Method and device for segmental declaration of electricity quantity and price in forward market with participation of wind power storage power station

Abstract

The application provides a power and price segmentation declaration method and device for a wind storage power station participating in a forward market, wherein the method comprises the following steps: constructing a forward market declaration model of the wind storage power station based on an initial energy of the storage and a wind power output-out clearing price scenario set, a target function of the forward market declaration model comprising a current period net income and a next period storage energy value function; converting the forward market declaration model into a segmented linear optimization problem about accumulated declaration power; and solving the segmented linear optimization problem by using a dynamic programming algorithm to obtain a stepwise power and price declaration curve. The method generates a stepwise power and price declaration curve for the wind storage power station under the premise of not depending on information of other market subjects, not depending on accurate probability distribution assumptions, and not needing to call an external commercial optimization solver, and the stepwise power and price declaration curve takes into account current income, power shortage risk and future storage value, and has high calculation efficiency and strong engineering implementability, and the economic income and operation reliability of the wind storage power station in the forward market are improved.

Description

Technical Field

[0001] This invention relates to the field of new energy power generation technology, and in particular to a method and apparatus for segmented declaration of electricity price for wind power and energy storage power stations participating in the pre-market. Background Technology

[0002] Driven by the "dual carbon" goals, wind power, as a non-hydropower renewable energy source with high technological maturity and great commercial potential, is accelerating its large-scale development. However, the inherent volatility and unpredictability of wind power generation pose a severe challenge to the real-time balance of the power system. Therefore, integrating energy storage systems with wind farms to form wind-storage power plants has become a key path to improve the controllability of wind power and enhance grid friendliness. Through the flexible adjustment capabilities of energy storage, wind-storage power plants have the potential to increase profitability in the short-term electricity market.

[0003] Under the current market mechanism, market participants are required to submit tiered electricity volume-price declaration curves for the next hour. The market operator conducts the clearing process with the goal of maximizing social welfare, determining a unified clearing price and the winning bid volume for each participant. Wind-storage power plants must generate electricity according to the winning bid plan; if the actual output is insufficient, a penalty for power shortage must be paid.

[0004] The existing wind power and energy storage power station application strategies have three main shortcomings: (1) Most of the tiered application methods are derived from thermal power scenarios and do not take into account the cross-time coupling effect caused by the uncertainty of wind power and the limited capacity of energy storage. It is difficult to balance short-term benefits and long-term energy storage value. Moreover, the stochastic programming model on which they rely is often transformed into a mixed integer linear programming problem, which has low solution efficiency and poor scalability; (2) The method based on bi-level programming requires mastering the information of all market participants, which is not feasible in actual operation; (3) Simplifying the application mode (such as a single power-price pair or only reporting the quantity) is likely to result in zero winning bids or high fines, while model predictive control, reinforcement learning and other methods are limited by multi-step prediction errors, strong probability assumptions or complex parameter tuning, which are not practical for engineering.

[0005] Therefore, there is an urgent need for a pre-market tiered electricity pricing method that is suitable for wind-storage power plants, balances computational efficiency and economy, and does not rely on information from other market participants. Summary of the Invention

[0006] This invention provides a method and apparatus for segmented declaration of electricity price for wind power plants participating in the pre-market, which solves the problems of low solution efficiency, poor engineering practicality, and difficulty in balancing short-term benefits and long-term energy storage value caused by existing wind power plant electricity price declaration methods in the pre-market due to reliance on full market information, strong probability assumptions, or complex optimization models. It reduces the risk caused by the uncertainty of market clearing results and significantly improves solution efficiency.

[0007] On one hand, this invention provides a method for segmented electricity price declaration for wind-storage power stations participating in the pre-market, comprising: obtaining the initial energy storage capacity and wind power output-clearing price scenario set of the wind-storage power station in the current period; constructing a pre-market declaration model for the wind-storage power station based on the initial energy storage capacity and the wind power output-clearing price scenario set, wherein the objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage value function of the next period; transforming the pre-market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume; and solving the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity price declaration curve.

[0008] Furthermore, the energy storage value function is obtained through offline training and is used to characterize the long-term economic value of the remaining energy stored in the next period during the online application stage.

[0009] Furthermore, the energy storage energy value function is obtained through offline training, specifically including: constructing a frequency distribution histogram of wind power-electricity price joint based on historical wind power output data and market electricity price data; and solving for the energy storage energy value function based on the frequency distribution histogram of wind power-electricity price joint using the value iteration method of infinite time-discounted stochastic dynamic programming.

[0010] Furthermore, the step of transforming the pre-market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume includes: for each wind power output-clearing price scenario, representing the power shortage penalty as a piecewise linear function concerning the winning bid electricity volume, and representing the energy storage energy value function as a piecewise linear function concerning the winning bid electricity volume; unifying the power shortage penalty and the energy storage energy value function into a piecewise linear function concerning the cumulative declared electricity volume, so as to reconstruct the objective function of the pre-market declaration model into the sum of multiple piecewise linear functions, thereby transforming it into a piecewise linear optimization problem concerning the cumulative declared electricity volume.

[0011] Furthermore, the step of using dynamic programming to solve the piecewise linear optimization problem to obtain the tiered electricity price declaration curve includes: dividing the tiered electricity price declaration curve into multiple continuous electricity declaration segments and defining a cumulative declaration amount for each segment; under the constraint that the cumulative declaration amount of each electricity declaration segment satisfies the non-decreasing constraint, recursively calculating the optimal declaration amount for each electricity declaration segment through dynamic programming; and determining the tiered electricity price declaration curve based on the optimal declaration amount.

[0012] Furthermore, the net revenue for the current period is a function of the tiered electricity price declaration, and its value is determined based on the wind power output scenario, the clearing price scenario, and the power shortage penalty caused by failure to meet the bid volume.

[0013] Furthermore, the wind power output-clearing price scenario set includes pairs of wind power output scenarios and clearing price scenarios, and is associated with corresponding scenario probabilities.

[0014] Furthermore, the step of using a dynamic programming algorithm to solve the piecewise linear optimization problem to obtain a tiered electricity price declaration curve includes: submitting the tiered electricity price declaration curve to the electricity market operator to participate in the pre-market clearing.

[0015] Furthermore, the aforementioned pre-market declaration model is used to optimize the tiered electricity price declaration strategy for wind-storage power stations under the dual uncertainties of wind power output and market electricity price.

[0016] Furthermore, the aforementioned pre-market reporting model is defined as follows: ; ; ; in, Indicates time period The declared electricity volume corresponding to each electricity price This indicates the scenario number within the wind power output - clearing electricity price scenario cluster. This represents the total number of scenarios where wind power output is concentrated in the clearing electricity price scenario. This indicates a concentrated scenario of wind power output - clearing electricity price. The probability of its occurrence, This indicates the clearing price scenarios included in the wind power output-clearing price scenario set. Indicates the winning bid amount. This indicates the application and scheduling cycle for wind and energy storage power stations. Representing a scene The fine for power outage Representing a scene The energy value function of energy storage in the next time period. Indicates the first The declared electricity volume of the section, This indicates the total installed capacity of the wind turbine units. This indicates the maximum charging or discharging power of the energy storage power station. Indicates the current segment number. This indicates the total number of segments in the tiered application process. Indicates the first The electricity price declared by the section, Indicates the penalty coefficient. Representing a scene The maximum discharge energy of the energy storage power station This indicates the wind power output scenarios included in the wind power output - clearing price scenario set. This indicates the initial energy of the energy storage system during the current time period. This indicates the charging efficiency of the energy storage power station. Representing a scene The charging power of the energy storage station Representing a scene The discharge power of the energy storage power station This indicates the discharge efficiency of the energy storage power station. This indicates the minimum allowable energy of the energy storage power station. This indicates the maximum permissible energy of the energy storage power station. This indicates the charging status of the energy storage power station. This indicates the discharge state of the energy storage power station.

[0017] Secondly, the present invention also provides a device for segmented electricity price declaration for wind-storage power stations participating in the pre-market, comprising: an initial energy storage and scenario acquisition module, used to acquire the initial energy storage of the wind-storage power station and the wind power output-clearing price scenario set in the current period; a pre-market declaration model construction module, used to construct a pre-market declaration model for the wind-storage power station based on the initial energy storage and the wind power output-clearing price scenario set, wherein the objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage value function of the next period; a pre-market declaration model conversion module, used to convert the pre-market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume; and an electricity price declaration curve solving module, used to solve the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity price declaration curve.

[0018] Thirdly, the present invention also provides an electronic 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 segmented electricity price declaration method for wind and energy storage power stations participating in the pre-market as described above.

[0019] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the segmented declaration method for electricity price in the pre-market when a wind-storage power station participates in the market as described above.

[0020] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the segmented declaration method for electricity price in the pre-market when a wind-storage power station participates in the market, as described in any of the above-described methods.

[0021] This invention provides a method for segmented electricity price declaration for wind-storage power stations participating in the pre-market. It obtains the initial energy storage capacity and wind power output-clearing price scenario set of the wind-storage power station in the current time period, and constructs a pre-market declaration model based on these data. The objective function of the pre-market declaration model includes the net revenue of the current time period and the energy storage value function of the next time period. This transforms the pre-market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume, and uses a dynamic programming algorithm to solve the piecewise linear optimization problem, resulting in a stepped electricity price declaration curve. This method generates a stepped electricity price declaration curve for wind-storage power stations that balances current revenue, power shortage risk, and future energy storage value, without relying on information from other market participants, precise probability distribution assumptions, or external commercial optimization solvers. It boasts high computational efficiency, strong engineering feasibility, and significantly improves the economic benefits and operational reliability of wind-storage power stations in the pre-market. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating the method for segmented declaration of electricity price in the pre-market for wind and energy storage power stations, as provided in this embodiment of the invention.

[0024] Figure 2 This is a schematic diagram of the structure and energy flow relationship of a wind-storage power station provided in an embodiment of the present invention.

[0025] Figure 3 This is a graph showing the relationship between the energy storage capacity and the winning bid volume for the next hour during the online electricity price application for a wind-storage power station, as provided in this embodiment of the invention.

[0026] Figure 4 This is a schematic diagram of the relationship between the energy storage energy and the winning bid amount in the next hour in the offline training method of the energy storage value function.

[0027] Figure 5 This is a schematic diagram of the structure of the electricity price segmentation declaration device for wind and energy storage power stations participating in the pre-market provided in this embodiment of the invention.

[0028] Figure 6 This is a schematic diagram of the physical structure of the electronic device provided in the embodiment of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0030] It should be noted that, every hour, the wind-storage power station needs to submit a tiered electricity price declaration curve, provided that the actual wind power output and clearing price for the next hour are unknown. After market clearing, the wind-storage power station is required to generate electricity according to the bid-winning capacity. If the actual power generation is lower than the bid-winning capacity, a penalty must be paid for the shortfall.

[0031] Most existing methods incorporating tiered pricing strategies are designed for thermal power plants. However, the uncertainty of wind energy and the limited capacity of energy storage make the pricing strategies for wind-storage power plants fundamentally different from those for thermal power plants. Traditional pricing strategies face two limitations in wind-storage power plants. First, in terms of modeling, thermal power plants only require short-term price forecasts and have a constant generating capacity; in contrast, wind-storage power plants face issues of fluctuating output power and limited energy storage capacity, meaning that current pricing decisions can affect future supply capacity, thus requiring a trade-off between short-term and long-term profits. Second, in terms of computation and solution, tiered electricity pricing with energy storage is typically modeled as a stochastic programming problem, requiring the establishment and solution of a mixed-integer linear programming problem; and due to the uncertainties of wind power and prices, market pricing for wind-storage power plants requires more scenarios. Therefore, for current market electricity pricing, the solution efficiency and scalability of existing mathematical optimization models need improvement.

[0032] Modeling market mechanisms using a two-level programming framework typically requires information on all market participants to solve the problem. However, for real-world wind-storage power plants, obtaining information on other participants' application details or strategies is impractical. Therefore, the two-level programming method is only suitable for academic research or market mechanism design, and not for actual market applications in wind-storage power plants.

[0033] For research on wind-storage power plants, most existing application methods only provide a single power-price pair or only report the quantity. For the single power-price pair application mode, if the power plant's bid is higher than the unknown market clearing price, it may not receive any cleared power generation; if the bid is set to zero, the corresponding application will definitely be accepted by the market, thus achieving a quantity-only application mode. However, real-time power shortages caused by the uncertainty of wind power may lead to increased penalties. Regarding solution methods, model predictive control algorithms are based on time-series predictions within a specific time window, executed in a rolling cycle, generating a series of power-price application values, and only using the application value for the next time period during actual application. The performance of model predictive control largely depends on the prediction accuracy, and multi-step prediction often results in large errors. Existing dynamic programming algorithms rely on complex probabilistic assumptions about wind power and energy prices, which may be difficult to obtain. Reinforcement learning algorithms based on neural networks are highly sensitive to complex parameter tuning in practice.

[0034] In view of this, the present invention proposes a method for segmented declaration of electricity price for wind-storage power stations participating in the pre-market, specifically, Figure 1 The diagram illustrates a flowchart of the method for segmented declaration of electricity price for wind and energy storage power plants participating in the pre-market, as provided in an embodiment of the present invention.

[0035] like Figure 1 As shown, the method includes: S110, obtaining the initial energy of the wind-storage power station and the wind power output-clearing price scenario set for the current period; S120, constructing a pre-market declaration model for the wind-storage power station based on the initial energy of the energy storage and the wind power output-clearing price scenario set, wherein the objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage value function of the next period; S130, transforming the pre-market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume; S140, solving the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity price declaration curve.

[0036] Figure 2 A schematic diagram illustrating the structure and energy flow relationship of a wind-storage power station provided in an embodiment of the present invention is shown. Figure 2 In China, the total installed capacity of wind turbine units is Wind turbine units during the period The actual output is Furthermore, the amount is uncertain when declaring the electricity price for the corresponding time period; the maximum and minimum allowable energy of the energy storage power station are respectively and The maximum charging or discharging power of the energy storage power station is The application and dispatch cycle for wind and energy storage power stations is 1 hour, i.e. .

[0037] For time period Given that the initial energy of the energy storage is At that time, wind-storage power stations need to submit segmented declarations for electricity volume and price. Each declaration-clearing-operation cycle is divided into three stages.

[0038] The first stage is the application stage, where applications are given in ascending order. Individual electricity price It is necessary to determine the declared electricity volume (power) corresponding to each electricity price value. The price of each unit of electricity Indicates that the wind-storage power station is willing to... Price supply The energy, and the electricity price for each electricity unit can be bid on separately for each declared electricity unit.

[0039] The second stage is the clearing stage, during which the market clears out and discloses the clearing price. Of the above-mentioned declared electricity volume, the portion with a price lower than the clearing price is considered a successful bid; otherwise, it is considered a unsuccessful bid. The total successful bid volume is recorded as follows: The market's assessment of the total electricity volume awarded in the bids is based on... The unit energy price is used to pay the winning bid revenue to the wind-storage power station.

[0040] The third stage is the operation stage, which determines the actual wind power output. Subsequently, the wind-storage power station must, under the premise of meeting physical constraints, strive to achieve the awarded power volume as much as possible. If the actual power supply is lower than the bid-winning amount, the wind-storage power station must compensate for the total energy below the bid-winning amount according to the following regulations: The unit energy price is used to pay the fine, of which It is a constant greater than 1.

[0041] The method for segmented declaration of electricity price for wind and energy storage power stations participating in the pre-market provided in this embodiment of the invention is applied to the first stage, namely the declaration stage.

[0042] The following will provide a detailed description of steps S110-S140 and related steps.

[0043] S110: Obtain the initial energy of the wind-storage power station and the wind power output-clearing price scenario set for the current time period.

[0044] At the start of the current time period, the local monitoring system of the wind-storage power station reads the real-time status data of the energy storage system to obtain the initial energy of the energy storage for the current time period. Initial energy storage This reflects the remaining available battery power before the start of the current period. Simultaneously, based on historical wind power data, numerical weather forecast results, and historical electricity market clearing price series, scenario generation techniques such as K-means clustering or Monte Carlo sampling are used to construct a wind power output-clearing price scenario set.

[0045] The wind power output-clearing price scenario set contains S discrete scenarios, each scenario It consists of a pair of values: wind power output and the corresponding market-cleared electricity value, and is assigned a scenario probability to characterize the joint uncertainty of wind power output and market-cleared electricity price in the future and its probability of occurrence.

[0046] It should be noted that the wind power output-clearing price scenario set serves as the input basis for subsequent modeling and does not need to rely on the bidding information of other market participants.

[0047] Based on obtaining the initial energy storage capacity of the wind power storage station and the wind power output-clearing price scenario set in step S110, step S120 is further executed.

[0048] S120, Based on the initial energy of the energy storage and the wind power output-clearing price scenario set, construct a pre-market declaration model for the wind-storage power station. The objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage value function of the next period.

[0049] In this step, with the goal of maximizing overall economic benefits, a pre-market application model for wind-storage power plants is constructed.

[0050] The current market declaration model uses the tiered electricity price declaration curve as the decision variable. Its objective function consists of two parts: the first part is the net revenue for the current period, which is calculated based on the submitted tiered electricity price declaration curve, wind power output in each scenario, clearing price, and power shortage penalties caused by the actual power generation not reaching the bid volume; the second part is the energy storage value function for the next period, which is obtained in advance through offline training and is used to quantify the long-term economic value that the remaining energy of the energy storage system can bring in the future after completing the power generation task of the current period in different scenarios.

[0051] By introducing the energy value function of the next time period, the pre-market bidding model can effectively avoid short-sighted decisions caused by only pursuing the current period's returns, and can achieve coordinated optimization between short-term returns and long-term energy storage availability.

[0052] Based on step S120, which constructs a pre-market declaration model for wind-storage power stations based on the initial energy of energy storage and the wind power output-clearing price scenario set, step S130 is further executed.

[0053] S130, the current market declaration model is transformed into a piecewise linear optimization problem concerning the cumulative declared electricity volume.

[0054] Since the pre-market reporting model in step S120 involves random variables, nonlinear penalty terms, and value functions, direct solution is highly complex. Therefore, this step performs a structured transformation of the pre-market reporting model.

[0055] First, for each wind power output-clearing price scenario, the power shortage penalty is represented as a piecewise linear function of the winning bid amount. Second, the energy storage value function for the next time period is similarly approximated as a piecewise linear function of the winning bid amount. Further, leveraging the tiered bidding characteristic, the winning bid amount is expressed as the sum of the bid amounts for each segment, and the cumulative bid amount is defined as a new decision variable. Finally, the entire objective function is reconstructed as the sum of multiple piecewise linear functions of the cumulative bid amount, thus transforming the original stochastic programming problem (pre-market bidding model) into a piecewise linear optimization problem containing only continuous variables and with explicit monotonic constraints (the cumulative bid amount is sequentially non-decreasing). This transformation significantly reduces the mathematical complexity of the problem, enabling efficient solution.

[0056] Further, proceed to step S140.

[0057] S140, the piecewise linear optimization problem is solved using a dynamic programming algorithm to obtain the tiered electricity price declaration curve.

[0058] In this step, a dedicated dynamic programming algorithm is used to solve the piecewise linear optimization problem in step S130 to obtain the final tiered electricity price declaration curve.

[0059] The dedicated dynamic programming algorithm used in this step recursively advances from the last electricity declaration segment, utilizing the convexity or concavity of the piecewise linear function to retain the optimal candidate solution at each step, ultimately backtracking to obtain the optimal cumulative declared electricity for each segment. Subsequently, according to the preset segmented electricity price structure, such as equally spaced segments, the optimal cumulative declared electricity is converted into a complete tiered electricity price declaration curve. The tiered electricity price declaration curve contains M consecutive electricity segments, each corresponding to one declared electricity amount and one declared electricity price, conforming to the format requirements of the electricity market for strategic bidding.

[0060] Subsequently, the tiered electricity price declaration curves obtained will be automatically submitted to the market operator through the power trading platform to participate in the pre-market clearing.

[0061] In this embodiment, by obtaining the initial energy storage capacity and wind power output-clearing price scenario set of the wind-storage power station in the current time period, a pre-time market declaration model for the wind-storage power station is constructed based on the initial energy storage capacity and wind power output-clearing price scenario set. The objective function of the pre-time market declaration model includes the net revenue of the current time period and the energy storage value function of the next time period. The pre-time market declaration model is then transformed into a piecewise linear optimization problem concerning the cumulative declared electricity volume. A dynamic programming algorithm is used to solve the piecewise linear optimization problem, resulting in a tiered electricity price declaration curve. This method generates a tiered electricity price declaration curve for the wind-storage power station that balances current revenue, power shortage risk, and future energy storage value, without relying on information from other market participants, precise probability distribution assumptions, or external commercial optimization solvers. It boasts high computational efficiency, strong engineering feasibility, and significantly improves the economic benefits and operational reliability of the wind-storage power station in the pre-time market.

[0062] Based on the above embodiments, the following will further describe in detail the construction of the pre-market declaration model for wind-storage power stations in step S120.

[0063] Assume there exists a... discrete points The piecewise linear function formed by connecting the elements is used to represent the initial energy of energy storage. The long-term economic value of energy storage systems. Indicates time period The actual value of wind power output and clearing electricity price, among which, This represents the actual wind power output (wind power output scenario). To clear the actual value of the electricity price (clearing the electricity price scenario). Assume that information regarding electricity price can be obtained before the electricity price is declared. S scenarios This refers to the wind power output-clearing electricity price scenario set, where the scenario probability for each scenario is as follows: .

[0064] Based on the aforementioned assumptions, a pre-market electricity price declaration model for wind and energy storage power stations (i.e., the pre-market declaration model mentioned above) is established, as shown in equations (1)-(8).

[0065] (1).

[0066] (2).

[0067] (3).

[0068] (4).

[0069] (5).

[0070] (6).

[0071] (7).

[0072] (8).

[0073] In equations (1)-(8), Indicates time period The declared electricity volume corresponding to each electricity price This indicates the scenario number within the wind power output - clearing electricity price scenario cluster. This represents the total number of scenarios where wind power output is concentrated in the clearing electricity price scenario. This indicates a concentrated scenario of wind power output - clearing electricity price. The probability of its occurrence, This indicates the clearing price scenarios included in the wind power output-clearing price scenario set. Indicates the winning bid amount. This indicates the application and scheduling cycle for wind and energy storage power stations. Representing a scene The fine for power outage Representing a scene The energy value function of energy storage in the next time period. Indicates the first The declared electricity volume of the section, This indicates the total installed capacity of the wind turbine units. This indicates the maximum charging or discharging power of the energy storage power station. Indicates the current segment number. This indicates the total number of segments in the tiered application process. Indicates the first The electricity price declared by the section, Indicates the penalty coefficient. Representing a scene The maximum discharge energy of the energy storage power station This indicates the wind power output scenarios included in the wind power output - clearing price scenario set. This indicates the initial energy of the energy storage system during the current time period. This indicates the charging efficiency of the energy storage power station. Representing a scene The charging power of the energy storage station Representing a scene The discharge power of the energy storage power station This indicates the discharge efficiency of the energy storage power station. This indicates the minimum allowable energy of the energy storage power station. This indicates the maximum permissible energy of the energy storage power station. This indicates the charging status of the energy storage power station. This indicates the discharge state of the energy storage power station.

[0074] Equation (1) shows the objective function as the expected value of the sum of the net income of the current period and the energy storage value of the next period; Equation (2) limits the upper and lower limits of each declared electricity volume; Equation (3) describes the calculation method of the winning bid electricity volume; Equation (4) describes the calculation method of the power shortage penalty; Equation (5) describes the calculation method of the energy storage energy of the next period; Equation (6) limits the upper and lower limits of the energy storage energy of the next period; Equations (7)-(8) limit the upper and lower limits of energy storage charging and discharging and complementarity.

[0075] Based on the above embodiments, the following will further describe in detail the process of transforming the pre-market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume in step S130.

[0076] The current market declaration model is transformed into a piecewise linear optimization problem concerning the cumulative declared electricity volume. This includes: for each wind power output-clearing price scenario, representing the power shortage penalty as a piecewise linear function of the winning bid electricity volume, and representing the energy storage energy value function as a piecewise linear function of the winning bid electricity volume; unifying the power shortage penalty and the energy storage energy value function as a piecewise linear function of the cumulative declared electricity volume, so as to reconstruct the objective function of the current market declaration model into the sum of multiple piecewise linear functions, thus transforming it into a piecewise linear optimization problem concerning the cumulative declared electricity volume.

[0077] Specifically, firstly, the second term in the objective function (the defect penalty term) Writing about the winning bid electricity volume The piecewise linear function is as shown in equation (9).

[0078] (9).

[0079] At the same time, the third term in the objective function (energy storage value function) This refers to the amount of electricity won in the bid. A piecewise linear function. Figure 3 The figure shows the relationship between the energy storage capacity and the winning bid volume for the next hour when the online power price application for the wind-storage power station is submitted, as provided in the embodiment of the present invention.

[0080] according to Figure 3 Regarding the winning bid volume The piecewise linear function is formed by connecting the following 5 points in sequence: , , , , Thus, the third term in the objective function (energy storage value function) can be... Also written as the winning bid volume A piecewise linear composite function.

[0081] Therefore, the current market declaration model can be initially transformed into a standard form, as shown in equations (10)-(12).

[0082] (10).

[0083] (11).

[0084] (12).

[0085] Wherein, objective function Overall, the winning bid volume can be written as The piecewise linear function is as shown in equation (13).

[0086] (13).

[0087] Furthermore, the power shortage penalty and the energy storage value function are uniformly expressed as piecewise linear functions of the cumulative declared electricity volume. This reconstructs the objective function of the current market declaration model into the sum of multiple piecewise linear functions, transforming it into a piecewise linear optimization problem of the cumulative declared electricity volume.

[0088] Specifically, record The S scenarios are then categorized into M classes based on the range of clearing electricity prices, satisfying... ,in, Based on this, the objective function in the standard form is divided into M groups, defined as follows (14).

[0089] (14).

[0090] Then the standard form of the market declaration model is further transformed into the simplest form, which is a piecewise linear optimization problem for the cumulative declared electricity volume, as shown in equations (15)-(16).

[0091] (15).

[0092] (16).

[0093] In its simplest form, each term in the objective function They are continuous variables A piecewise linear function. Assume a piecewise linear function. Let Γ be the set of x-coordinates of the boundary points of all segments, containing J distinct values. Then the optimal solution must satisfy... .

[0094] It should be noted that this embodiment introduces the concept of cumulative reported electricity consumption. As a decision variable, its core advantage lies in its ability to handle arbitrary clearing electricity prices. The winning bid volume Exactly equal to satisfy Maximum cumulative declared electricity Therefore, by classifying scenarios into sets according to their clearing electricity price ranges, it can be determined that the winning bid volume for all scenarios within the same set is equal to... This decouples the complex summation relationship that originally relied on the ladder structure into independent objective function terms for each segment. This lays the foundation for achieving efficient solutions to subsequent dynamic programming problems.

[0095] Based on the above embodiments, the process of solving the piecewise linear optimization problem using a dynamic programming algorithm in step S140 will be described in detail below.

[0096] A dynamic programming algorithm is used to solve the piecewise linear optimization problem to obtain the tiered electricity price declaration curve. This includes: dividing the tiered electricity price declaration curve into multiple continuous electricity declaration segments and defining the cumulative declaration electricity for each segment; under the constraint that the cumulative declaration electricity of each electricity declaration segment satisfies the non-decreasing constraint, calculating the optimal declaration electricity for each electricity declaration segment recursively through dynamic programming; and determining the tiered electricity price declaration curve based on the optimal declaration electricity.

[0097] Specifically, to solve the piecewise linear optimization problem, a recursive solution is implemented based on dynamic programming. Establish with As decision variables, with The problem is a recursive optimization of the parameters, as shown in equations (17)-(18).

[0098] (17).

[0099] (18).

[0100] In equations (17)-(18), we define .

[0101] Based on the algorithm flow in Table 1 below, according to Find all function values ​​in order And record the corresponding optimal solution. The value is obtained in the end. This represents the optimal value of the objective function in the simplest form of the pre-market electricity price declaration model for wind and energy storage power stations, which is also the optimal value of the objective function in a piecewise linear optimization problem. Based on the optimal solution value... The optimal declared power capacity for a wind-storage power station can be obtained using the following formula. ,Right now This allows us to obtain a tiered electricity price declaration curve.

[0102] Table 1

[0103] Based on the above embodiments, the following will further describe in detail the offline training process of the energy storage value function for the next time period in the objective function of the pre-market declaration model in step S120.

[0104] The energy storage value function is obtained through offline training and is used to characterize the long-term economic value of the remaining energy stored in the next period during the online application stage.

[0105] The energy storage value function is obtained through offline training, specifically including: constructing a joint frequency distribution histogram of wind power and electricity price based on historical wind power output data and market electricity price data; and solving for the energy storage value function using the value iteration method of infinite time-discounted stochastic dynamic programming based on the joint frequency distribution histogram of wind power and electricity price.

[0106] Specifically, when solving the online electricity price declaration strategy for wind and energy storage power stations, it is necessary to obtain the data in advance from... discrete points Piecewise linear functions formed by connecting the pieces , used to represent the initial value of the stored energy The long-term economic value of the energy storage system at that time. Assuming the energy storage capacity... Within its feasible range The internal values ​​are uniformly selected. The energy storage value function is derived through the following process. Offline training.

[0107] First, based on sufficient historical data of wind power and market electricity prices at one-hour intervals, a frequency distribution histogram is constructed to estimate the probabilistic characteristics of the random variables. Specifically, representative values ​​of historical wind power and market electricity prices are selected. For wind power, feasible intervals are defined. Unevenly divided into Let the intervals be denoted as . Each interval must contain an equal number of historical data sample points; for market electricity prices, the interval between the minimum and maximum historical data values ​​is unevenly divided into... Let the intervals be denoted as . Each interval must contain an equal number of historical data sample points. This applies to two-dimensional random variables composed of wind power and market electricity prices. It is divided into the following categories: Areas: Among them, the region The representative value (denoted as) () represents the mean of all sample points falling within this region. This represents the total number of sample pairs consisting of wind power output and market electricity price.

[0108] Secondly, assess the probability of these representative values ​​occurring. Representative values The probability of occurrence (denoted as) The value is equal to the proportion of the number of sample points falling into that region to the total sample size.

[0109] Based on the above steps, construct a frequency distribution histogram. .

[0110] Furthermore, based on the frequency distribution histogram From the given initial value Set out to seek energy storage The corresponding value function value Based on the value iteration theory of infinite-time discounted stochastic dynamic programming, in obtaining the th... The function value at the next iteration Then, the following optimization problems (19)-(21) are established to solve the th problem. The function value at the next iteration .

[0111] (19).

[0112] (20).

[0113] (twenty one).

[0114] Each sample represents a value The corresponding objective function is shown in equation (22).

[0115] (twenty two).

[0116] Among them, the function It is composed of discrete points A piecewise linear function formed by connecting the pieces; Represents the stored energy for subsequent periods; The future value discount factor has a range of values. Appropriate selection is required.

[0117] It should be noted here that optimization problems (19)-(21) are essentially the same as the pre-market reporting model mentioned above. The difference is that the scenario in the pre-market reporting model is s, while the "scenario" in optimization problems (19)-(21) is z, which is sampled from the constructed frequency distribution histogram. Based on this, the meaning of the relevant parameters in optimization problems (19)-(21) can be referred to the meaning of the corresponding parameters in the pre-market reporting model.

[0118] Figure 4 This diagram illustrates the relationship between the energy storage capacity and the winning bid amount in the next hour using the offline training method for the energy storage value function. According to... Figure 4 , about The piecewise linear function is formed by connecting the following 5 points in sequence: , , , , .

[0119] Following the same method as steps S130-S140, solve the... The function value at the next iteration The optimization problem is transformed into its simplest form and solved using the algorithm flow shown in Table 1.

[0120] Using the optimization problems (19)-(21) mentioned above, an algorithm (Algorithm 2) is established as shown in Table 2 below. Through value iteration of infinite time-discounted stochastic dynamic programming, the energy storage capacity is finally obtained. The corresponding value function value .in, This is a constant representing the upper limit of error and needs to be selected appropriately.

[0121] It should be noted that the optimization problem-solving process called in step 3 of Algorithm 2 is exactly the same as that of Algorithm 1 described in step S140. That is, offline training and online application share the same dynamic programming solution engine, only the input data (scene set vs. histogram representative value) and the objective are different. This design can significantly reduce the system implementation complexity and ensure the consistency of the solution logic.

[0122] It should also be noted that in actual engineering deployments, the number of power reporting segments M is usually set to 3-5 segments to meet market rule requirements, and the number of scenarios S can be set to 10-50 to balance accuracy and efficiency, as well as the number of discrete energy points. A value between 100 and 200 is sufficient to guarantee the accuracy of the value function and the upper limit of error. It can be set to 10 -3 ~10 -2Yuan / MWh. All of the aforementioned parameters can be flexibly configured according to the specific scale of the wind-storage power station and market rules, without relying on specific hardware or external software.

[0123] Complete the energy value function of energy storage After offline training, the data is stored in the local database of the wind-storage power station as a piecewise linear function. Therefore, during online electricity price declarations, the system will calculate the initial energy of the energy storage based on the current time period. The corresponding data can be retrieved in real time by looking up a table. The value is then embedded into the objective function of the prior market declaration model. This design achieves efficient coupling between offline training and prior declaration decision-making, avoiding the computational burden of solving high-dimensional dynamic programming problems online while ensuring accurate quantification of long-term benefits.

[0124] The method for segmented electricity price declaration for wind-storage power stations participating in the pre-market provided in this invention has two aspects. First, for online electricity price declarations by wind-storage power stations, a pre-market declaration model based on single-step probability prediction and the energy storage value function is constructed. The objective function consists of the net revenue for the current period and the energy storage value function for the next period, used to avoid short-sighted decision-making. For the pre-market declaration model, this invention employs a dynamic programming algorithm with polynomial time complexity. This algorithm requires no external optimization software and can efficiently find the global optimal solution.

[0125] On the other hand, for offline training of the energy storage value function, it is implemented through infinite-time-discount stochastic dynamic programming and solved using a value iteration method based on frequency distribution histograms. In this invention, the optimization problem involved in offline training can be implemented using the same dynamic programming algorithm as online application, thus achieving higher solution efficiency than existing methods (Algorithm 1 has a computational time complexity of O(n), which is polynomial, while existing methods typically require solving mixed-integer linear programming, whose computational time complexity is usually exponential). Compared with the probability transition matrix description commonly used in reinforcement learning, this invention uses frequency distribution histograms to represent probabilistic characteristics, eliminating the need to describe state transitions, making it easier to obtain and more adaptable to uncertainty.

[0126] Corresponding to the method for segmented declaration of electricity price for wind and energy storage power plants participating in the pre-market as described in the above embodiments, the present invention also provides a device for segmented declaration of electricity price for wind and energy storage power plants participating in the pre-market.

[0127] Specifically, Figure 5 The diagram shows a schematic representation of the structure of the electricity price segmentation declaration device for wind and energy storage power stations participating in the pre-market, provided in an embodiment of the present invention.

[0128] like Figure 5As shown, the device includes: an initial energy storage and scenario acquisition module 510, used to acquire the initial energy storage of the wind-storage power station and the wind power output-clearing price scenario set in the current period; a pre-market declaration model construction module 520, used to construct a pre-market declaration model for the wind-storage power station based on the initial energy storage and the wind power output-clearing price scenario set, wherein the objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage value function of the next period; a pre-market declaration model conversion module 530, used to convert the pre-market declaration model into a piecewise linear optimization problem about the cumulative declared electricity volume; and an electricity volume and price declaration curve solving module 540, used to solve the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity volume and price declaration curve.

[0129] In this embodiment, the initial energy storage and scenario acquisition module 510 acquires the initial energy storage energy and wind power output-clearing price scenario set of the wind-storage power station in the current period. The pre-market declaration model construction module 520 constructs a pre-market declaration model for the wind-storage power station based on the initial energy storage energy and wind power output-clearing price scenario set. The objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage energy value function of the next period. Then, the pre-market declaration model conversion module 530 transforms the pre-market declaration model into a piecewise linear optimization problem about the cumulative declared electricity volume. The electricity volume and price declaration curve solution module 540 uses a dynamic programming algorithm to solve the piecewise linear optimization problem to obtain a stepped electricity volume and price declaration curve. This device generates a stepped electricity volume and price declaration curve for the wind-storage power station that takes into account current revenue, power shortage risk, and future energy storage value without relying on information from other market participants, precise probability distribution assumptions, or external commercial optimization solvers. It has high computational efficiency, strong engineering feasibility, and significantly improves the economic benefits and operational reliability of the wind-storage power station in the pre-market.

[0130] It should be noted that the electricity price segmentation declaration device for wind and energy storage power stations participating in the pre-market provided in this embodiment of the invention can be referred to in correspondence with the electricity price segmentation declaration method for wind and energy storage power stations participating in the pre-market described in the above embodiments, and will not be repeated here.

[0131] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call logic instructions in the memory 630 to execute a segmented electricity price declaration method for wind power and energy storage power stations participating in the pre-market. This method includes: obtaining the initial energy storage capacity and wind power output-clearing price scenario set of the wind power and energy storage power station in the current period; constructing a pre-market declaration model for the wind power and energy storage power station based on the initial energy storage capacity and the wind power output-clearing price scenario set, wherein the objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage energy value function of the next period; transforming the pre-market declaration model into a piecewise linear optimization problem about the cumulative declared electricity volume; and solving the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity price declaration curve.

[0132] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0133] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the segmented electricity price declaration method for wind power stations participating in the pre-market provided by the above methods. The method includes: obtaining the initial energy storage capacity and wind power output-clearing price scenario set of the wind power station in the current period; constructing a pre-market declaration model for the wind power station based on the initial energy storage capacity and the wind power output-clearing price scenario set, wherein the objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage value function of the next period; transforming the pre-market declaration model into a piecewise linear optimization problem about the cumulative declared electricity volume; and solving the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity price declaration curve.

[0134] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a method for segmented electricity price declaration for wind power stations participating in the pre-market, as provided by the methods described above. This method includes: obtaining the initial energy storage capacity and wind power output-clearing price scenario set of the wind power station in the current time period; constructing a pre-market declaration model for the wind power station based on the initial energy storage capacity and the wind power output-clearing price scenario set, wherein the objective function of the pre-market declaration model includes the net revenue of the current time period and the energy storage value function of the next time period; transforming the pre-market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume; and solving the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity price declaration curve.

[0135] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0136] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for segmented declaration of electricity price for wind-storage power stations participating in the pre-market, characterized in that, include: Obtain the initial energy storage capacity and wind power output-clearing price scenario set of the wind-storage power station in the current time period; Based on the initial energy of the energy storage and the wind power output-clearing price scenario set, a pre-market declaration model for the wind-storage power station is constructed. The objective function of the pre-market declaration model includes the net revenue of the current period and the energy storage energy value function of the next period. The aforementioned pre-market reporting model is transformed into a piecewise linear optimization problem concerning the cumulative reported electricity volume; The piecewise linear optimization problem is solved using a dynamic programming algorithm to obtain the tiered electricity price declaration curve.

2. The method for segmented declaration of electricity price in the pre-market for wind-storage power stations as described in claim 1, characterized in that, The energy storage value function is obtained through offline training and is used to characterize the long-term economic value of the remaining energy stored in the next period during the online application stage.

3. The method for segmented declaration of electricity price in the pre-market for wind-storage power stations as described in claim 1, characterized in that, The energy storage value function is obtained through offline training and specifically includes: Based on historical wind power output data and market electricity price data, a frequency distribution histogram of wind power-electricity price linkage is constructed. Based on the frequency distribution histogram of the wind power-electricity price joint, the energy storage value function is obtained by using the value iteration method of infinite time-discounted stochastic dynamic programming.

4. The method for segmented declaration of electricity price in the pre-market for wind-storage power stations as described in claim 1, characterized in that, The step of transforming the pre-market reporting model into a piecewise linear optimization problem concerning the cumulative reported electricity volume includes: For each wind power output-clearing electricity price scenario, the power shortage penalty is represented as a piecewise linear function of the winning bid electricity volume, and the energy storage value function is also represented as a piecewise linear function of the winning bid electricity volume. The power shortage penalty and the energy storage value function are uniformly expressed as piecewise linear functions of the cumulative declared electricity volume, so as to reconstruct the objective function of the current market declaration model into the sum of multiple piecewise linear functions, thus transforming it into a piecewise linear optimization problem of the cumulative declared electricity volume.

5. The method for segmented declaration of electricity price in the pre-market for wind-storage power stations as described in claim 1, characterized in that, The stepwise electricity pricing curve is obtained by solving the piecewise linear optimization problem using a dynamic programming algorithm, including: The tiered electricity price declaration curve is divided into multiple consecutive electricity declaration segments, and the cumulative declared electricity volume is defined for each segment. Under the constraint that the cumulative declared electricity volume of each electricity declaration segment satisfies the condition of sequential non-decreasing, the optimal declared electricity volume of each electricity declaration segment is calculated recursively through dynamic programming. The tiered electricity price declaration curve is determined based on the optimal declared electricity volume.

6. The method for segmented declaration of electricity price before a wind-storage power station participates in the pre-market, as described in claim 1, is characterized in that... The net revenue for the current period is a function of the tiered electricity price declaration, and its value is determined based on the wind power output scenario, the clearing price scenario, and the power shortage penalty caused by failure to meet the bid volume.

7. The method for segmented declaration of electricity price in the pre-market for wind-storage power stations as described in claim 1, characterized in that, The process involves using a dynamic programming algorithm to solve the piecewise linear optimization problem, resulting in a tiered electricity price declaration curve, followed by: The tiered electricity price declaration curve is submitted to the electricity market operator to participate in the pre-market clearing.

8. The method for segmented declaration of electricity price in the pre-market for wind-storage power plants according to any one of claims 1-7, characterized in that, The aforementioned pre-market declaration model is used to optimize the tiered electricity price declaration strategy for wind-storage power stations under the dual uncertainties of wind power output and market electricity price.

9. The method for segmented declaration of electricity price in the pre-market for wind-storage power plants according to any one of claims 1-7, characterized in that, The pre-market reporting model is defined as follows: ； ； ； in, Indicates time period The declared electricity volume corresponding to each electricity price This indicates the scenario number within the wind power output - clearing electricity price scenario cluster. This represents the total number of scenarios where wind power output is concentrated in the clearing electricity price scenario. This indicates a concentrated scenario of wind power output - clearing electricity price. The probability of its occurrence, This indicates the clearing price scenarios included in the wind power output-clearing price scenario set. Indicates the winning bid amount. This indicates the application and scheduling cycle for wind and energy storage power stations. Representing a scene The fine for power outage Representing a scene The energy value function of energy storage in the next time period. Indicates the first The declared electricity volume of the section, This indicates the total installed capacity of the wind turbine units. This indicates the maximum charging or discharging power of the energy storage power station. Indicates the current segment number. This indicates the total number of segments in the tiered application process. Indicates the first The electricity price declared by the section, Indicates the penalty coefficient. Representing a scene The maximum discharge energy of the energy storage power station This indicates the wind power output scenarios included in the wind power output - clearing price scenario set. This indicates the initial energy of the energy storage system during the current time period. This indicates the charging efficiency of the energy storage power station. Representing a scene The charging power of the energy storage station Representing a scene The discharge power of the energy storage power station This indicates the discharge efficiency of the energy storage power station. This indicates the minimum allowable energy of the energy storage power station. This indicates the maximum permissible energy of the energy storage power station. Indicates the charging status of the energy storage power station. This indicates the discharge state of the energy storage power station.

10. A device for segmented declaration of electricity price for wind-storage power stations participating in the pre-market, characterized in that, include: The initial energy and scenario acquisition module for energy storage is used to acquire the initial energy of the wind power storage station and the wind power output-clearing price scenario set for the current period. The pre-market declaration model construction module is used to construct a pre-market declaration model for wind power storage power stations based on the initial energy of energy storage and the wind power output-clearing price scenario set. The objective function of the pre-market declaration model includes the net income of the current period and the energy storage energy value function of the next period. The current market declaration model conversion module is used to convert the current market declaration model into a piecewise linear optimization problem concerning the cumulative declared electricity volume; The electricity price declaration curve solving module is used to solve the piecewise linear optimization problem using a dynamic programming algorithm to obtain a stepped electricity price declaration curve.

Images