min{F2(x):x∈S} (14)

[0157] F 12 =min{F2(x):F1(x)=F 11 ,x∈S} (15)

[0158] F 21 =min{F1(x):F2(x)=F 22 ,x∈S} (16)

[0159] In the formula, F 11 and F 22 All are minimum values ​​under a single objective, that is, minimum values ​​when only F1(x) or F2(x) are considered. 12 To minimize risk while minimizing cost and single objective, F 21 To minimize cost while minimizing risk, we choose cost as the primary objective and risk as a secondary objective and constraint, dividing its range into p equal intervals. By combining auxiliary variables ε and slack variables s, we transform the multi-objective optimization problem into a single-objective optimization problem.

[0160] ε=lb+(k+r) / p,k=0,1,...,p (17)

[0161]

[0162] stF2(x)+s=ε,s∈R + (19)

[0163] In the formula, lb is the minimum value of the risk objective, p is the number of intervals into which the risk objective is divided, r is the range of the risk objective, α is a sufficiently small number, and s is the non-negative slack variable corresponding to the risk objective. The above describes the construction of the cost and risk multi-objective function; the following section lists the relevant constraints and variables.

[0164] Energy storage battery capacity-power function:

[0165] 0.2W ESS ≤P ESS ≤W ESS (20)

[0166] In the formula, W ESS P ESS This formula represents the capacity and rated power of the energy storage battery, and it limits the relationship between energy storage capacity and power.

[0167] Energy storage system capacity and charging / discharging power constraints:

[0168]

[0169]

[0170]

[0171] In the formula, η、 D max D min The values ​​are the energy storage system's charge at time t and t-1, the charge / discharge efficiency, the charge / discharge power, and the maximum charge / discharge depth, respectively. Equation (21) shows the relationship between the energy storage capacity and the charge / discharge power. Equation (22) limits the range of the energy storage capacity at time t. Equation (23) ensures that the energy storage capacity is equal at the beginning and end of an operating cycle (24h in this invention).

[0172]

[0173]

[0174] In the formula, P ESS The rated power configured for the energy storage system, u t The value is 0 to 1. When the value is 1, it can only be charged, and when the value is 0, it can only be discharged. Equations (24) and (25) limit the range of charging and discharging power of the energy storage system and ensure that the energy storage system does not charge and discharge at the same time.

[0175] Because a 0-1 variable u is introduced in equations (24) and (25) t And P ESS These are also decision variables, leading to the emergence of nonlinear constraints. Therefore, the Big-M method is used to decouple equations (24) and (25).

[0176]

[0177]

[0178]

[0179]

[0180] In the formula, M is a sufficiently large positive number that achieves the decoupling of nonlinear constraints. At this time, constraints (24) to (25) are transformed into constraints (26) to (29).

[0181] Power balance constraints:

[0182]

[0183] Power exchange constraints with the grid:

[0184]

[0185]

[0186] In the formula, P max u represents the maximum power exchanged between the photovoltaic-storage charging station and the power grid. e The value is a variable ranging from 0 to 1. When the value is 1, electricity can only be purchased from the grid, and when the value is 0, electricity can only be sold to the grid, thus ensuring that the photovoltaic and energy storage charging stations do not purchase or sell electricity from the grid at the same time.

[0187] CVaR risk constraints:

[0188] z se ≥0 (33)

[0189]

[0190] Equation (34) measures the risk value considering the expected cost for each scenario, describing the high cost risk of investment, maintenance, and operation costs of photovoltaic storage capacity configuration when facing deterministic scenarios composed of uncertainties. After steps S1, S2, and S3, the planning model proposed in this invention is transformed into a mixed-integer linear programming model, and therefore the Gurobi solver can be called to solve the model.

[0191] In step S4 of this embodiment of the invention, in the traditional linear weighted method, the confidence level α = 0.9 and the risk preference coefficient increases by 0.05, while in the augmented ε-constraint method, the interval p = 20, and the number of solution sets is 21. The simulation results obtained after solving are as follows.

[0192] Comparison of solution sets between linear weighted method and augmented ε-constraint method, as follows: Figure 5 As shown.

[0193] from Figure 5 The Pareto efficient front in the solution set shows that as investment, maintenance, and operating costs decrease, the corresponding risk increases. A negative cost indicates that the photovoltaic-storage charging station is profitable during actual operation, while a negative risk indicates the minimum possible return after commissioning. The actual physical meanings of cost, risk, return, and their positive and negative signs are explained below: Figure 5 The cost corresponding to point A is -281,085 yuan, which means that the investor's expected return is 281,085 yuan; the risk is -141,492 yuan, and the return is no less than 141,492 yuan at a 90% confidence level.

[0194] At the efficient frontier, as risk increases and cost decreases, it indicates a focus on cost while ignoring risk, reflecting the investment trend of risk-seeking investors; conversely, as cost increases and risk decreases, it indicates a focus on risk while ignoring cost, reflecting the investment trend of risk-averse investors. Furthermore, the efficient frontier obtained by the linear weighting method shows that a uniformly distributed set of weight coefficients β does not guarantee a uniform distribution of the efficient solution set {F1, F2}. Therefore, the mapping of the Pareto efficient set is insufficient, and different weight combinations will produce the same efficient solution, for example, when β = 0.9, 0.95 or β = 0.05, ..., 0.55. Thus, it can be concluded that the Pareto efficient frontier obtained using the augmented ε-constraint method has better distribution and can provide investment schemes with more easily adjustable cost and risk solution set gradients.

[0195] The efficient solution sets obtained by the two methods are not comparable because the results are two different mappings of the same Pareto boundary. However, they are comparable at boundary points B, β=1 and A, β=0. At these boundaries, scheme B and scheme A dominate the schemes β=1 and β=0, respectively. For the upper left boundary, this is equivalent to the same risk, but scheme B has a lower cost than the scheme with β=1; for the lower right boundary, this is equivalent to the same cost, but scheme A has a lower risk than the scheme with β=0. In summary, the superiority of the augmented ε-constraint method is proven.

[0196] Cost and risk analysis, and analysis of different capacity configurations: Analysis is conducted based on the capacity configuration of the solution set obtained by the augmented ε-constraint method. The results of photovoltaic capacity and energy storage capacity power configurations are as follows: Figure 6 As shown.

[0197] Because the uncertainties in photovoltaic power output and electric vehicle charging demand affect planning results, this invention uses a scenario-based approach to handle these uncertainties. The optimal result that minimizes investment, maintenance, and operating costs is actually the probabilistic sum of costs across all scenarios—a projected cost value. However, the actual cost value corresponds to a cost distribution, which may result in a high probability of significant costs for investors. To overcome this ambiguity, this invention introduces CVaR (Continuous Cost Risk Assessment) to quantify the risk of these costs, allowing investors to make subjective decisions by autonomously weighing the relationship between costs and risks when investing.

[0198] The uncertainty of photovoltaic (PV) output and electric vehicle (EV) charging demand can cause fluctuations in operating costs across various scenarios, creating operating cost risks and further impacting the capacity configuration of PV energy storage, thus leading to investment and maintenance cost risks. As shown in the figure, as investors increasingly prioritize risk management, they tend to reduce the capacity allocation for both PV and energy storage, but the proportion of energy storage within PV capacity will increase. Reducing capacity allocation can mitigate investment and maintenance cost risks to some extent, but it increases operating costs, thereby increasing operating cost risk. Therefore, the change in energy storage capacity is relatively small, ensuring that the system cost reduction risk from reduced investment outweighs the increased operating cost risk, resulting in a downward trend in overall risk. This process will increase electricity purchases from the grid to meet EV charging needs, reduce PV capacity to smooth out fluctuations, and decrease electricity sales to the grid.

[0199] Conversely, when investors prioritize cost and ignore risk, they may invest heavily in photovoltaic and energy storage capacity. Although this increases investment and maintenance costs, ensuring the charging needs of electric vehicles are met reduces electricity purchases from the grid and increases electricity sales to the grid, thus lowering operating costs. Furthermore, the reduction in operating costs outweighs the increase in investment and maintenance costs, resulting in overall cost reduction. However, due to increased investment and the volatility of photovoltaic power, investors also face greater risks due to uncertainty.

[0200] This invention quantifies the cost risks brought to photovoltaic-storage charging stations by uncertainties, establishes a multi-objective capacity optimization configuration model for cost and risk, and constructs an augmented ε-constraint method to handle multi-objectives.

[0201] This invention measures the investment, maintenance, and operating cost risks of photovoltaic (PV) power output and electric vehicle charging demand to PV-storage charging stations, visually demonstrating the relationship between risk and cost—higher costs correlate with lower risk—and providing investment and operational solutions for investors with different risk appetites. The augmented ε-constraint method yields a more uniform Pareto efficient frontier distribution and better boundary points. It also provides a more detailed breakdown of investment and operational solutions for different risk appetites, facilitating investors' control over risk and cost.

[0202] The photovoltaic-storage-charging station capacity configuration system based on multiple objectives of cost and risk provided in this embodiment of the invention includes:

[0203] The optimized configuration model building module is used to establish a multi-objective capacity optimization configuration model for photovoltaic-storage charging stations that includes costs and risks. The risk function is the investment, maintenance and operation cost risk quantified by conditional risk value, and the cost function is the sum of expected investment, maintenance and operation costs considering charging demand and typical photovoltaic output scenarios.

[0204] The augmented ε-constraint module is used to combine the augmented ε-constraint method with cost as the primary objective and risk as a secondary objective as the constraint.

[0205] The objective decision-making scheme selection module is used to solve the model to obtain the Pareto frontier of cost and risk and the corresponding configuration capacity under different risk preferences, and uses the entropy weight-TOPSIS method to select objective decision-making schemes.

[0206] To demonstrate the inventiveness and technical value of the technical solution of this invention, this section provides specific product or related technology application examples of the technical solution claimed.

[0207] The technical solution of this invention has been analyzed through numerical simulation.

[0208] 1. Example Parameters

[0209] The models established in sections 3.1 and 3.2 were solved using the Gurobi solver in Matlab. The probability of typical photovoltaic power output scenarios in all four seasons is 0.25. Monte Carlo sampling was used to obtain the charging demand of EVs, and k-means was used to cluster the EV charging demand into four categories, as follows: Figure 7 As shown, the probabilities of the four charging power categories after clustering are 0.238, 0.262, 0.262, and 0.238, respectively. Considering the time perspective, all photovoltaic-storage charging stations use fast charging piles to provide charging services to EVs, and it is assumed that there will be no queuing for charging, meaning that EVs will be charged immediately upon arrival at the photovoltaic-storage charging station. The EV charging electricity price (a), the electricity purchase and sale price from the grid by the photovoltaic-storage charging station (b), and (c) are as follows. Figure 8 As shown, the maximum power exchanged between the photovoltaic-storage charging station and the grid is 200kW. The initial capacity of the energy storage battery is 50% of its total capacity. Among the three pricing options, the EV charging price is higher than the electricity purchase price and significantly higher than the electricity sales price, ensuring that the photovoltaic-storage charging station provides charging services to EVs as much as possible, rather than selling electricity to the grid.

[0210] 2. Examples, Results, and Analysis

[0211] 2.1 Comparison of solution sets between linear weighted method and augmented ε-constraint method

[0212] To compare the advantages and disadvantages of different methods for handling multi-objective optimization problems involving risk management, their respective effective frontiers are plotted as follows: Figure 5 As shown. In the traditional linear weighted method, the confidence level α = 0.9, and the risk preference coefficient increases by 0.05, while in the augmented ε-constraint method, the interval p = 20, and the number of solution sets is 21.

[0213] from Figure 5 The Pareto efficient front shows that as investment, maintenance, and operating costs decrease in the solution set, the corresponding risk increases. A negative cost indicates that the photovoltaic-storage charging station is profitable during actual operation, while a negative risk indicates the minimum possible return after commissioning. The actual physical meanings of cost, risk, return, and their positive and negative signs are explained below: Figure 5 Point A corresponds to a cost of -281,085 yuan, indicating that the investor's expected return is 281,085 yuan; the risk is -141,492 yuan, meaning that the return will not be less than 141,492 yuan at a 90% confidence level.

[0214] At the efficient frontier, as risk increases and cost decreases, it indicates a focus on cost while ignoring risk, reflecting the investment trend of risk-seeking investors; conversely, as cost increases and risk decreases, it indicates a focus on risk while ignoring cost, reflecting the investment trend of risk-averse investors. Furthermore, the efficient frontier obtained by the linear weighting method shows that a uniformly distributed set of weight coefficients β does not guarantee a uniform distribution of the efficient solution set {F1, F2}. Therefore, the mapping of the Pareto efficient set is insufficient, and different weight combinations will produce the same efficient solution, for example, when β = 0.9, 0.95 or β = 0.05, ..., 0.55. Thus, it can be concluded that the Pareto efficient frontier obtained using the augmented ε-constraint method has better distribution and can provide investment schemes with more easily adjustable cost and risk solution set gradients.

[0215] The efficient solution sets obtained by the two methods are not comparable because the results are two different mappings of the same Pareto boundary. However, they are comparable at boundary points B, β=1 and A, β=0. At these boundaries, scheme B and scheme A dominate the schemes β=1 and β=0, respectively. For the upper left boundary, this is equivalent to the same risk, but scheme B has a lower cost than the scheme with β=1; for the lower right boundary, this is equivalent to the same cost, but scheme A has a lower risk than the scheme with β=0. In summary, the superiority of the augmented ε-constraint method is proven.

[0216] 2.2 Cost and Risk Analysis and Different Capacity Configurations

[0217] This invention analyzes the capacity configuration based on the solution set obtained by the augmented ε-constraint method. The photovoltaic capacity and energy storage capacity power configuration results are as follows: Figure 6 As shown.

[0218] Because the uncertainties in photovoltaic output and EV charging demand affect planning results, this invention uses a scenario-based approach to handle these uncertainties. The optimal result that minimizes investment, maintenance, and operating costs is actually the probabilistic sum of costs across all scenarios—an expected cost value. However, the actual cost value corresponds to a cost distribution, which may result in a high probability of significant costs for investors. To overcome this ambiguity, this invention introduces CVaR (Continuous Cost Risk Assessment) to quantify the risk of these costs, allowing investors to make subjective decisions by autonomously weighing the relationship between costs and risks when investing.

[0219] The uncertainty of photovoltaic power output and EV charging demand can cause fluctuations in operating costs across various scenarios, creating operating cost risks and further affecting the capacity configuration of photovoltaic energy storage, thus leading to investment and maintenance cost risks. Figure 6 It can be observed that as investors increasingly prioritize risk management, they will reduce the capacity allocation for both solar PV and energy storage, but the proportion of energy storage within solar PV capacity will increase. Reducing capacity allocation can mitigate investment and maintenance cost risks to some extent, but it increases operating costs, thereby increasing operating cost risk. Therefore, the change in energy storage capacity is relatively small. This ensures that the system cost risk reduced by lowering investment outweighs the increased operating cost risk, resulting in a downward trend in overall risk. This process will increase electricity purchases from the grid to meet EV charging needs, reduce solar PV capacity to mitigate fluctuations, and decrease electricity sales to the grid.

[0220] Conversely, when investors prioritize cost and ignore risk, they may invest heavily in photovoltaic (PV) and energy storage capacity. Although this increases investment and maintenance costs, it also reduces operating costs by decreasing electricity purchases from the grid and increasing electricity sales after meeting EV charging needs. Furthermore, the reduction in operating costs outweighs the increase in investment and maintenance costs, resulting in overall cost reduction. However, due to increased investment and the volatility of PV power, investors also face greater risks due to uncertainty.

[0221] 2.3 Screening of Objective Decision-Making Schemes Based on Entropy Weight-TOPSIS Method

[0222] First, the entropy weight method is used to calculate... Figure 5 The weights of cost and risk objectives in the solution set of the augmented ε-constraint method are: W1 = 0.5461, W2 = 0.4539. The comprehensive evaluation values ​​of each solution set are obtained according to the TOPSIS method as follows: Figure 9 As shown.

[0223] Figure 6 The solution set index in ascending order of size and Figure 5 The solution sets for medium-risk values ​​correspond one-to-one from right to left. Figure 6 The capacity configurations, with risk values ​​decreasing from left to right, correspond one-to-one. A higher overall evaluation value indicates a better solution set; the highest overall evaluation value is 0.6176. Figure 9 The 7th solution set represents the objective decision-making scheme; (Compare with...) Figure 9 The capacity configuration results for each device in the objective decision-making scheme are: P PV =265kW,W ESS =352kWh, P ESS =145kW, cost is -271,521 yuan, risk is -148,031 yuan.

[0224] 2.4 Comparison of the results of objective decision-making schemes and typical schemes

[0225] In actual investment, based on the investor's subjective wishes, there are two typical approaches that aim to minimize either cost or risk. Figure 9 The first and 21st solution sets represent these two typical schemes, respectively. The investment result of the first solution set is: P PV =315kW,W ESS =354kWh, P ESS =146kW, cost -281085 yuan, risk -141492 yuan, the investment result of the 21st solution set is: P PV =112kW,W ESS =272kWh, P ESS =112kW, cost -234071 yuan, risk -163290 yuan. The cost is the expected cost, including the operating costs of each scenario. The actual cost is related to the scenario probability, and the quantified risk value represents the maximum possible cost. This invention has 16 scenarios. In the simulation results, the cost of the 13th scenario in each scheme exceeds the risk value, meaning the cost of this scenario will exceed the expected cost. The photovoltaic output and EV charging demand in the 13th scenario are as follows: Figure 10 As shown, the operational status of the objective decision-making scheme, the investment scheme considering only cost, and the investment scheme considering only risk in this scenario are as follows: Figure 11 , Figure 12 , Figure 13 As shown.

[0226] Because investment schemes that only consider risk excessively avoid risk, the capacity allocation of photovoltaics and energy storage is far lower than that of objective decision-making schemes and investment schemes that only consider cost. Figure 11 and Figure 12 , Figure 13 The comparison reveals that in an investment plan that only considers risk, the capacity of the configured photovoltaic and energy storage is very low. When photovoltaic power generation is weak (between 1:00 and 7:00), this plan can only meet the charging needs of EVs by purchasing electricity from the grid. When photovoltaic power generation is strong (between 11:00 and 15:00), this plan not only cannot sell electricity to the grid to generate revenue, but can also only store a small amount of energy. Therefore, while this plan's overly conservative investment reduces risk, it significantly increases operating costs.

[0227] Compared to an investment plan that only considers cost, the objective decision-making plan involves a reduction in photovoltaic (PV) capacity while maintaining essentially the same energy storage capacity. The objective decision-making plan reduces the PV capacity allocation, therefore... Figure 11 and Figure 12The comparison shows that the objective decision-making scheme reduces the electricity sold to the grid. Although this increases operating costs, it reduces investment costs by lowering the photovoltaic capacity configuration. Therefore, the investment, maintenance, and operating cost risks in the risk item can be interpreted as follows: the reduction in risk value due to lower investment and maintenance costs outweighs the increase in risk value due to higher operating costs. Thus, this invention reduces overall cost risk.

[0228] In summary, the objective decision-making plan, compared to the investment plan that only considers cost, reduces risk by 4.6% while increasing cost by only 3.4%; the optimal investment plan, compared to the investment plan that only considers risk, reduces cost by 16.0% while increasing risk by only 9.3%.

[0229] The embodiments of the present invention have achieved some positive results during the research and development or use process, and have indeed great advantages compared with the prior art. The following content describes them in conjunction with the data, charts and other information of the experimental process.

[0230] 1. EV charging demand and photovoltaic-storage charging station functions

[0231] 1.1EV Uncertainty Charging Demand Function

[0232] The charging demand of an EV is determined by the initial charging time and the initial SOC. The initial SOC of an EV approximately follows a log-normal distribution, while the initial charging time approximately follows a normal distribution.

[0233]

[0234]

[0235] In the formula, S OC1 Initial SOC for EV charging; t1 is the initial charging time; and Mean and standard deviation of the logarithm of the initial SOC variable for EV charging; and The mean and standard deviation of the logarithm of the initial SOC variable for charging the EV.

[0236] 1.2 Typical Photovoltaic Output Function

[0237] P PV (t)=p er (t)P PV (3)

[0238] In the formula, P PV (t), P PV p er (t) represents the photovoltaic output power at time t, the percentage of photovoltaic power output at time t, and the photovoltaic configuration capacity, respectively. The typical percentage of photovoltaic power output across all four seasons is shown below. Figure 3 As shown.

[0239] 1.3 Energy Storage Battery Capacity-Power Function

[0240] Energy storage batteries can store electrical energy during periods of low electricity prices or when the photovoltaic power generation is greater than the EV charging power; and release electrical energy when the photovoltaic power generation is lower than the EV charging power to meet the EV's charging needs.

[0241] 0.2W ESS ≤P ESS ≤W ESS (4)

[0242] In the formula, W ESS P ESS This formula represents the capacity and rated power of the energy storage battery, and it limits the relationship between energy storage capacity and power.

[0243] 1.4 Investment and Maintenance Cost Function of Photovoltaic Storage System

[0244] The investment and maintenance costs of a photovoltaic and energy storage system include two aspects: the initial investment cost at equivalent annual value and the annual operation and maintenance costs.

[0245] C inv =(C PV P PV +C ESS,W W ESS +C ESS,P P ESS C RF (5)

[0246]

[0247]

[0248] C cost =C inv +C OM (8)

[0249] In the formula, C inv C PV C ESS,W C ESS,P C RF represents the equivalent annual investment cost of the photovoltaic and energy storage system, the unit capacity investment cost of the photovoltaic and energy storage systems, the unit power investment cost of the energy storage system, and the equivalent annual investment factor, respectively; r and m represent the discount rate and the corresponding system's service life, respectively. C OM , C cost These represent the annual maintenance cost of the photovoltaic and energy storage system, the annual maintenance cost per unit capacity of the photovoltaic and energy storage systems, and the investment and maintenance cost of the photovoltaic and energy storage system, respectively.

[0250] 2. Multi-objective capacity optimization configuration model for photovoltaic-storage-charging stations considering CVaR measurement risk

[0251] 2.1 Scenario Description

[0252] The multi-objective capacity optimization configuration problem of photovoltaic-storage-charging stations, which includes uncertainties such as photovoltaic output and EV charging demand, is a stochastic programming problem. This paper employs a scenario-based approach to handle the uncertainties of photovoltaic output and EV charging demand. By simulating various possible scenarios through numerous scenarios, the stochastic programming problem is ultimately transformed into a deterministic programming problem. Let the photovoltaic output scenario set be s = {s...} i i = 1, 2, ..., n s}, EV charging demand scenario set e={e j i = 1, 2, ..., n e}; where n s and e j These represent the total number of scenarios for photovoltaic power output and EV charging demand, respectively. In the following analysis, "se" in all superscript letters indicates that the value is at the s-th position. i One photovoltaic power output scenario, the eth j In individual EV charging demand scenarios.

[0253] 2.2 Objective function for investment, maintenance and operating costs

[0254] Minimize the investment, maintenance, and operating costs of photovoltaic-storage charging stations:

[0255]

[0256]

[0257] In the formula, R se , P s se , π(s) represents the annual operating cost of scenario se, the electricity purchased and sold from the grid by the photovoltaic-storage charging station, and the charging power of the EV, respectively; a, b, and c represent the electricity purchase and sale price from the grid by the photovoltaic-storage charging station and the charging price of the EV, respectively; T represents the number of operating hours (24 hours in this invention); π(s) i ), π(e j ) are respectively the sth i One photovoltaic power output scenario, the eth j The probability of individual EV charging demand scenarios.

[0258] 2.3 Objective Function for CVaR Risk Measurement

[0259] Minimize the risks faced by photovoltaic-storage-charging stations:

[0260]

[0261] min{F2(x)=C CVaR} (12)

[0262] In the formula, α is the confidence level, and the maximum potential loss risk at confidence level α is C. VaR And C CVaR To exceed C VaR The average loss of a portion, i.e., the cost risk in this invention, z se As a dummy variable. Due to the uncertainty of photovoltaic output and EV charging demand, while minimizing the expected cost, there may be a maximum cost, that is, the risk of high cost. In order to overcome this ambiguity, CVaR is introduced to quantify the cost risk value. Equation (11) represents the risk related to expected investment, maintenance and operation costs. The specific risk items are shown in 2.4 CVaR risk constraints.

[0263] 2.4 Constraints

[0264] 2.4.1 Energy storage system power and charging / discharging power constraints

[0265]

[0266]

[0267]

[0268] In the formula, η、 D max D min Equation (13) represents the energy storage system's energy at times t and t-1, its charge / discharge efficiency, charge / discharge power, and maximum charge / discharge depth, respectively. Equation (14) defines the energy storage capacity range at time t, and Equation (15) ensures that the energy storage capacity is equal at the beginning and end of an operating cycle (24h in this invention).

[0269]

[0270]

[0271] In the formula, P ESS The rated power configured for the energy storage system, u t The value is 0 to 1. When the value is 1, it can only be charged, and when the value is 0, it can only be discharged. Equations (16) and (17) limit the range of charging and discharging power of the energy storage system and ensure that the energy storage system does not charge and discharge at the same time.

[0272] Because a 0-1 variable u is introduced in equations (16) and (17) t And PESS These are also decision variables, leading to the emergence of nonlinear constraints. Therefore, the big-M method is used to decouple equations (16) and (17).

[0273]

[0274]

[0275]

[0276]

[0277] In the formula, M is a sufficiently large positive number that achieves the decoupling of nonlinear constraints. At this time, constraints (16) to (17) are transformed into constraints (18) to (21).

[0278] 2.4.2 Power Balance Constraints

[0279]

[0280] 2.4.3 Power Exchange Constraints with the Power Grid

[0281]

[0282] 0≤P s se (t)≤(1-u e )P max (twenty four)

[0283] In the formula, P max u represents the maximum power exchanged between the photovoltaic-storage charging station and the power grid. e The value is a variable ranging from 0 to 1. When the value is 1, electricity can only be purchased from the grid, and when the value is 0, electricity can only be sold to the grid, thus ensuring that the photovoltaic and energy storage charging stations do not purchase or sell electricity from the grid at the same time.

[0284] 2.4.4 CVaR Risk Constraints

[0285] To facilitate the solution, dummy variables have been introduced, and to facilitate the calculation, the dummy variables are relaxed to the following two constraints.

[0286] zse≥0(25)

[0287]

[0288] Equation (26) measures the risk value for each scenario, considering the expected cost. It describes the high cost risk of investment, maintenance and operation costs of photovoltaic storage capacity configuration when facing a deterministic scenario composed of uncertainty.

[0289] 3. Multi-objective combination and solution including risk management

[0290] 3.1 Linear Weighted Method

[0291] The model established in this invention not only considers the economic objective of minimizing the total cost of photovoltaic-storage-charging stations, but also the risk objective of minimizing CVaR quantification risk. When solving multi-objective optimization problems involving risk management, linear weighting methods are often used, and risk preference coefficients are introduced to handle risk terms, transforming the multi-objective optimization problem into a single-objective optimization problem.

[0292] min{(1-β)F1(x)+βF2(x)} (27)

[0293] stx∈S (28)

[0294] The linear weighted method constructs a composite objective function containing cost and risk (27). In the formula: β is a weighting factor in the range [0,1], used to achieve a trade-off between cost and risk, i.e., the risk preference coefficient. By changing the parameter β, different investment schemes can be obtained, and an efficient frontier of cost and risk can be constructed. The larger β is, the more risk is valued. Such investors are risk-averse, i.e., they want to minimize risk as much as possible; the smaller β is, the more risk is ignored. Such investors are risk-seeking, i.e., they want to minimize cost as much as possible. S is the set of feasible regions of decision variables x that satisfy equations (4), (13)-(26).

[0295] Solving the multi-objective model constructed using the linear weighted method yields a Pareto front with poor distribution and boundary optimality. This means that when investors make choices based on the cost and risk values ​​of various investment options, it is difficult to control the gradient of investment options; considering only cost results in excessive risk, and considering only risk results in excessive cost.

[0296] 3.2 Augmented ε-constraint method

[0297] To improve the distribution of the solution set and the values ​​of the boundary points in multi-objective combinatorial problems constructed by the linear weighted method, an augmented ε-constraint method is proposed to construct a multi-objective model with risk management. This method maps the actual Pareto front of the multi-objective optimization problem, provides investment schemes with more easily adjustable cost and risk solution set gradients, and offers a solution set with better boundary points.

[0298] The augmented ε-constraint method uses a secondary objective as a constraint to optimize another primary objective, adjusting the value of the auxiliary variable ε within a certain range to solve the problem. First, the value range of each objective is calculated, namely the value range of the cost and risk objectives in this invention, as shown in Table 1.

[0299] Table 1. Range of Cost and Risk Target Values

[0300]

[0301] F in Table 1 11 and F 22 All are minimum values ​​under a single objective, that is, minimum values ​​when only F1(x) or F2(x) are considered. 12 To minimize risk while minimizing cost and single objective, F 21 To minimize cost while minimizing risk for a single objective, the calculation process is as follows:

[0302] F 11 =min{F1(x):x∈S} (29)

[0303] F 22 =min{F2(x):x∈S} (30)

[0304] F 12 =min{F2(x):F1(x)=F 11 ,x∈S} (31)

[0305] F 21 =min{F1(x):F2(x)=F 22 ,x∈S} (32)

[0306] The maximum and minimum values ​​in each column of Table 1 determine the range of each objective on the Pareto front. At this point, the cost objective is chosen as the primary objective, and the risk objective as the secondary objective and constraint. The range is divided into p equal intervals, and the auxiliary variable ε and slack variable s are used to transform the multi-objective optimization problem into a single-objective optimization problem.

[0307] ε=lb+(k+r) / p,k=0,1,...,p (33)

[0308]

[0309] stF2(x)+s=ε,s∈R + ,x∈S (35)

[0310] In the formula, lb is the minimum value of the risk objective, p is the number of intervals into which the risk objective is divided, r is the range of the risk objective, α is a sufficiently small number, and s is the non-negative slack variable corresponding to the risk objective.

[0311] 3.3 Screening of Objective Optimal Solutions Based on Entropy Weight-TOPSIS Method

[0312] Solving the multi-objective model constructed in 3.1 and 3.2 yields the Pareto front. Investors can subjectively choose a decision-making scheme based on their own trade-offs between costs and risks, or they can objectively determine the decision-making scheme using the entropy-weighted TOPSIS method. Its core idea is to standardize cost and risk indicators, objectively assign weights to cost and risk indicators based on the information entropy of the indicator data, and quantify the relative distance between each solution set and the positive and negative ideal solutions as a comprehensive evaluation value.

[0313] The first step is to use the range method to analyze each indicator X. ij Standardization is performed to eliminate the influence of orders of magnitude and dimensions.

[0314]

[0315] In the formula, i represents the solution number, j represents the measurement index, i.e., the cost or risk index; X ij and Y ij Represent the cost value, risk value, and standardized cost value and risk value, respectively; max(X ij ) and min(X ij Both ) represent the maximum or minimum value of cost or risk.

[0316] The second step is to calculate the Y values ​​for each indicator. ij Information entropy E j .

[0317]

[0318] The third step is to calculate the Y values ​​for each indicator. ij weight W j .

[0319]

[0320] The fourth step is to construct a weighted evaluation matrix S for each indicator.

[0321] S=(s ij ) n×m (39)

[0322] In the formula, n is the number of solutions in the Pareto solution set, i.e., p+1 in this invention; m is the index number, s ij =W j ×Y ij .

[0323] The fifth step is to determine the positive ideal solution k for each index based on the weighted evaluation matrix S. * and negative ideal solution k 0 Since both cost and risk are cost-related indicators, meaning the lower the better, then:

[0324]

[0325] Step 6: Calculate the solution sets and the positive ideal solution. and negative ideal solution Euclidean distance and

[0326]

[0327] Step 7: Calculate the overall evaluation value:

[0328]

[0329] Overall evaluation value R i The largest solution set is optimal.

[0330] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0331] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for configuring the capacity of photovoltaic-storage-charging stations based on multiple objectives of cost and risk, characterized in that, The method for configuring the capacity of photovoltaic-storage-charging stations based on multiple objectives of cost and risk includes: establishing a multi-objective capacity optimization configuration model for photovoltaic-storage-charging stations that includes cost and risk. In the model, the risk function is the investment, maintenance, and operating cost risk quantified by conditional risk value, and the cost function is the sum of expected investment, maintenance, and operating costs considering charging demand and typical photovoltaic output scenarios; combining the augmented ε-constraint method with cost as the primary objective and risk as the secondary objective; solving the model to obtain the Pareto fronts of cost and risk and the corresponding configuration capacity under different risk preferences; and using the entropy weight-TOPSIS method to select objective decision schemes. The method for configuring the capacity of photovoltaic-storage-charging stations based on multiple objectives of cost and risk includes the following steps: Step 1: Use Monte Carlo sampling to obtain electric vehicle charging demand scenarios and use typical photovoltaic output in four seasons to obtain photovoltaic power generation scenarios; combine scenario-based methods to construct uncertainty functions for electric vehicle charging demand and photovoltaic output. Step 2: Establish the investment and maintenance cost function, operating cost function, and CVaR risk measurement function for the photovoltaic and energy storage system. The investment and maintenance cost of the photovoltaic and energy storage system includes the initial investment cost at equal annual values ​​and the annual operating and maintenance cost. The cost of the entire planning and operation phase is the sum of the investment and maintenance cost and the operating cost of the photovoltaic and energy storage system. The risk value represents the high cost risk caused by uncertainty, and this value is quantified using CVaR theory. Step 3: Based on the objectives of minimizing costs and risks, and combined with the augmented ε-constraint method to handle multiple objectives, establish a multi-objective capacity optimization configuration model for photovoltaic-storage-charging stations that addresses costs and risks. Step 4: Use the Gurobi solver to solve the transformed mixed integer linear programming model. After the solution is completed, the capacity planning results of the photovoltaic-storage-charging station and the corresponding optimized operation strategy are obtained under different risk values. In step three, the augmented ε-constraint method is used to construct a multi-objective model for cost and risk, and the process of solving the multi-objective planning problem of photovoltaic-storage charging stations is as follows: The objective function that minimizes cost is: ； C cost This represents the investment and maintenance costs of a photovoltaic and energy storage system; n s n represents the total number of photovoltaic power generation scenarios; e This represents the total number of scenarios indicating electric vehicle charging demand; π(s) i ) indicates the first The probability of a photovoltaic power output scenario; π(e j ) indicates the first The probability of a specific electric vehicle charging demand scenario; R se The annual operating cost of scenario se; se indicates the stage. The photovoltaic power output scenario, the first In various electric vehicle charging demand scenarios; The objective function for minimizing risk alone is: ； For more than The average loss of a portion represents cost risk; The objective function of the CVaR programming problem introduced by the traditional linear weighting method is: ； Different planning schemes are obtained by artificially adjusting risk weights through weighted factors; among them... It is within the range The weighting factor, used to achieve a trade-off between cost and risk, represents the risk preference coefficient; by changing the parameters... Obtain different investment options and build an effective frontier of cost and risk; The larger the value, the more risk-averse the investor is; The smaller the value, the less risk is considered, indicating that the investor is risk-seeking. The augmented ε-constraint method handles multiple objectives related to cost and risk as follows: The augmented ε-constraint method optimizes another primary objective by using a secondary objective as a constraint. It solves the problem by adjusting the value of the auxiliary variable ε within a certain range and calculating the range of values ​​for each objective. ； ； ； ； In the formula, and All are minimum values ​​under a single objective, considering only... or The minimum value at that time; To minimize risk while minimizing cost and single objective, To minimize cost while minimizing risk as the primary objective, the scope is divided into two parts: cost as the primary objective and risk as a secondary objective and constraint. Equal intervals, combined with auxiliary variable ε and slack variable Transform the multi-objective optimization problem into a single-objective optimization problem; ； ； ； In the formula, It is the minimum value of the risk target. It is the number of intervals in which the risk targets are divided. It refers to the scope of risk objectives. It is a small enough number. These are the non-negative slack variables corresponding to the risk objectives; Energy storage battery capacity-power function: ； In the formula, , This indicates the capacity and rated power of the energy storage battery. Energy storage system capacity and charging / discharging power constraints: The relationship between energy storage capacity and charging / discharging power is as follows: ； , These represent the energy levels of the energy storage system at times t and t-1, respectively. For the charging and discharging efficiency of the energy storage system; , These represent the charging power and discharging power of the energy storage system at time t, respectively. The range of stored energy capacity at time t is: ； This refers to the minimum depth of discharge for the energy storage system. This represents the maximum depth of charge for the energy storage system. The initial and final charge levels are equal throughout the operating cycle: ； T represents the number of operating hours; ； ； In the formula, The rated power configured for the energy storage system, It is a variable from 0 to 1. When the value is 1, it can only charge; when the value is 0, it can only discharge. Decoupling using the big-M method: ； ； ； ； In the formula, M is a sufficiently large positive number to achieve decoupling of nonlinear constraints; Power balance constraints: ； Let se be the photovoltaic output power in the se-th scenario; The power purchased from the grid by the photovoltaic-storage charging station; The electricity sold to the grid by the photovoltaic-storage charging station; The charging power for electric vehicles; Power exchange constraints with the grid: ； ； In the formula, This represents the maximum power exchanged between the photovoltaic-storage charging station and the power grid. The value is a variable between 0 and 1. When the value is 1, electricity can only be purchased from the grid; when the value is 0, electricity can only be sold to the grid. CVaR risk constraints: ； ； z se It is a dummy variable; The risk value of the CVaR risk constraint metric is based on the expected cost for each scenario, describing the high cost risk of investment, maintenance and operation costs of photovoltaic storage capacity configuration when facing deterministic scenarios composed of uncertainties; after steps one to three, the planning model is transformed into a mixed integer linear programming model, and the model is solved by calling the Gurobi solver.

2. The photovoltaic-storage-charging station capacity configuration method based on multiple objectives of cost and risk as described in claim 1, characterized in that, In step one, the process of transforming the uncertainty of electric vehicle charging demand and photovoltaic output into the required research scenario is as follows: The charging demand of electric vehicles is determined by the initial charging time and the initial state of charge (SOC). The initial SOC of electric vehicles approximately follows a log-normal distribution, while the initial charging time approximately follows a normal distribution. ； ； In the formula, Initial SOC for charging electric vehicles; Initial charging time; and The mean and standard deviation of the logarithm of the initial SOC variable for charging electric vehicles; and The mean and standard deviation of the logarithm of the initial SOC variable for electric vehicle charging are calculated. The initial charging time and initial SOC of electric vehicles are sampled using the Monte Carlo method to obtain the charging demand of electric vehicles. The number of charging scenarios required for electric vehicles is obtained by combining k-means clustering. ； In the formula, , , These represent the photovoltaic output power at time t, the percentage of photovoltaic output at time t, and the photovoltaic configuration capacity, respectively. The scenario-based approach is used to address the uncertainties in photovoltaic power output and electric vehicle charging demand. A large number of scenarios are simulated to simulate various possible situations and transform the problem into a deterministic programming problem. A set of photovoltaic power output scenarios is defined. Electric vehicle charging demand scenarios ;in, and These represent the total number of scenarios for photovoltaic power output and electric vehicle charging demand, respectively.

3. The photovoltaic-storage-charging station capacity configuration method based on multiple objectives of cost and risk as described in claim 1, characterized in that, In step two, the process of describing the cost function and risk measurement function when configuring the capacity of the photovoltaic storage system is as follows: The investment and maintenance costs of a photovoltaic and energy storage system include both the initial investment cost (equivalent annual value) and the annual operation and maintenance costs. ； ； ； ； ； In the formula, , , , , These represent the equivalent annual investment cost of the photovoltaic and energy storage system, the unit capacity investment cost of the photovoltaic and energy storage systems, the unit power investment cost of the energy storage system, and the equivalent annual investment coefficient, respectively. , These represent the discount rate and the corresponding system's service life, respectively. , , , These represent the annual maintenance cost of the photovoltaic and energy storage system, the annual maintenance cost per unit capacity of the photovoltaic and energy storage systems, and the investment and maintenance cost of the photovoltaic and energy storage system, respectively. , , , Scenes Annual operating cost, power purchased from the grid by the photovoltaic-storage charging station, power sold to the grid by the photovoltaic-storage charging station, and charging power of electric vehicles, superscript All indicate that the value is in the first position. The photovoltaic power output scenario, the first In various electric vehicle charging demand scenarios; , , These are the electricity purchase and sale price from the power grid for photovoltaic and energy storage charging stations, and the electricity price for charging electric vehicles, respectively. The running hours are set to 24 hours. , The first The photovoltaic power output scenario, the first The probability of a specific electric vehicle charging demand scenario; ； In the formula, At the confidence level, at the confidence level The maximum potential loss risk is ; For more than The average loss of a portion represents cost risk; CVaR is introduced as a dummy variable to quantify the cost risk value, representing the risk associated with expected investment, maintenance, and operating costs.

4. A photovoltaic-storage-charging station capacity configuration system based on cost and risk multi-objectives, applying the cost- and risk-based multi-objective capacity configuration method for photovoltaic-storage-charging stations as described in any one of claims 1 to 3, characterized in that, The photovoltaic-storage-charging station capacity configuration system based on multiple cost and risk objectives includes: The optimized configuration model building module is used to establish a multi-objective capacity optimization configuration model for photovoltaic-storage charging stations that includes costs and risks. The risk function is the investment, maintenance and operation cost risk quantified by conditional risk value, and the cost function is the sum of expected investment, maintenance and operation costs considering charging demand and typical photovoltaic output scenarios. The augmented ε-constraint module is used to combine the augmented ε-constraint method with cost as the primary objective and risk as a secondary objective as the constraint. The objective decision-making scheme selection module is used to solve the model to obtain the Pareto frontier of cost and risk and the corresponding configuration capacity under different risk preferences, and uses the entropy weight-TOPSIS method to select objective decision-making schemes.

5. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the capacity configuration method for photovoltaic-storage-charging stations based on multiple cost and risk objectives as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the capacity configuration method for a photovoltaic-storage-charging station based on multiple cost and risk objectives as described in any one of claims 1 to 3.

7. An information data processing terminal, characterized in that, The information data processing terminal includes the photovoltaic energy storage charging station capacity configuration system based on multiple objectives of cost and risk as described in claim 4.

Images