Power distribution network adequacy improvement strategy generation method, system, equipment, medium and product

Abstract

The invention discloses a power distribution network adequacy improvement strategy generation method, system and device, a medium and a product, and relates to the field of power distribution network adequacy improvement strategy generation, and the method comprises the steps: employing a generative adversarial network, depicting the output uncertainty of distributed new energy according to the historical output data of the distributed new energy, and obtaining the output uncertainty of the distributed new energy; depicting the uncertainty of the running boundary of the electric vehicle cluster caused by the uncertainty of the willingness of the user; establishing an adequacy improvement strategy double-layer optimization model considering multi-dimensional uncertainty; the uncertainty of distributed new energy output is considered in an upper layer model, a two-stage robust model is adopted to formulate a standby strategy, node marginal electricity price considering the uncertainty is generated, and a standby management price mechanism is formed; and considering the operation boundary uncertainty of the electric vehicle cluster in the lower layer model, introducing a data-driven distribution robust chance constraint form, generating a power distribution network adequacy improvement strategy that the electric vehicle cluster provides standby for the power distribution network, and ensuring safe power utilization of users in the novel power system.

Description

Technical Field

[0001] This application relates to the field of generating strategies to improve the adequacy of distribution networks, and in particular to a method, system, device, medium, and product for generating such strategies. Background Technology

[0002] With a high proportion of distributed renewable energy being integrated into the distribution network, the randomness and volatility of its output pose new challenges to the power balance of the distribution network. The high proportion of renewable energy has led to an increasingly prominent issue of power supply security in the distribution network. The traditional model relying on source-grid side regulation is unsustainable, and the effective utilization of the flexibility of large-scale electric vehicles on the load side has become a new approach to solving the power supply security problem. However, current demand-side resources are mainly used for peak shaving, and the impact of the spatiotemporal distribution of high-proportion distributed renewable energy output on the adequacy of the distribution network has not yet been taken into account. There is insufficient preventive measures for the direct factors and indirect risks that may cause risks to the adequacy of the distribution network, resulting in an unreasonable design of strategies to improve the adequacy of the distribution network. Summary of the Invention

[0003] The purpose of this application is to provide a method, system, device, medium and product for generating distribution network adequacy improvement strategies, so as to solve the problem of unreasonable design of distribution network adequacy improvement strategies.

[0004] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for generating a distribution network adequacy enhancement strategy, including: Generative adversarial networks are used to characterize the uncertainty of distributed renewable energy output based on historical power output data. Based on price incentives, user response intention samples are generated to characterize the uncertainty of the electric vehicle cluster operation boundary caused by the uncertainty of user intentions; A two-layer optimization model for sufficiency enhancement strategy considering multidimensional uncertainties is established; the two-layer optimization model for sufficiency enhancement strategy includes an upper-layer model and a lower-layer model; In the upper-level model, considering the uncertainty of the output of the distributed new energy, a two-stage robust model is adopted to formulate a backup strategy and generate a node marginal electricity price that takes into account the uncertainty, forming a backup management price mechanism to guide electric vehicle users to provide backup. In the lower-level model, based on the reserve management price mechanism and considering the uncertainty of the electric vehicle cluster's operating boundary, a data-driven sub-Blu-ray opportunity constraint is introduced to generate a distribution network adequacy improvement strategy for the electric vehicle cluster to provide reserves to the distribution network.

[0005] Secondly, this application provides a distribution network adequacy enhancement strategy generation system, comprising: The distributed renewable energy output uncertainty characterization module is used to characterize the output uncertainty of distributed renewable energy based on historical output data of distributed renewable energy using generative adversarial networks. The electric vehicle cluster operation boundary uncertainty characterization module is used to generate user response intention samples based on price incentives, and characterize the uncertainty of the electric vehicle cluster operation boundary caused by the uncertainty of user intentions; A two-layer optimization model establishment module for sufficiency enhancement strategy is used to establish a two-layer optimization model for sufficiency enhancement strategy that considers multidimensional uncertainties; the two-layer optimization model for sufficiency enhancement strategy includes an upper-layer model and a lower-layer model; The upper-level model is used to consider the uncertainty of the output of the distributed new energy sources. It adopts a two-stage robust model to formulate a backup strategy and generate a node marginal electricity price that takes into account the uncertainty, forming a backup management price mechanism to guide electric vehicle users to provide backup. The lower-level model is used to generate a distribution network adequacy improvement strategy for the electric vehicle cluster to provide backup to the distribution network based on the backup management price mechanism, taking into account the uncertainty of the electric vehicle cluster's operating boundary, and introducing a data-driven sub-Bruker opportunity constraint form.

[0006] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for generating a distribution network adequacy enhancement strategy.

[0007] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for generating a distribution network adequacy enhancement strategy.

[0008] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for generating a distribution network adequacy enhancement strategy.

[0009] According to the specific embodiments provided in this application, this application has the following technical effects: Addressing the issue of sufficient power supply in power distribution networks, this application uses electric vehicle (EV) users as a typical demand-side resource. Considering the uncertainty of distributed renewable energy output and user willingness, and taking into account the safe and stable operation of the power distribution network, it incorporates the multidimensional uncertainties of distributed renewable energy and user willingness on the demand side. A two-layer optimization model for improving sufficient power supply considering multidimensional uncertainties is established. This model reasonably quantifies the direct and indirect factors that cause risks to the sufficiency of the power distribution network, and designs a reasonable strategy for improving the sufficiency of the power distribution network. This effectively addresses the insufficient power supply caused by fluctuations in renewable energy, thereby helping to ensure the safe electricity use of users in the new power system and contributing to the achievement of dual-carbon goals. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic diagram of a method for generating a distribution network adequacy enhancement strategy according to an embodiment of this application; Figure 2 A schematic diagram of a generative adversarial network provided in an embodiment of this application; Figure 3 A schematic diagram illustrating the uncertainty of the operating boundary of an electric vehicle cluster provided in an embodiment of this application; Figure 4 This is a schematic diagram of a two-layer optimization model for an adequacy enhancement strategy provided in an embodiment of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] To make the objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] like Figure 1 As shown in the figure, this application provides a method for generating a distribution network adequacy improvement strategy, including: S1: Using generative adversarial networks, based on historical power output data of distributed renewable energy sources, the uncertainty of distributed renewable energy output is characterized.

[0015] S2: Generate user response intention samples based on price incentives to characterize the uncertainty of the electric vehicle cluster operation boundary caused by the uncertainty of user intentions.

[0016] S3: Establish a two-layer optimization model for sufficiency enhancement strategy that considers multidimensional uncertainty; the two-layer optimization model for sufficiency enhancement strategy includes an upper-layer model and a lower-layer model.

[0017] S4: In the upper-level model, considering the uncertainty of the output of the distributed new energy, a two-stage robust model is adopted to formulate a backup strategy and generate a node marginal electricity price that takes into account uncertainty, forming a backup management price mechanism to guide electric vehicle users to provide backup.

[0018] S5: In the lower-level model, based on the backup management price mechanism and considering the uncertainty of the electric vehicle cluster's operating boundary, a data-driven sub-Blu-ray opportunity constraint is introduced to generate a distribution network adequacy improvement strategy for the electric vehicle cluster to provide backup to the distribution network.

[0019] In an exemplary embodiment, compared to traditional scene generation methods, the generative adversarial network (GAN) scene generation method directly utilizes historical data to train network parameters to obtain scenes that more closely resemble real-world distributions. Based on the collected historical data of distributed renewable energy output, a large number of wind and solar power output datasets closely resembling real-world distributions can be generated through a generator and an adversarial network, providing support for the generation of subsequent scheduling strategies. S1 specifically includes: Generative adversarial networks are used to generate a set of landscape output scenes, such as Figure 2 As shown. Clustering methods are used to analyze the generated set of wind and solar power output scenes. From the samples A discrete typical scenario And this probability distribution is used as the initial probability distribution.

[0020] A comprehensive norm constraint is constructed to restrict the value of the scenario probability distribution, ensuring that it fluctuates within a reasonable range, as shown in equation (1).

[0021] (1) In the formula: and They represent the first The probability distribution of each scenario and the initial probability distribution; , They are respectively 1-norm and -norm; and Used to characterize the conservatism of a model, indicating It can deviate The maximum value. Probability distribution. Satisfy the confidence constraints of equations (2)-(3).

[0022] (2) (3) Let the right sides of equations (2) and (3) represent the confidence levels of the uncertainty probability, respectively. and ,but and It can be calculated using equation (4).

[0023] (4) In one exemplary embodiment, user response intention samples are generated based on price incentives. Considering that the response intentions of electric vehicle users affect the feasible domain of energy and power, a VB model is used to characterize the uncertainty of the electric vehicle cluster operation boundary caused by the uncertainty of user intentions through aggregation. The price incentives refer to price incentives provided by different EV travel scenarios and by the electric vehicle aggregator (EVA).

[0024] S2 specifically includes: S21: Based on the price incentive, define the user response intention state that follows a binomial distribution and generate user response intention samples.

[0025] S22: Based on the user response intention samples, and according to the user response intention status, the EV cluster is divided into EV clusters that receive control and EV clusters that do not accept control.

[0026] S23: Determine the feasible domain aggregation model based on the different types of EV clusters.

[0027] S24: Based on the feasible region clustering model, characterize the uncertainty of the electric vehicle cluster operation boundary caused by the uncertainty of user intentions, such as... Figure 3 As shown.

[0028] In practical applications, different travel scenarios of EV and the incentives provided by EVA may affect user decisions. In order to characterize user response intentions, this application defines the user response intention state that follows a binomial distribution, as shown in Equation (5).

[0029] (5) In the formula: Total number of users For users The state of mind; For users willingness The mathematical expectation is The 0-1 distribution, The user's willingness to respond follows a 0-1 distribution.

[0030] It is believed that the EV user's willingness to respond is related to the incentives provided to the EV by EVA. Based on equation (5), the EV cluster is divided into an EV cluster that accepts regulation and an EV cluster that does not accept regulation. They expressed their unwillingness to accept regulation. This indicates acceptance of regulation. For EV clusters that accept regulation, when the DSO issues a regulation signal, the EVs adjust their power levels according to the signal. For EV clusters that do not accept regulation, they complete the charging process according to the planned power.

[0031] The energy and power boundaries of a cluster of EVs can be generated by aggregating individual EVs: (6) (7) In the formula: The number of EVs in the cluster; For the first The car is Grid connection status during different time periods and The first The car is Maximum and minimum electrical energy and maximum and minimum power during a given time period.

[0032] Meanwhile, in order to correct the step phenomenon of cluster EV power value caused by EV grid connection and off-grid periods, power change is introduced for correction, as shown in equation (8).

[0033] (8) in, This represents the step value that occurs in the VB model when electric vehicle users connect to or disconnect from the grid; Let be the maximum energy value of the i-th vehicle in time period t+1; This indicates the grid connection status of the i-th vehicle during time period t+1; This indicates the grid connection status of the i-th vehicle during time period t.

[0034] Therefore, the feasible domain aggregation model of the cluster EV can be obtained. For both EV clusters that accept regulation and EV clusters that do not accept regulation, the feasible domain aggregation model can be generated according to equations (9)-(10).

[0035] (9) (10) In the formula: Energy for electric vehicles when they are connected to the grid. These represent the maximum and minimum charging power of the electric vehicle cluster during time period t, respectively. These represent the maximum and minimum energy values ​​of the electric vehicle cluster during time period t, respectively. Let K be the power value of the electric vehicle cluster during time period k. This represents the energy step value of the electric vehicle cluster during time period k due to the electric vehicles going in and out of the grid.

[0036] First, the electric vehicle cluster generates multiple sets of user intention data by sampling the response intentions of electric vehicle users in the cluster. Then, through the feasible domain aggregation model, it generates operating parameters for the EV cluster that accepts regulation and operating parameters for the EV cluster that does not accept regulation. Each set of user intentions corresponds to a set of operating parameters, thus generating a data sample set.

[0037] As the number of user intention samplings increases, although it still exhibits randomness, it reveals certain patterns in the operating boundary of electric vehicles. When the number of samplings reaches a certain level, these patterns become apparent. Through the above sampling, a set of fuzzy data samples of the electric vehicle operating boundary can be obtained to reflect uncertainty.

[0038] In an exemplary embodiment, a two-layer optimization model for an adequacy enhancement strategy considering multidimensional uncertainties is established. At the upper layer, the power distribution system operator establishes a two-stage robust model with the objective of minimizing total operating costs, considering the uncertainty of wind and solar power output. At the lower layer, considering the uncertainty of electric vehicle operating boundaries, an optimization model is established with the objective of maximizing the revenue of electric vehicle aggregators. S3 specifically includes: A two-layer optimization framework considering the risk of distribution network supply security, namely, a two-layer optimization model for sufficiency improvement strategies, is as follows: Figure 4 As shown, considering the supply guarantee risks and grid constraints brought about by the uncertainty of wind and solar power, the distribution network optimizes its scheduling with the goal of minimizing its total operating cost. The scheduling results are used to update the control signals of each electric vehicle cluster and are then distributed to each cluster. When electric vehicle clusters submit their reserve requests based on the reserve compensation price issued by the distribution network, aiming to maximize their own revenue, the clusters quantify the willingness of electric vehicle users through price compensation incentives, thus encouraging users to participate in the reserve market.

[0039] In an exemplary embodiment, at the upper layer, considering the uncertainty of distributed renewable energy output, a two-stage robust model is used to formulate a reserve strategy and generate a node marginal electricity price that considers uncertainty. Based on this, a reserve management price mechanism is formed to guide electric vehicle users to provide reserves. Due to limitations in line transmission capacity, relying solely on the distribution network to purchase electricity from the main grid cannot compensate for the power deficit caused by renewable energy prediction errors. At this time, load shedding occurs in the distribution network, indicating insufficient distribution network adequacy and a supply guarantee problem. Therefore, the load shedding cost is defined as the supply guarantee risk cost. During day-ahead dispatching, considering the uncertainty of wind and solar power, the supply guarantee risk cost is expressed as the sum of the product of scenario probability and load shedding cost. A two-stage robust model is used to incorporate the supply guarantee risk cost into the objective function, optimizing the generation of distribution network reserve strategy and price compensation strategy. Considering the spatiotemporal distribution characteristics of distributed renewable energy, a reserve management price mechanism is formed to guide electric vehicle users to provide reserves. S4 specifically includes: Due to limitations in transmission capacity and other factors, relying solely on the distribution network to purchase electricity from the main grid cannot compensate for the power deficit caused by renewable energy forecasting errors. In such cases, load shedding occurs in the distribution network, indicating insufficient sufficiency and a supply guarantee issue. Therefore, the load shedding cost is defined as the supply guarantee risk cost. During day-ahead dispatching, considering the uncertainties of wind and solar power, the supply guarantee cost is incorporated into the analysis of uncertain scenarios. The sum of the scenario probability and the load shedding cost is expressed as the supply guarantee risk cost, which measures the supply guarantee risk of the distribution network.

[0040] The objective function of the upper-level model is to minimize the total operating cost, which includes: wholesale market electricity purchase cost, gas turbine generator cost, standby service cost for electric vehicle clusters, load shedding risk cost, and wind and solar curtailment cost. This application models the split-bar optimization scheduling as a two-stage, three-layer optimization problem. The first stage uses wholesale market electricity purchase, gas turbine generator cost, electric vehicle cluster output, and wind and solar curtailment as decision variables to minimize the wholesale market electricity purchase cost, gas turbine generator cost, standby service cost for electric vehicle clusters, and wind and solar curtailment cost. The second stage seeks the worst-case probability distribution that maximizes the minimum load shedding risk cost, as shown in Equation (11).

[0041] (11) (12) (13) (14) (15) (16) (17) In the formula: F DSO This represents the total operating cost of the power distribution network operator; , These are the variables for the first stage and the variables for the second stage, respectively. For the first One scenario; For the first Second-stage variables in each scenario; The above is a comprehensive norm constraint on the probability distribution of the scene; Given a set Time optimization variables The feasible domain; The cost of purchasing electricity from the wholesale market for DSOs; Cost of generating electricity from gas turbine units; and These are the costs of EVA's upper and lower backup services, respectively. For load shedding costs; For load shedding costs; For time period labels; For the distribution system operator (DSO) in Electricity was purchased from the wholesale market during certain periods. For the first Taiwan gas turbine unit During specific time periods; and The first An EVA in The amount of upward and downward adjustment during the time period; and The distribution network is the first Each node Load shedding and wind / solar curtailment during specific time periods; for Wholesale market electricity purchase price during certain time periods; and This represents the power generation cost coefficient of a gas turbine unit; and These are the clearing prices for both upper and lower reserves in the power grid; and These are the load shedding penalty coefficient and the wind / solar curtailment penalty coefficient, respectively.

[0042] First, in order to ensure the safe operation of the distribution network, the constraints shown in equations (18)-(23) should be satisfied.

[0043] (18) (19) (20) (twenty one) (twenty two) (twenty three) In the formula: This refers to the system's rated voltage amplitude. This refers to the voltage amplitude at the substation's outlet. This represents the maximum permissible voltage deviation at each node in the distribution network. , These are the line resistance and reactance between node i and node j, respectively; , These represent the active and reactive power of the branch from node i to node j, respectively. , These represent the active and reactive power of node i, respectively. Let be the square of the branch voltage from node i to node j; Let be the square of the branch current from node i to node j.

[0044] Secondly, each distribution network node should satisfy the node power balance constraints shown in equations (24)-(25).

[0045] (twenty four) (25) In the formula: p j For nodes j active power, and The distribution network is the first Each node Active and reactive loads during a given time period For the distribution network Each node Forecasted output of new energy sources during the period As the output baseline of EVA, For the first The uncertainty of each scenario.

[0046] The power generation of a gas turbine unit includes active power constraints and reactive power constraints. The amount of electricity that the DSO purchases from the wholesale market is subject to a maximum constraint, as shown in equations (26)-(28).

[0047] (26) (27) (28) In the formula: , These are the upper and lower limits of the active power output of the i-th gas turbine unit, respectively. , These are the upper and lower limits of the reactive power output of the i-th generator set, respectively. The maximum amount of electricity that a DSO can purchase from the wholesale market, subject to line capacity constraints.

[0048] Furthermore, the upper and lower reserves that electric vehicles can provide satisfy the constraints shown in equations (29)-(31).

[0049] (29) (30) (31) In the formula, and These refer to the upper and lower backups that EVA can provide, respectively.

[0050] In order to ensure power supply and maximize the use of new energy sources, there are maximum value constraints on the load shedding and wind and solar curtailment of each node in the distribution network, as shown in equations (32)-(34).

[0051] (32) (33) (34) In the formula: Equation (31) indicates that load shedding and wind / solar curtailment are not allowed to occur simultaneously, representing the maximum allowable load shedding value.

[0052] The uncertain nodal marginal electricity price is derived from the Lagrange function. Let represent the Lagrange function corresponding to (24), with the remaining constraints remaining implicit. Therefore, the Lagrange function is... L The simplified representation is as follows: (35) In the formula: Indicates the number of iterations; This represents the total number of iterative solutions; Indicates the first The worst-case scenario obtained in the next iteration; For the dual variable corresponding to constraint (24).

[0053] The marginal electricity price at an uncertain node can be defined as the marginal cost of the uncertainty increment in the network. Therefore, for a given amount of uncertainty... The uncertain nodal marginal electricity price is defined as: (36) in, The dual variable is the corresponding dual variable in the two-level optimization model of the sufficiency improvement strategy.

[0054] Because it is node The price in the s-th iteration, i.e., the marginal electricity price at the node considering uncertainty, is used to generate a reserve compensation price by aggregating these ULMPs. This price can effectively guide the entities in the lower-level model to provide reserve capacity to the distribution network through specific payment methods. (37) This application explicitly considers the scenario where electric vehicle clusters and distributed renewable energy sources may be connected to different nodes in the distribution network topology. This spatial inconsistency means that the reserve compensation price specified in equation (36) may lead to biased uncertainty estimations of the electric vehicle cluster nodes. To address this limitation, we propose the following improved method for reserve compensation pricing.

[0055] In response to the above situation, taking the uncertain marginal electricity price of distributed new energy access nodes as a reference, a price mechanism as shown in the following formula (38) is set to manage the reserve capacity, and to avoid excessive losses caused by line transmission, the principle of "local consumption" is established.

[0056] (38) In the formula: A backup management price mechanism; A reserve management price mechanism; Marginal electricity prices at nodes with uncertainties related to backup power; Marginal electricity prices at nodes with uncertainties related to backup; The weight of the impact of renewable energy nodes on management nodes; l For distribution network node serial numbers; This represents the set of nodes where renewable energy exists. This indicates the degree to which the reserve compensation price of a node is affected by its optimal load management plan for renewable energy access nodes.

[0057] In an exemplary embodiment, considering the uncertainty of the cluster operation boundary caused by the uncertainty of user response willingness, a data-driven sub-Bruker chance constraint method is introduced to generate a strategy for the electric vehicle cluster to provide backup to the distribution network. Since the precise distribution of the uncertainty of the electric vehicle cluster operation boundary is difficult to characterize, a data-driven sub-Bruker chance constraint method is adopted. This method can mine the potential distribution information of random variables from the available data support, and can achieve easily tractable convex reconstruction. Simultaneously, the conservatism of the model can be adjusted by controlling relevant parameters. S5 specifically includes: In the lower-level model, considering the uncertainty of the cluster operation boundary caused by the uncertainty of user response willingness, a data-driven sub-Bruker chance constraint method is introduced to generate a strategy for electric vehicle clusters to provide backup to the distribution network.

[0058] Because user intentions are represented by 0-1 variables, the power and energy boundaries of EV clusters exhibit significant uncertainty after aggregation. To describe the uncertainty arising from these user intentions, a data-driven, distributed bar chance constraint method is employed for modeling. Unlike mainstream stochastic optimization methods, this method can mine the latent distribution information of random variables from available data, achieve easily tractable convex reconstruction, and adjust the model's conservatism by controlling relevant parameters.

[0059] Based on the obtained fuzzy data sample set, establish parameters. , , as well as Based on a data-driven, segmented probability constraint expression, this method constructs a fuzzy set of probability distributions by identifying the worst-case scenario among distributions adjacent to the sample distribution. This is based on N uncertain observations obtained through sampling. ,in Define the known empirical probability distribution of the samples as ,in Therefore The Dirac distribution centered on the support set. Fuzzy sets and empirical distributions Distance between As shown in the following formula: (39) In the formula: Q is and Regarding the margin and The joint probability distribution of fuzzy sets, therefore, fuzzy sets As shown below: (40) in, For the distribution of random variables; The empirical distribution of user response intention samples; It is a random variable; Let be the radius of the fuzzy set.

[0060] At the same time, in order to eliminate the potential possibility that the worst-case distribution within the set takes an unrealistic value, support sets can be used. Adding supporting information that may correspond to uncertain physical information, typically using multifaceted sets. To indicate, among which and This is the coefficient vector.

[0061] Based on the above probability distribution fuzzy set For EV clusters that are subject to regulation, the parameters , , , as well as It can be represented as a data-driven, distributed bar chance constraint, as shown below.

[0062] (41) (42) (43) (44) In the formula: For probability operators, For the selected confidence interval; For EV clusters to be regulated Planned power for the time period; , The EV clusters under control are respectively The available upper and lower backups for different time periods.

[0063] For electric vehicle clusters that do not accept regulation, the energy boundary constraints are as shown in equations (41)-(42), and the power boundary constraints are as shown in equations (45)-(46).

[0064] (45) (46) In the formula: For electric vehicle clusters that do not accept regulation Total power over a given period.

[0065] The electric vehicle cluster aims to maximize its own revenue, as shown in the objective function below, which includes electricity sales revenue, reserve revenue, and charging costs. Electric vehicle clusters subject to regulation can provide reserve capacity, specifically divided into upper and lower reserves. Electric vehicle clusters not subject to regulation do not provide upper or lower reserve services to the distribution network and only complete charging according to the planned power output.

[0066] (47) in: (48) (49) (50) (51) (52) (53) (54) In the formula: F The total revenue obtained for the electric vehicle cluster; The revenue obtained by the regulated electric vehicle cluster; The revenue gained by electric vehicle clusters that do not accept regulation; Revenue generated from providing charging services to unregulated electric vehicle fleets; Revenue generated by electric vehicle clusters providing charging services to regulated electric vehicle users; Revenue generated from providing backup services for electric vehicle clusters; Charging costs for regulated electric vehicle fleets; The charging cost for an unregulated electric vehicle fleet; t is the time period label. and These are the start and end times for optimization, respectively. For time intervals; For electric vehicle users who are subject to regulation Total power over the time period; and For electric vehicle clusters The upper and lower reserve control signals for different time periods; for Electricity prices for time-of-use electric vehicle clusters and electric vehicle user charging service transactions; for The charging electricity price for electric vehicle clusters trading with the grid during specific time periods; The charging compensation price provided by the electric vehicle cluster to electric vehicles participating in the regulation.

[0067] In the process of optimizing the electric vehicle cluster strategy, in addition to satisfying the opportunity constraints shown in equations (41)-(46), the electric vehicle cluster parameters should also satisfy the constraints shown in equation (55). The power of the electric vehicle cluster under control during the charging process is shown below, and the following equation is the constraint of the electric vehicle charging and discharging power on the reserve capacity.

[0068] (55) This application provides a distribution network adequacy enhancement strategy generation system, including: The distributed renewable energy output uncertainty characterization module is used to characterize the output uncertainty of distributed renewable energy based on historical output data of distributed renewable energy using generative adversarial networks.

[0069] The electric vehicle cluster operation boundary uncertainty characterization module is used to generate user response intention samples based on price incentives, and characterize the uncertainty of the electric vehicle cluster operation boundary caused by the uncertainty of user intentions.

[0070] The module for establishing a two-layer optimization model for sufficiency enhancement strategy is used to establish a two-layer optimization model for sufficiency enhancement strategy that considers multidimensional uncertainties; the two-layer optimization model for sufficiency enhancement strategy includes an upper-layer model and a lower-layer model.

[0071] The upper-level model is used to consider the uncertainty of the output of the distributed new energy sources. It adopts a two-stage robust model to formulate a backup strategy and generate a node marginal electricity price that takes into account the uncertainty, thus forming a backup management price mechanism to guide electric vehicle users to provide backup power.

[0072] The lower-level model is used to generate a distribution network adequacy improvement strategy for the electric vehicle cluster to provide backup to the distribution network based on the backup management price mechanism, taking into account the uncertainty of the electric vehicle cluster's operating boundary, and introducing a data-driven sub-Bruker opportunity constraint form.

[0073] This application is applicable to various demand-side resources such as distributed energy storage, smart building air conditioning, and electric heating.

[0074] This application fully considers the impact of the complex interplay of uncertainties in distributed renewable energy output and electric vehicle user willingness on the adequacy risk of the distribution network. This invention introduces marginal electricity prices at uncertain nodes to form a reserve price management mechanism, which can effectively guide demand-side resources to provide reserves to the distribution network, improving its adequacy. Simultaneously, considering the uncertainty brought by the uncertainty of electric vehicle user willingness to the operational boundary of electric vehicle clusters, it innovatively adopts data-driven distributed robust opportunity constraints to solve the problem of the difficulty in accurately characterizing the precise distribution of this uncertainty, generating a robust reserve scheduling strategy. Compared with existing technologies, this invention fully considers the impact of the complex interplay of multidimensional uncertainties on the adequacy of the distribution network. Using the method of this invention, the adequacy of the distribution network can be achieved, reducing the risk of day-ahead reserve strategy failure and ensuring the electricity safety of users in the new distribution system.

[0075] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments. The computer device can be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, memory, and I / O interface are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device stores data to be processed. The I / O interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with an external terminal via a network connection. When the computer program is executed by the processor, it implements the above-described methods.

[0076] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0077] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0078] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0079] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0080] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by hardware related to computer program instructions. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0081] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0082] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0083] This application uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. In summary, the content of this specification should not be construed as a limitation of this application.

Claims

1. A power distribution network adequacy improvement strategy generation method, characterized by, The application relates to a power supply system operation method and device. The power supply system operation method comprises the following steps: An uncertainty of distributed new energy output is described by using a generative adversarial network according to historical output data of the distributed new energy; A user response willingness sample is generated according to a price incentive, and an uncertainty of an electric vehicle (EV) cluster operation boundary caused by user willingness uncertainty is described, which specifically comprises the following steps: A user response willingness state subject to a binomial distribution is defined according to the price incentive, and a user response willingness sample is generated; Based on the user response willingness sample, the EV cluster is divided into an EV cluster receiving regulation and an EV cluster not receiving regulation according to the user response willingness state; 2. The power distribution grid adequacy improvement strategy generation method of claim 1, wherein, A feasible region aggregation model is determined according to different types of EV clusters; The uncertainty of the EV cluster operation boundary caused by the user willingness uncertainty is described according to the feasible region aggregation model. The upper model is a two-stage robust model established by a power supply system operator to minimize total operation cost under the condition of considering the uncertainty of the distributed new energy output; The lower model is an optimization model established to maximize EV aggregator revenue under the condition of considering the uncertainty of the EV cluster operation boundary. The node marginal price considering the uncertainty is: The reserve management price mechanism is:

3. The power distribution grid adequacy improvement strategy generation method of claim 1, wherein, In the lower model, the power supply system operation method further comprises the following steps: For the EV cluster receiving regulation, a data-driven distributed robust chance constraint form is determined based on the probability distribution fuzzy set; 4. The power distribution grid adequacy improvement strategy generation method of claim 1, wherein, According to the data-driven distributed robust chance constraint form, the power supply system operation method generates a power supply system adequacy improvement strategy for the EV cluster to provide reserve to the power grid, wherein the EV cluster receiving regulation provides reserve capacity to the power grid; the reserve capacity comprises upper reserve capacity and lower reserve capacity; the EV cluster not receiving regulation does not provide upper reserve capacity to the power grid and only completes charging behavior according to a planned power. wherein, is the price of the s-th iteration of the node, i.e. the node marginal price considering the uncertainty; is the price of the s-th iteration of the node, i.e. the node marginal price considering the uncertainty; L is the Lagrangian function; is the worst case scenario obtained in the s-th iteration; is the worst case scenario obtained in the s-th iteration; is the dual variable corresponding to the adequacy boosting strategy bi-level optimization model; The objective function taking maximizing self-revenue of the EV cluster as an optimization target is: wherein, is the upper reserve management price mechanism; is the lower reserve management price mechanism; is the upper reserve related uncertainty nodal marginal price; is the lower reserve related uncertainty nodal marginal price; is the renewable energy source node impact weight on the management node; l is the distribution network node sequence number; is the set of nodes where renewable energy sources exist.

5. The power distribution grid adequacy improvement strategy generation method of claim 2, wherein, The application relates to a power supply system operation method and device. In the lower layer model, based on the reserve management price mechanism, considering the uncertainty of the electric vehicle cluster operation boundary, a probability distribution fuzzy set is constructed; the probability distribution fuzzy set is: ; wherein, is the distribution of a random variable; P is an empirical distribution; is the empirical distribution of the user response willingness sample; is a support set; is the distance between P and ; is a random variable; is the radius of the fuzzy set; The power supply system operation method comprises the following steps: An uncertainty of distributed new energy output is described by using a generative adversarial network according to historical output data of the distributed new energy; 6. The power distribution grid adequacy improvement strategy generation method according to claim 5, characterized by, A user response willingness sample is generated according to a price incentive, and an uncertainty of an electric vehicle (EV) cluster operation boundary caused by user willingness uncertainty is described, which specifically comprises the following steps: wherein, F total revenue obtained by the cluster of electric vehicles; revenue obtained by the cluster of electric vehicles subject to regulation, ; revenue obtained by the cluster of electric vehicles providing charging services to users of electric vehicles subject to regulation; revenue obtained by the cluster of electric vehicles providing backup services; charging cost of the cluster of electric vehicles subject to regulation; revenue obtained by the cluster of electric vehicles not subject to regulation, ; charging cost of the cluster of electric vehicles not subject to regulation; revenue obtained by the cluster of electric vehicles not subject to regulation providing charging services.

7. A power distribution network adequacy improvement strategy generation system characterized by, A user response willingness state subject to a binomial distribution is defined according to the price incentive, and a user response willingness sample is generated; Based on the user response willingness sample, the EV cluster is divided into an EV cluster receiving regulation and an EV cluster not receiving regulation according to the user response willingness state; A feasible region aggregation model is determined according to different types of EV clusters; The uncertainty of the EV cluster operation boundary caused by the user willingness uncertainty is described according to the feasible region aggregation model. The upper model is a two-stage robust model established by a power supply system operator to minimize total operation cost under the condition of considering the uncertainty of the distributed new energy output; The lower model is an optimization model established to maximize EV aggregator revenue under the condition of considering the uncertainty of the EV cluster operation boundary. The node marginal price considering the uncertainty is: The reserve management price mechanism is: In the lower model, the power supply system operation method further comprises the following steps: For the EV cluster receiving regulation, a data-driven distributed robust chance constraint form is determined based on the probability distribution fuzzy set; According to the data-driven distributed robust chance constraint form, the power supply system operation method generates a power supply system adequacy improvement strategy for the EV cluster to provide reserve to the power grid, wherein the EV cluster receiving regulation provides reserve capacity to the power grid; the reserve capacity comprises upper reserve capacity and lower reserve capacity; the EV cluster not receiving regulation does not provide upper reserve capacity to the power grid and only completes charging behavior according to a planned power. The objective function taking maximizing self-revenue of the EV cluster as an optimization target is: The application relates to a power supply system operation method and device. The power supply system operation method comprises the following steps: An uncertainty of distributed new energy output is described by using a generative adversarial network according to historical output data of the distributed new energy; A user response willingness sample is generated according to a price incentive, and an uncertainty of an electric vehicle (EV) cluster operation boundary caused by user willingness uncertainty is described, which specifically comprises the following steps: A user response willingness state subject to a binomial distribution is defined according to the price incentive, and a user response willingness sample is generated; Based on the user response willingness sample, the EV cluster is divided into an EV cluster receiving regulation and an EV cluster not receiving regulation according to the user response willingness state; A feasible region aggregation model is determined according to different types of EV clusters; The uncertainty of the EV cluster operation boundary caused by the user willingness uncertainty is described according to the feasible region aggregation model. The upper model is a two-stage robust model established by a power supply system operator to minimize total operation cost under the condition of considering the uncertainty of the distributed new energy output; The lower model is an optimization model established to maximize EV aggregator revenue under the condition of considering the uncertainty of the EV cluster operation boundary. The node marginal price considering the uncertainty is: The reserve management price mechanism is: In the lower model, the power supply system operation method further comprises the following steps: For the EV cluster receiving regulation, a data-driven distributed robust chance constraint form is determined based on the probability distribution fuzzy set; According to the data-driven distributed robust chance constraint form, the power supply system operation method generates a power supply system adequacy improvement strategy for the EV cluster to provide reserve to the power grid, wherein the EV cluster receiving regulation provides reserve capacity to the power grid; the reserve capacity comprises upper reserve capacity and lower reserve capacity; the EV cluster not receiving regulation does not provide upper reserve capacity to the power grid and only completes charging behavior according to a planned power. The objective function taking maximizing self-revenue of the EV cluster as an optimization target is: The distributed new energy output uncertainty characterization module is configured to characterize distributed new energy output uncertainty by using a generative adversarial network and according to historical output data of the distributed new energy; The electric vehicle cluster operation boundary uncertainty characterization module is configured to generate user response willingness samples according to price incentives, and to characterize electric vehicle cluster operation boundary uncertainty caused by user willingness uncertainty; The adequacy improvement strategy bi-level optimization model establishment module is configured to establish an adequacy improvement strategy bi-level optimization model considering multi-dimensional uncertainty; the adequacy improvement strategy bi-level optimization model includes an upper model and a lower model; The upper model is configured to consider the distributed new energy output uncertainty, to formulate a backup strategy by using a two-stage robust model and to generate a node marginal price considering uncertainty, to form a backup management price mechanism, and to guide electric vehicle users to provide backup; The lower model is configured to consider the electric vehicle cluster operation boundary uncertainty based on the backup management price mechanism, to introduce a data-driven distributed robust chance constraint form, and to generate a distribution network adequacy improvement strategy for the electric vehicle cluster to provide backup to the distribution network.

8. A computer device comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that the processor executes the computer program to implement the distribution network adequacy improvement strategy generation method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the distribution network adequacy improvement strategy generation method of any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the distribution network adequacy improvement strategy generation method of any one of claims 1-6.

Images