A power auxiliary service providing method and system considering new energy prediction error

Abstract

A power auxiliary service providing method and system considering new energy prediction error. The method calculates the total cost of frequency modulation auxiliary service based on the clearing price and the winning capacity of the frequency modulation in the thermal power unit, calculates the allocated frequency modulation auxiliary service fee of the power generation side and the user side based on the total cost of the frequency modulation auxiliary service and the on-grid power of the power generation side and the user side, calculates the allocated peak shaving auxiliary service fee and the allocated standby auxiliary service fee of the power generation side and the user side based on the power correction coefficient, constructs a characteristic function based on the prediction error, quantifies the excess function according to the characteristic function, solves the excess function by using the kernel method, and forms a new energy auxiliary service fee allocation scheme. The scheme improves the fairness, accuracy and stability of the power auxiliary service fee allocation mechanism.

Description

Technical Field

[0001] This invention belongs to the field of new energy ancillary services, and specifically relates to a method and system for providing power ancillary services that takes into account the prediction error of new energy. Background Technology

[0002] With the continuous increase in the installed capacity of new energy sources, the high proportion of new energy grid connection has led to a surge in demand for ancillary services such as frequency regulation, peak shaving, and reserve. The "Basic Rules for the Electricity Ancillary Services Market" clearly requires the establishment of an ancillary service cost transmission mechanism based on the principle of "whoever benefits, bears the cost," promoting a reasonable allocation of costs among market participants. However, the existing allocation mechanism still relies primarily on the electricity generated and fed into the grid, or the electricity consumed by users, without considering the crucial influencing factor of new energy generation forecasting errors. The randomness of new energy output leads to frequent forecasting errors, which are one of the main reasons for the increased demand for system ancillary services. The rapid development of new energy generation has brought about profound changes in grid operation. Its intermittent and fluctuating characteristics have increased the demand for ancillary services, including frequency regulation, peak shaving, and reserve. Due to the additional demand for peak shaving services caused by the characteristics of new energy generation, forecasting errors in new energy generation will exacerbate the grid's peak shaving demand. Therefore, new energy should share the system's peak shaving costs based on its generation forecasting errors. The forecasting errors in new energy generation require the system to provide additional reserves, thus also necessitating the allocation of system reserve costs. The explicitness and reasonable allocation of these costs are of great significance to the formation of new energy pricing and the healthy development of the electricity market.

[0003] The power generation side is the main provider of ancillary services. For example, thermal power plants provide frequency regulation and peak shaving services by adjusting their power generation. To compensate for the additional costs incurred by power generation companies in providing ancillary services, such as equipment wear and tear and increased fuel consumption, these costs need to be reasonably allocated to other beneficiaries through a cost-sharing mechanism. The cost-sharing portion is allocated by thermal power units and renewable energy power generation units (wind power and photovoltaic) that do not provide ancillary services according to the proportion of their power generation during the spot trading cycle to their total power generation. The larger the power generation, the more costs may be allocated. This method is simple and easy to implement, but it may not fully reflect the differences in the ability of different power generation types to provide ancillary services.

[0004] The "user side" refers to electricity users who participate in the electricity market and directly sign power purchase agreements with power generation companies or electricity sales companies. Market-based electricity users enjoy the stable and reliable power supply provided by ancillary services. Without ancillary services, parameters such as voltage and frequency of the power system may not remain within normal ranges, affecting users' normal production and business activities. According to the market principle of "whoever benefits, bears the cost," market-based electricity users should share the costs of ancillary services. This helps cultivate users' market awareness and sense of responsibility, and promotes fair competition in the electricity market. The cost of ancillary services is shared based on the proportion of electricity consumption by market-based electricity users to total electricity consumption within a certain period. The higher the electricity consumption, the higher the cost. This method is relatively simple to implement, but it may not fully consider the actual needs and varying degrees of benefit users receive from ancillary services.

[0005] Traditional cost-sharing mechanisms rely solely on either the electricity generated and fed into the grid or the electricity consumed by users, failing to consider the actual impact of renewable energy generation forecasting errors on ancillary service demands such as frequency regulation, peak shaving, and reserve. This results in a mismatch between the costs borne by renewable energy plants with large forecasting errors (i.e., those contributing significantly to ancillary service demands) and their responsibilities, while plants with high forecasting accuracy may bear more responsibility, violating the market principle of "whoever causes it, bears it." Furthermore, existing methods lack quantitative means to address forecasting errors among different renewable energy plants, making it impossible to accurately allocate the cost-sharing ratio for each plant, leading to a "one-size-fits-all" approach to renewable energy cost-sharing. In addition, traditional cooperative game theory allocation methods (such as the Shapley value method) in multi-entity (multiple renewable energy plants + conventional power sources) cost-sharing scenarios are prone to situations where some alliances are highly dissatisfied with the allocation plan, potentially triggering market player exit risks and affecting the long-term stability of the cost-sharing mechanism. The traditional model not only violates the principle of matching responsibilities emphasized by the new regulations, but also creates an imbalance in the cost-sharing between power plants with large prediction errors and those with high accuracy. In addition, it lacks the means to accurately quantify errors, and the traditional allocation method is prone to causing dissatisfaction among some market participants. It is significantly different from the requirements for fairness and stability in the construction of a unified national electricity market. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for providing electricity ancillary services that takes into account the prediction errors of new energy sources, thereby resolving the technical problems of fairness, accuracy, and stability in existing electricity ancillary service cost-sharing mechanisms.

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution.

[0008] This invention first discloses a method for providing ancillary services for electricity that takes into account the prediction error of new energy sources. The method includes the following steps:

[0009] Step 1: Calculate the total cost of frequency regulation ancillary services based on the clearing price and winning capacity of the winning bid for frequency regulation of thermal power units. Based on the total cost of frequency regulation ancillary services and the on-grid electricity on the generation side and the user side, construct a cost-sharing model for frequency regulation ancillary services on the generation side and the user side.

[0010] Step 2: Determine the power correction coefficient based on the waveform overlap between the actual power output and the equivalent power output curve of the new energy power station, and construct a peak-shaving ancillary service cost sharing model for the power generation side and the user side based on the power correction coefficient.

[0011] Step 3: Construct a cost-sharing model for backup ancillary services on the generation and user sides based on the grid-connected electricity volume on the generation and user sides;

[0012] Step 4: Based on the frequency regulation ancillary service cost sharing model, peak shaving ancillary service cost sharing model, and reserve ancillary service cost sharing model, construct a feature function based on prediction error, quantify the excess function according to the feature function, and solve the excess function using the nucleolus method to form a new energy ancillary service cost sharing scheme.

[0013] The present invention further includes the following preferred embodiments:

[0014] The construction of the frequency regulation ancillary service cost-sharing model between the generation side and the user side further includes:

[0015] Calculate the generation-side apportionment of frequency regulation ancillary service costs:

[0016] (1)

[0017] In the formula, For the first The cost of frequency modulation ancillary services shared by each unit that does not provide frequency modulation ancillary services; This indicates the number of generating units that do not provide frequency regulation ancillary services; This represents the total cost of meeting the demand for frequency modulation ancillary services within the billing cycle; This indicates the proportion of frequency regulation ancillary service costs borne by the power generation side; For the first The power consumption of units that do not provide frequency modulation auxiliary services;

[0018] Set supply and demand balance constraints for frequency modulation:

[0019] (2)

[0020] In the formula, This indicates the total number of periods for daily spot clearing; Indicates the number of thermal power units; This indicates the clearing price of the winning bid for frequency regulation in thermal power units; For the first Taiwan thermal power units The winning bid capacity for frequency modulation at any given time.

[0021] The construction of the frequency regulation ancillary service cost-sharing model between the generation side and the user side further includes:

[0022] Calculate the user-side apportionment of FM ancillary service costs:

[0023] (3)

[0024] In the formula, For the first The cost of frequency modulation ancillary services shared by each user; This represents the total number of users participating in the frequency modulation cost sharing; For the first Metrics for each user; This represents the sum of metrics for all users who participated in the cost-sharing.

[0025] The determination of the power correction coefficient based on the waveform overlap between the actual power output and the equivalent power output curve of the new energy power station includes:

[0026] Calculate the similarity between the waveform overlap of the actual power output curve and the equivalent power output curve of a renewable energy power station.

[0027] (4)

[0028] In the formula, Indicates new energy power stations Actual output during the time period; Indicates new energy power stations Equivalent output over a given time period;

[0029] The power correction factor is calculated as follows:

[0030] (6).

[0031] The construction of the peak-shaving ancillary service cost-sharing model between the generation side and the user side includes:

[0032] Calculate the peak-shaving cost allocation on the generation side:

[0033] (7)

[0034] In the formula, For the first The cost of peak shaving ancillary services shared by each unit that does not provide peak shaving ancillary services; This indicates the number of generating units that did not provide peak-shaving ancillary services; This represents the total cost of meeting peak-shaving ancillary service needs within a single statistical period. This indicates the proportion of peak-shaving ancillary service costs borne by the power generation side; For the first The on-grid electricity consumption of units that do not provide peak-shaving auxiliary services;

[0035] Calculate the user-side peak-shaving market compensation cost:

[0036] (8)

[0037] In the formula, This represents the total number of users participating in the peak-shaving cost sharing. For the first Metrics for individual users For the first Peak shaving ancillary service fees shared by each user.

[0038] The construction of the backup ancillary service cost-sharing model between the generation side and the user side further includes:

[0039] Calculate the cost allocation for backup ancillary services on the generation side:

[0040] (9)

[0041] In the formula, For the first The cost of backup ancillary services shared by each flight unit that does not provide backup ancillary services; This indicates the number of units that did not provide backup ancillary services; This represents the total cost of meeting the standby ancillary service requirements within the billing cycle. This indicates the proportion of the cost of backup ancillary services borne by the power generation side; For the first The power consumption of units that do not provide backup auxiliary services;

[0042] Calculate the cost sharing of user-side backup ancillary services:

[0043] (10)

[0044] In the formula, For the first The cost of backup ancillary services is shared by each user.

[0045] The construction of a feature function based on prediction error, and the quantization of the excess function based on the feature function, further includes:

[0046] Calculate the normalized root mean square error of each new energy power station, and then calculate the characteristic value of each alliance; The evaluation metric for each new energy power station is the normalized root mean square error. :

[0047] (15)

[0048] In the formula, Indicates a range of values; Indicates the number of time periods in a day; For the first A new energy station Actual power during the time period; For the first A new energy station Predicted power for the time period;

[0049] Determine each participant, i.e., the set of new energy power stations. The error assessment index for each new energy power station is normalized to the root mean square error. Defined as each alliance, i.e., new energy power station eigenvalues;

[0050] definition For the alliance The maximum total benefit that all members can guarantee through internal cooperation without cooperating with other players;

[0051] Two types of core constraints are defined:

[0052] Among them, the cost-sharing scheme This represents the share received by each participant, satisfying both individual rationality and collective validity:

[0053] (16)

[0054] (17)

[0055] In the formula, For the first The shared costs of each station; Indicates user The cost required to complete the service or facility independently; This represents the total cost of joint construction and sharing by all users;

[0056] For each alliance Define an out-of-value Indicates alliance For the cost-sharing scheme Dissatisfaction level:

[0057] (18)

[0058] In the formula, Indicates alliance The excess function value; Indicates alliance The characteristic function values; Indicates alliance The total cost shared by all stations within the area.

[0059] This invention also discloses a power ancillary service provision system that takes into account the prediction error of new energy sources, utilizing the aforementioned method for providing power ancillary services that takes into account the prediction error of new energy sources, comprising:

[0060] The frequency regulation cost allocation module is used to calculate the total cost of frequency regulation ancillary services based on the clearing price and winning capacity of the winning bid for frequency regulation of thermal power units, and to construct a frequency regulation ancillary service cost allocation model for the generation side and the user side based on the total cost of frequency regulation ancillary services and the on-grid electricity of the generation side and the user side.

[0061] The peak-shaving cost sharing module is used to determine the power correction coefficient based on the waveform overlap between the actual power generation output and the equivalent power output curve of the new energy power station, and to construct a peak-shaving ancillary service cost sharing model between the power generation side and the user side based on the power correction coefficient.

[0062] The standby cost sharing module is used to construct a standby ancillary service cost sharing model for the generation side and the user side based on the on-grid electricity volume on the generation side and the user side.

[0063] The solution module is used to construct a feature function based on the prediction error based on the frequency regulation ancillary service cost sharing model, peak shaving ancillary service cost sharing model and reserve ancillary service cost sharing model, quantify the excess function according to the feature function, and solve the excess function using the nucleolus method to form a new energy ancillary service cost sharing scheme.

[0064] Accordingly, this application also discloses a terminal, including a processor and a storage medium;

[0065] The storage medium is used to store instructions;

[0066] The processor is configured to operate according to the instructions to perform the steps of the aforementioned method for providing power ancillary services taking into account new energy prediction errors.

[0067] Accordingly, this application also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the aforementioned method for providing power auxiliary services taking into account new energy prediction errors.

[0068] The beneficial effects of this invention are that, compared with the prior art, it provides a method and system for providing power ancillary services that takes into account the prediction errors of new energy sources. Based on the fairness framework of the nucleoli method, it accurately transforms the prediction errors of new energy power plants into apportioned responsibilities. This not only solves the shortcomings of the traditional "apportionment based on grid-connected electricity" method, which ignores individual differences, but also ensures the feasibility of the solution through clear quantitative indicators and constraints. The apportionment results not only achieve a reasonable allocation of ancillary service costs, but also guide new energy power plants to optimize prediction technologies, reduce the demand for system ancillary services, and contribute to the economical and efficient operation of the power system. Attached Figure Description

[0069] Figure 1 This is a flowchart of the method for providing power auxiliary services that takes into account the prediction error of new energy sources in this invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0071] The embodiments described in this application are merely some, not all, embodiments of the present invention. Based on the spirit of the present invention, other embodiments obtained by those skilled in the art without inventive effort are all within the protection scope of the present invention.

[0072] To address the shortcomings of existing technologies, this invention proposes a method and system for providing power ancillary services that takes into account the prediction error of new energy sources. The method calculates the prediction error of power generation at new energy power plants and uses it as a characteristic value for calculating the ancillary service cost allocated to new energy sources, thereby obtaining the different responsibility allocation ratios for each new energy power plant due to the error calculation.

[0073] See Figure 1 As shown, the method for providing ancillary power services that takes into account the prediction error of new energy sources disclosed in this invention includes the following steps:

[0074] Step 1: Calculate the total cost of frequency regulation ancillary services based on the clearing price and winning capacity of the winning bid for frequency regulation of thermal power units. Based on the total cost of frequency regulation ancillary services and the on-grid electricity on the generation side and the user side, construct a cost-sharing model for frequency regulation ancillary services on the generation side and the user side.

[0075] Frequency regulation ancillary service cost allocation refers to the allocation of frequency regulation service fees paid by grid operators to maintain frequency stability based on the contributions of generating units, energy storage systems, and other components to the grid's frequency regulation capabilities. Since grid frequency stability is crucial to power system security, a reasonable allocation of frequency regulation ancillary service costs ensures that all stakeholders fairly share the cost of maintaining grid stability.

[0076] The frequency regulation ancillary service fee compensation mechanism follows the principle of "whoever benefits, bears the cost." If a certain amount of revenue is obtained, a corresponding fee must be provided. One method is to use the grid connection and electricity consumption of the generation side and the user side as the allocation standard.

[0077] The specific formula for allocating frequency regulation ancillary service costs on the generation side is as follows:

[0078] (1)

[0079] In the formula, For the first The cost of frequency modulation ancillary services shared by each unit that does not provide frequency modulation ancillary services; This indicates the number of generating units that do not provide frequency regulation ancillary services; This represents the total cost of meeting the demand for frequency modulation ancillary services within the billing cycle; This indicates the proportion of frequency regulation ancillary service costs borne by the power generation side; this value is predetermined by market rules. For the first The amount of electricity connected to the grid by units that do not provide frequency modulation auxiliary services.

[0080] Based on this, the present invention further sets supply and demand balance constraints for frequency regulation, and the overall purchase cost of frequency regulation services in the ancillary service market area (or the total frequency regulation revenue of winning frequency regulation units in the area) is as follows:

[0081] (2)

[0082] In the formula, This indicates the total number of periods for daily spot clearing; Indicates the number of thermal power units; This indicates the clearing price of the winning bid for frequency regulation in thermal power units; For the first Taiwan thermal power units The winning bid capacity for frequency modulation at any given time.

[0083] The specific formula for the user-side cost sharing of FM ancillary services is as follows:

[0084] (3)

[0085] In the formula, For the first The cost of frequency modulation ancillary services shared by each user; This represents the total number of users participating in the frequency modulation cost sharing; For the first Metrics for each user; This represents the sum of metrics for all users who participated in the cost-sharing.

[0086] Step 2: Determine the power correction coefficient based on the waveform overlap between the actual power output and the equivalent power output curve of the new energy power station, and construct a peak-shaving ancillary service cost sharing model between the power generation side and the user side based on the power correction coefficient.

[0087] The cost of peak shaving ancillary services on the power generation side should be shared by new energy power plants such as wind and solar power plants and thermal power units that do not provide peak shaving ancillary services. The standard for sharing the cost should be based on their respective on-grid electricity volume in the power system.

[0088] The additional peak-shaving ancillary services resulting from the grid connection of renewable energy generation are due to the inability of the grid-connected renewable energy capacity to meet the increased load demand and the curtailment of wind and solar power. Therefore, renewable energy power plants are responsible for the cost of these additional peak-shaving ancillary services and bear the corresponding expenses. Because renewable energy power plants have different capacity reliability (reflecting the contribution of renewable energy plant grid connection to system adequacy, i.e., the size of the load that the renewable energy plant can handle), their shared ancillary service costs differ. For example, renewable energy power plants with lower reliability contribute less to the power system and must bear more peak-shaving ancillary service costs, thus highlighting fairness.

[0089] Because renewable energy power plants do not possess peak-shaving capabilities, the standard for allocating peak-shaving ancillary service costs for renewable energy power plants differs from the mechanism for thermal power units based on their on-grid electricity volume. When the output curve of a renewable energy power plant changes accordingly with the load curve, the power output of the renewable energy power plant changes accordingly. This corresponding change pattern can alleviate the pressure on the system. By using the equal power volume-load-following method, the output curve of the renewable energy power plant is transformed into an equivalent output curve of renewable energy with the same load fluctuation as the power grid. By comparing the waveform overlap between the actual power generation output of the renewable energy power plant and the equivalent output curve, the impact of the renewable energy power plant's output on the peak-shaving pressure of the power system can be determined, and the cost of peak-shaving ancillary services for renewable energy power plants can be allocated using the similarity factor standard.

[0090] The higher the overlap between the actual output curve and the equivalent output curve of a renewable energy power station, the higher their corresponding similarity. The larger the value, the better it matches the changes in the power system load, and the lower the difficulty of peak shaving for the power system.

[0091] (4)

[0092] In the formula, Indicates new energy power stations Actual output during the time period; Indicates new energy power stations Equivalent output over a given time period.

[0093] Through the revised power generation standards for new energy power plants The cost of peak-shaving ancillary services will be shared:

[0094] (5)

[0095] In the formula, This represents the power correction factor; Indicates new energy power station exist Actual power generation during the period.

[0096] , The smaller the value, the more difficult it is for the system to regulate peak loads. The corresponding power correction factor is:

[0097] (6)

[0098] The formula for calculating the peak-shaving cost sharing on the power generation side is as follows:

[0099] (7)

[0100] In the formula, For the first The cost of peak shaving ancillary services shared by each unit that does not provide peak shaving ancillary services; This indicates the number of generating units that did not provide peak-shaving ancillary services; This represents the total cost of meeting peak-shaving ancillary service needs within a single statistical period. This indicates the proportion of peak-shaving ancillary service costs borne by the power generation side; this value is predetermined by market rules. For the first The electricity generated by generating units that do not provide peak-shaving auxiliary services will be charged according to the revised electricity standard. To conduct statistics.

[0101] The compensation fee for peak shaving market on the user side is:

[0102] (8)

[0103] In the formula, This represents the total number of users participating in the peak-shaving cost sharing. For the first The metering metrics for individual users adopt the revised electricity consumption standard. Perform statistics; For the first Peak shaving ancillary service fees shared by each user.

[0104] Step 3: Construct a cost-sharing model for backup ancillary services on the generation and user sides based on the grid-connected electricity volume on the generation and user sides.

[0105] The specific formula for allocating the cost of backup ancillary services on the generation side is as follows:

[0106] (9)

[0107] In the formula, For the first The cost of backup ancillary services shared by each flight unit that does not provide backup ancillary services; This indicates the number of units that did not provide backup ancillary services; This represents the total cost of meeting the standby ancillary service requirements within the billing cycle. This indicates the proportion of the cost of backup ancillary services borne by the power generation side; this value is predetermined by market rules. For the first The amount of electricity connected to the grid by units that do not provide backup auxiliary services.

[0108] The specific formula for cost sharing of user-side backup ancillary services is as follows:

[0109] (10)

[0110] In the formula, For the first The cost of backup ancillary services is shared by each user.

[0111] Step 4: Based on the frequency regulation ancillary service cost sharing model, peak shaving ancillary service cost sharing model, and reserve ancillary service cost sharing model, construct a feature function based on prediction error, quantify the excess function according to the feature function, and solve the excess function using the nucleolus method to form a new energy ancillary service cost sharing scheme.

[0112] Nucleolus is a method for solving the problem of benefit distribution in cooperative games, especially in multi-player cooperative games, where it can provide a relatively fair and stable distribution scheme. The core idea of ​​Nucleolus is to find a reasonable distribution point by minimizing the "dissatisfaction" of all participants.

[0113] The mathematical formula for the nucleolus method is as follows, assuming there is a cooperative game. , among which is Participant group It is a characteristic function, representing any subset. The goal of the nucleolus method is to find an assignment vector. So that each participant Allocation The following conditions must be met:

[0114] 1) Individual rationality: Each participant's allocation is at least equal to their payoff if they acted alone, i.e.

[0115] (11)

[0116] 2) Collective rationality: The total distribution among all participants equals the total benefit of the entire cooperation, i.e.

[0117] (12)

[0118] 3) Minimize maximum dissatisfaction: The nucleolus method finds a reasonable allocation point by minimizing the "dissatisfaction" of all participants. (Definition of the first...) The dissatisfaction level of each participant is:

[0119] (13)

[0120] The goal of the nucleolus method is to find an assignment vector. To minimize the "dissatisfaction" of all participants, that is...

[0121] (14)

[0122] The nucleolar method has a unique solution, meaning that in a given cooperative game, it can find only one unique allocation vector. The nucleolar method attempts to achieve a relatively fair allocation among all participants by minimizing the "dissatisfaction" of all players. The solution of the nucleolar method is generally considered stable because it minimizes the maximum value of the excess function of all subsets, thus reducing the risk of cooperation breakdown.

[0123] The specific solution steps are as follows:

[0124] Step 4.1 Define the cooperative game theory system

[0125] (1) Define all new energy power stations participating in the sharing of ancillary service costs as the set of game participants. (Where n is the total number of new energy power stations). Each new energy power station, as a cooperative entity, shares the incremental costs of ancillary services caused by the uncertainty of its own power output, and there is no individual decision to leave the alliance.

[0126] The core objective of the game is to determine the cost-sharing ratio of each station while satisfying the dual requirements of "full coverage of total ancillary service costs" and "matching individual cost-sharing responsibilities with actual impacts," thereby minimizing the dissatisfaction of any single station or station alliance and achieving a balance between fairness and economy.

[0127] (2) Key premise assumptions:

[0128] The power generation prediction error of all new energy power plants can be quantified by a unified index, and the error data is real and traceable (calculated based on historical actual output and predicted output).

[0129] The incremental cost of ancillary services has been clearly defined (i.e., the total additional costs incurred in steps 1-3 due to the grid connection of new energy sources, such as frequency regulation, peak shaving, and backup).

[0130] Step 4.2 Constructing Feature Functions Based on Prediction Error

[0131] The characteristic function is the core input of the nucleolus method, used to quantify the responsibility weight of each participant or alliance for "ancillary service cost generation". This invention constructs the characteristic function based on the power generation prediction error of new energy power plants:

[0132] (1) Quantitative index of prediction error: Normalized root mean square error (NRMSE)

[0133] Based on the historical actual data of the new energy power stations, the normalized root mean square error of each new energy power station is calculated using formula (15), and the characteristic value of each alliance is calculated. The evaluation metric for each new energy power station is the normalized root mean square error. :

[0134] (15)

[0135] In the formula, Indicates a range of values; Indicates the number of time periods in a day; For the first A new energy station Actual power during the time period; For the first A new energy station Predicted power for a given time period.

[0136] (2) Define the characteristic function

[0137] Determine each participant, i.e., the set of new energy power stations. Then, the error evaluation index for each new energy power station is normalized to the root mean square error. Defined as each alliance, i.e., new energy power station eigenvalues ​​( It can represent a single station, a combination of multiple stations, or all stations), normalized root mean square error. It measures the error between the predicted and actual values ​​of power generation at new energy power plants.

[0138] Where the definition , indicating the alliance The maximum total benefit (or minimum total cost) that all members can guarantee through internal cooperation without cooperating with other players (i.e., players in N / S). It is a special Among them, the alliance It is a collection of all players It represents the maximum total revenue that the entire system can generate, or the minimum total cost that it must bear, when all players cooperate (forming a "grand alliance").

[0139] Step 4.3 Set constraints for the allocation scheme

[0140] To ensure the feasibility and rationality of the cost-sharing scheme, the individual and collective rationality requirements of the nuclear law must be met, and two types of core constraints are set:

[0141] Among them, the cost-sharing scheme This represents the share of the value obtained by each participant, and must satisfy the conditions of equations (16) and (17) to satisfy individual rationality and collective effectiveness.

[0142] (16)

[0143] (17)

[0144] In the formula, For the first The shared costs of each station; Indicates user The cost required to complete the service or facility independently; This represents the total cost of joint construction and sharing by all users.

[0145] Step 4.4 Quantifying the excess function and dissatisfaction

[0146] The nucleolus method quantifies the dissatisfaction of each alliance with the cost-sharing plan through an excess function. The higher the dissatisfaction, the more serious the alliance believes that its share of costs is mismatched with its responsibilities.

[0147] (1) Definition of excess function

[0148] For each alliance Its excess value , indicating the alliance For the cost-sharing scheme Dissatisfaction level:

[0149] (18)

[0150] In the formula, Indicates alliance The excess function value (dimensionless; a positive value indicates that the alliance believes the shared cost is lower than its due responsibility, indicating high dissatisfaction; a negative value indicates that the shared cost is higher than its responsibility, indicating low dissatisfaction). Indicates alliance The characteristic function value (total responsibility weight); Indicates alliance The total cost shared by all stations within the area.

[0151] (2) Calculation of excess credits for the entire league

[0152] Iterate through all possible station alliances (including single stations, pairwise combinations, three-by-three combinations, and so on up to the entire alliance), calculate the excess function value for each alliance, and form a complete insufficiency matrix (excluding the empty set).

[0153] Step 4.5 Nucleolus Solving and Proportion Determination

[0154] The goal of the nucleolus method is to find an allocation scheme that minimizes the maximum excess function value of all coalitions. This scheme is called the nucleolus solution, which can be solved using a linear programming model.

[0155] The goal of the nucleolus method is to find an assignment vector. This minimizes the maximum value of the excess function for all subsets, i.e.

[0156] (19)

[0157] (1) Construction of linear programming model

[0158] Introducing auxiliary variables (Representing the maximum excess function value across all alliances), construct the following optimization model:

[0159] (20)

[0160] (twenty one)

[0161] (twenty two)

[0162] In addition, equation (17) must also be satisfied.

[0163] (2) Calculation of the optimal allocation ratio

[0164] The optimal cost allocation for each station is obtained by solving the above model using a linear programming algorithm. Further calculate the sharing ratio (That is, the proportion of the cost allocated to a single station to the total cost):

[0165] (twenty three)

[0166] The final allocation ratio Error with station prediction There is a positive correlation: the larger the prediction error of the power plant, the higher the cost-sharing ratio, which is in line with the fair principle of "whoever causes the demand for ancillary services should bear the corresponding costs". At the same time, economic incentives force new energy power plants to improve the accuracy of power generation prediction.

[0167] The beneficial effects of this invention are that, compared with the prior art, it provides a method and system for providing power ancillary services that takes into account the prediction errors of new energy sources. Based on the fairness framework of the nucleoli method, it accurately transforms the prediction errors of new energy power plants into apportioned responsibilities. This not only solves the shortcomings of the traditional "apportionment based on grid-connected electricity" method, which ignores individual differences, but also ensures the feasibility of the solution through clear quantitative indicators and constraints. The apportionment results not only achieve a reasonable allocation of ancillary service costs, but also guide new energy power plants to optimize prediction technologies, reduce the demand for system ancillary services, and contribute to the economical and efficient operation of the power system.

[0168] This invention can be a system, method, and / or computer program product. This invention also discloses a power ancillary service provision system that takes into account renewable energy forecasting errors, based on the aforementioned method for providing power ancillary services taking into account renewable energy forecasting errors, comprising:

[0169] The frequency regulation cost allocation module is used to calculate the total cost of frequency regulation ancillary services based on the clearing price and winning capacity of the winning bid for frequency regulation of thermal power units, and to construct a frequency regulation ancillary service cost allocation model for the generation side and the user side based on the total cost of frequency regulation ancillary services and the on-grid electricity of the generation side and the user side.

[0170] The peak-shaving cost sharing module is used to determine the power correction coefficient based on the waveform overlap between the actual power generation output and the equivalent power output curve of the new energy power station, and to construct a peak-shaving ancillary service cost sharing model between the power generation side and the user side based on the power correction coefficient.

[0171] The standby cost sharing module is used to construct a standby ancillary service cost sharing model for the generation side and the user side based on the on-grid electricity volume on the generation side and the user side.

[0172] The solution module is used to construct a feature function based on the prediction error based on the frequency regulation ancillary service cost sharing model, peak shaving ancillary service cost sharing model and reserve ancillary service cost sharing model, quantify the excess function according to the feature function, and solve the excess function using the nucleolus method to form a new energy ancillary service cost sharing scheme.

[0173] Based on the spirit of this invention, those skilled in the art will readily conceive that a computer program product can be obtained based on the aforementioned method for providing auxiliary power services taking into account new energy prediction errors. The computer program product may include a computer-readable storage medium on which computer-readable program instructions are loaded to cause a processor to implement various aspects of this disclosure. That is, this application also includes a terminal comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to perform the steps of the aforementioned method for providing auxiliary power services taking into account new energy prediction errors.

[0174] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0175] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0176] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0177] 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 it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for providing ancillary electricity services taking into account the prediction error of new energy sources, characterized in that, Includes the following steps: Step 1: Calculate the total cost of frequency regulation ancillary services based on the clearing price and winning capacity of the winning bid for frequency regulation of thermal power units. Based on the total cost of frequency regulation ancillary services and the on-grid electricity on the generation side and the user side, construct a cost-sharing model for frequency regulation ancillary services on the generation side and the user side. Step 2: Determine the power correction coefficient based on the waveform overlap between the actual power output and the equivalent power output curve of the new energy power station, and construct a peak-shaving ancillary service cost sharing model for the power generation side and the user side based on the power correction coefficient. Step 3: Construct a cost-sharing model for backup ancillary services on the generation and user sides based on the grid-connected electricity volume on the generation and user sides; Step 4: Based on the frequency regulation ancillary service cost sharing model, peak shaving ancillary service cost sharing model and reserve ancillary service cost sharing model, construct a feature function based on prediction error, quantify the excess function according to the feature function, and solve the excess function using the nucleolus method to form a new energy ancillary service cost sharing scheme. The construction of a feature function based on prediction error, and the quantization of the excess function based on the feature function, further includes: Calculate the normalized root mean square error of each new energy power station, and then calculate the characteristic value of each alliance; The evaluation metric for each new energy power station is the normalized root mean square error. : （15） In the formula, Indicates a range of values; Indicates the number of time periods in a day; For the first A new energy station Actual power during the time period; For the first A new energy station Predicted power for the time period; Determine each participant, i.e., the set of new energy power stations. The error assessment index for each new energy power station is normalized to the root mean square error. Defined as each alliance, i.e., new energy power station eigenvalues; definition For the alliance The maximum total benefit that all members can guarantee through internal cooperation without cooperating with other players; Two types of core constraints are defined: Among them, the cost-sharing scheme This represents the share received by each participant, satisfying both individual rationality and collective validity: （16） （17） In the formula, For the first The shared costs of each station; Indicates user The cost required to complete the service or facility independently; This represents the total cost of joint construction and sharing by all users; For each alliance Define an out-of-value Indicates alliance For the cost-sharing scheme Dissatisfaction level: （18） In the formula, Indicates alliance The excess function value; Indicates alliance The characteristic function values; Indicates alliance The total cost shared by all stations within the area; The step of solving the excess function using the nucleolus method to form a cost-sharing scheme for new energy auxiliary services further includes: The objective of finding an allocation scheme that minimizes the maximum excess function value for all alliances can be solved using a linear programming model: The goal of the nucleolus method is to find an assignment vector. This minimizes the maximum value of the excess function for all subsets, i.e. （19） Introducing auxiliary variables To represent the maximum excess function value across all alliances, the following optimization model is constructed: （20） （21） （22） The optimal cost allocation for each station is obtained by solving the above model using a linear programming algorithm. Further calculate the sharing ratio That is, the proportion of the cost allocated to a single station to the total cost: （23） The final allocation ratio Error with station prediction They show a positive correlation.

2. The method for providing ancillary power services taking into account the prediction error of new energy sources according to claim 1, characterized in that, The construction of the frequency regulation ancillary service cost-sharing model between the generation side and the user side further includes: Calculate the generation-side apportionment of frequency regulation ancillary service costs: （1） In the formula, For the first The cost of frequency modulation ancillary services shared by each unit that does not provide frequency modulation ancillary services; This indicates the number of generating units that do not provide frequency regulation ancillary services; This represents the total cost of meeting the demand for frequency modulation ancillary services within the billing cycle; This indicates the proportion of frequency regulation ancillary service costs borne by the power generation side; For the first The power consumption of units that do not provide frequency modulation auxiliary services; Set supply and demand balance constraints for frequency modulation: （2） In the formula, This indicates the total number of periods for daily spot clearing; Indicates the number of thermal power units; This indicates the clearing price of the winning bid for frequency regulation in thermal power units; For the first Taiwan thermal power units The winning bid capacity for frequency modulation at any given time.

3. The method for providing power ancillary services taking into account the prediction error of new energy sources according to claim 2, characterized in that, The construction of the frequency regulation ancillary service cost-sharing model between the generation side and the user side further includes: Calculate the user-side apportionment of FM ancillary service costs: （3） In the formula, For the first The cost of frequency modulation ancillary services shared by each user; This represents the total number of users participating in the frequency modulation cost sharing. For the first Metrics for each user; This represents the sum of metrics for all users who participated in the cost-sharing.

4. The method for providing ancillary power services taking into account the prediction error of new energy sources according to claim 3, characterized in that, The determination of the power correction coefficient based on the waveform overlap between the actual power output and the equivalent power output curve of the new energy power station includes: Calculate the similarity between the waveform overlap of the actual power output curve and the equivalent power output curve of a renewable energy power station. （4） In the formula, Indicates new energy power stations Actual output during the time period; Indicates new energy power stations Equivalent output over a given time period; The power correction factor is calculated as follows: （6）。 5. The method for providing ancillary power services taking into account the prediction error of new energy sources according to claim 4, characterized in that, The construction of the peak-shaving ancillary service cost-sharing model between the generation side and the user side includes: Calculate the peak-shaving cost allocation on the generation side: （7） In the formula, For the first The cost of peak shaving ancillary services shared by each unit that does not provide peak shaving ancillary services; This indicates the number of generating units that did not provide peak-shaving ancillary services; This represents the total cost of meeting peak-shaving ancillary service needs within a single statistical period. This indicates the proportion of peak-shaving ancillary service costs borne by the power generation side; For the first The on-grid power consumption of units that do not provide peak-shaving auxiliary services; Calculate the user-side peak-shaving market compensation cost: （8） In the formula, This represents the total number of users participating in the peak-shaving cost sharing. For the first Metrics for individual users For the first Peak shaving ancillary service fees shared by each user.

6. The method for providing ancillary power services taking into account the prediction error of new energy sources according to claim 5, characterized in that, The construction of the backup ancillary service cost-sharing model between the generation side and the user side further includes: Calculate the cost allocation for backup ancillary services on the generation side: （9） In the formula, For the first The cost of backup ancillary services shared by each flight unit that does not provide backup ancillary services; This indicates the number of units that did not provide backup ancillary services; This represents the total cost of meeting the standby ancillary service requirements within the billing cycle. This indicates the proportion of the cost of backup ancillary services borne by the power generation side; For the first The power consumption of units that do not provide backup auxiliary services; Calculate the cost sharing of user-side backup ancillary services: （10） In the formula, For the first The cost of backup ancillary services is shared by each user.

7. A power auxiliary service provision system that takes into account the prediction error of new energy sources, characterized in that, include: The frequency regulation cost allocation module is used to calculate the total cost of frequency regulation ancillary services based on the clearing price and winning capacity of the winning bid for frequency regulation of thermal power units, and to construct a frequency regulation ancillary service cost allocation model for the generation side and the user side based on the total cost of frequency regulation ancillary services and the on-grid electricity of the generation side and the user side. The peak-shaving cost sharing module is used to determine the power correction coefficient based on the waveform overlap between the actual power generation output and the equivalent power output curve of the new energy power station, and to construct a peak-shaving ancillary service cost sharing model between the power generation side and the user side based on the power correction coefficient. The standby cost sharing module is used to construct a standby ancillary service cost sharing model for the generation side and the user side based on the on-grid electricity volume on the generation side and the user side. The solution module is used to construct a feature function based on the prediction error based on the frequency regulation ancillary service cost sharing model, peak shaving ancillary service cost sharing model and reserve ancillary service cost sharing model, quantify the excess function according to the feature function, and solve the excess function using the nucleolus method to form a new energy ancillary service cost sharing scheme. The solution module is further used for: Calculate the normalized root mean square error of each new energy power station, and then calculate the characteristic value of each alliance; The evaluation metric for each new energy power station is the normalized root mean square error. : （15） In the formula, Indicates a range of values; Indicates the number of time periods in a day; For the first A new energy station Actual power during the time period; For the first A new energy station Predicted power for the time period; Determine each participant, i.e., the set of new energy power stations. The error assessment index for each new energy power station is normalized to the root mean square error. Defined as each alliance, i.e., new energy power station eigenvalues; definition For the alliance The maximum total benefit that all members can guarantee through internal cooperation without cooperating with other players; Two types of core constraints are defined: Among them, the cost-sharing scheme This represents the share received by each participant, satisfying both individual rationality and collective validity: （16） （17） In the formula, For the first The shared costs of each station; Indicates user The cost required to complete the service or facility independently; This represents the total cost of joint construction and sharing by all users; For each alliance Define an out-of-value Indicates alliance For the cost-sharing scheme Dissatisfaction level: （18） In the formula, Indicates alliance The excess function value; Indicates alliance The characteristic function values; Indicates alliance The total cost shared by all stations within the area; The step of solving the excess function using the nucleolus method to form a cost-sharing scheme for new energy auxiliary services further includes: The objective of finding an allocation scheme that minimizes the maximum excess function value for all alliances can be solved using a linear programming model: The goal of the nucleolus method is to find an assignment vector. This minimizes the maximum value of the excess function for all subsets, i.e. （19） Introducing auxiliary variables To represent the maximum excess function value across all alliances, the following optimization model is constructed: （20） （21） （22） The optimal cost allocation for each station is obtained by solving the above model using a linear programming algorithm. Further calculate the sharing ratio That is, the proportion of the cost allocated to a single station to the total cost: （23） The final allocation ratio Error with station prediction They show a positive correlation.

8. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method for providing power ancillary services taking into account the error of new energy prediction as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method for providing power auxiliary services taking into account the prediction error of new energy sources as described in any one of claims 1-6.

Images