A wind power simulation data electricity price generation method and system based on scene mapping

Abstract

The application relates to a wind power simulation data price generation method and system based on scene mapping, wherein the method comprises the following steps: obtaining historical data and simulation data, and recombining the historical data and the simulation data to obtain historical day data and simulation day data; clustering the historical day data to divide the historical day data into typical price scenes; clustering the historical day data under different typical price scenes to divide the historical day data into typical wind power scenes, and constructing a mapping relationship between the typical price scenes and the typical wind power scenes; calculating a joint probability based on a prior probability of the typical price scenes and a conditional probability of the typical wind power scenes; obtaining a posterior probability based on the simulation day data and the joint probability; performing weighted calculation according to the posterior probability of each typical price scene to obtain an original simulation expected price; and correcting the original simulation expected price based on the historical day data and the simulation day data to obtain a simulation expected price.

Description

Technical Field

[0001] This invention relates to the field of power system simulation and data processing technology, and mainly to a method and system for generating electricity prices from wind power simulation data based on scene mapping. Background Technology

[0002] In power system simulation and operation simulation analysis, wind power simulation software (such as WASP and WindSim) can generate 8760 hours of simulated power output data. Existing simulation data processing methods usually only analyze and process the power data itself, such as power prediction, volatility analysis, and probability distribution fitting. However, in actual power system operation, electricity price is a physical parameter of equal importance to power, and there is a complex temporal coupling relationship between the two. Fluctuations in wind power output will affect the real-time supply and demand balance of the system, and thus affect the formation of electricity prices. If the simulation data lacks time-aligned electricity price data, the simulation dataset is incomplete and cannot support subsequent application scenarios such as power system operation simulation and grid dispatch analysis.

[0003] Chinese invention patent application CN110826777A discloses a method, apparatus, equipment, and medium for analyzing transaction data in a wind power bidding market. The technical solution includes: acquiring historical raw market data; establishing a vector based on the historical raw market data; establishing a minimax difference method to calculate the historical raw data using this method, obtaining a calculated vector; acquiring a market prediction vector; comparing the predicted vector with the calculated vector; and using the predicted vector as the predicted value when it meets the range of the calculated vector. Finally, establishing a neural network algorithm to calculate the predicted value and obtaining the calculation result as the predicted electricity price. This invention acquires historical market data and uses a minimax difference algorithm from the similarity day method to predict the required data. After obtaining the required data, a neural network algorithm is built to calculate the predicted electricity price, thus assisting in bidding decisions. However, the above-mentioned technical solution uses the minimax difference method to define a fixed range and simply judges whether the predicted vector falls within the interval to obtain the predicted value. It can only output a single, definite electricity price prediction value and cannot quantify the probability of various electricity price fluctuations, making it difficult to characterize the random fluctuation distribution characteristics of electricity prices. At the same time, the neural network in the above-mentioned technical solution directly outputs the electricity price result based on the filtered predicted value, lacking a correction step for the electricity price prediction result in combination with the market bidding space. The predicted electricity price is not optimized in combination with the actual bidding boundary conditions, and the prediction result deviates from the real market bidding environment, making it difficult to directly support bidding decisions. In addition, the above-mentioned technical solution relies solely on raw market data to directly construct vectors for analysis, lacking refined means of dividing historical data by operating conditions and scenarios. Data from different operating conditions are mixed together, which can easily cause data patterns to interfere with each other and reduce the accuracy of subsequent prediction models.

[0004] Therefore, there is an urgent need for a simulation data electricity price generation method that can make full use of historical electricity prices, wind power output and bidding space data, as well as adaptive correction of bidding space. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a method and system for generating electricity prices from wind power simulation data based on scene mapping.

[0006] The technical solution of the present invention is as follows: On one hand, this invention proposes a method for generating electricity prices from wind power simulation data based on scene mapping, the method comprising: Acquire historical electricity price data, historical wind power data, historical bidding space data, simulated wind power data, and simulated bidding space data; reorganize the acquired data to obtain historical daily electricity price data, historical daily wind power data, historical daily bidding space data, simulated daily wind power data, and simulated daily bidding space data. Cluster historical daily electricity price data and classify typical electricity price scenarios based on the clustering results; calculate the prior probability of each typical electricity price scenario based on historical daily electricity price data; Historical daily wind power data under different typical electricity price scenarios are clustered, and typical wind power scenarios are divided according to the clustering results. A mapping relationship between typical electricity price scenarios and typical wind power scenarios is constructed. The conditional probability of typical wind power scenarios is calculated based on historical daily wind power data. The joint probability of typical electricity price scenarios and typical wind power scenarios is calculated based on prior probability and conditional probability. Based on simulated daily wind power data and joint probability, the posterior probability of simulated daily wind power data belonging to different typical electricity price scenarios is obtained; the original simulated expected electricity price is obtained by weighted calculation based on the posterior probability of each typical electricity price scenario. The original simulated expected electricity price is corrected based on historical bidding space daily data and simulated bidding space daily data to obtain the simulated expected electricity price.

[0007] Preferably, the acquired data is reorganized, and the specific steps are as follows: The acquired data is segmented according to a 24-hour calendar day to obtain the corresponding daily data.

[0008] Preferably, historical daily electricity price data is clustered, specifically as follows: The elbow rule is used to determine the optimal number of clusters for historical daily electricity price data. ; The K-means algorithm was used to cluster historical daily electricity price data to obtain... A typical electricity pricing scenario; Based on the The ratio of the number of historical daily electricity price data points for a typical electricity price scenario to the total number of historical daily electricity price data points is used to calculate the first... Prior probabilities for a typical electricity price scenario.

[0009] Preferably, historical daily wind power data under different typical electricity price scenarios are clustered, specifically as follows: Extract cluster labels from typical electricity pricing scenarios; Based on the clustering label mapping to historical wind power daily data, extract the corresponding historical wind power data subset for each typical electricity price scenario in the historical wind power daily data; The elbow rule is used to determine the optimal number of wind power clusters for a subset of historical wind power data. ; The K-means algorithm was used to cluster a subset of historical wind power data to obtain... A typical wind power scenario; Based on the The subset of historical wind power data belongs to the first The number of historical daily wind power data for a typical wind power scenario and the number of... The ratio of the total number of historical wind power daily data points in the first subset of historical wind power data is used to calculate the first... Under a typical electricity pricing scenario, the first Conditional probabilities for a typical wind power scenario.

[0010] Preferably, the mapping relationship between the typical electricity price scenario and the typical wind power scenario is as follows: each typical electricity price scenario corresponds to one or more typical wind power scenarios, and each typical wind power scenario belongs to only one typical electricity price scenario.

[0011] Preferably, the posterior probability of simulated daily wind power data belonging to different typical electricity price scenarios is obtained based on simulated daily wind power data and joint probability. The specific steps are as follows: The Euclidean distance between simulated daily wind power data and typical wind power scenarios is calculated, and the Euclidean distance is converted into corresponding matching weights using a Gaussian kernel function. Sort all matching weights in descending order and select the top [weights]. Each matching weight is used as the scene matching weight; Based on scenario matching weights and joint probability calculations, the posterior probability of simulated daily wind power data belonging to a typical electricity price scenario is obtained.

[0012] Preferably, the original simulated expected electricity price is corrected based on historical bidding space daily data and simulated bidding space daily data. The specific steps are as follows: The simulated daily bidding space data is input into the trained multilayer perceptron regression neural network to obtain the predicted daily price. Calculate the historical bid center for daily historical bid space data; Calculate the Mahalanobis distance between the simulated daily bidding space data and the historical bidding center; The adaptive correction weights are calculated based on the Mahalanobis distance. The daily average of the simulated expected electricity price is calculated based on the daily average of the adaptively adjusted weights, the predicted daily electricity price, and the original simulated expected electricity price. The original simulated expected electricity price is corrected based on the daily average value of the simulated expected electricity price to obtain the simulated expected electricity price.

[0013] Preferably, the adaptive correction weights are calculated based on the Mahalanobis distance, specifically as follows: Set the Mahalanobis distance threshold; If the Mahalanobis distance is less than or equal to the Mahalanobis distance threshold, the Mahalanobis distance is input into an exponential function with a decay coefficient to obtain the decayed Mahalanobis distance; the adaptive correction weight is obtained based on the product of the decayed Mahalanobis distance and the preset maximum correction coefficient. Otherwise, the adaptive adjustment weight is 0.

[0014] On the other hand, the present invention also provides a wind power simulation data electricity price generation system based on scene mapping, the system comprising: The data acquisition module acquires historical electricity price data, historical wind power data, historical bidding space data, simulated wind power data, and simulated bidding space data; it then reorganizes the acquired data to obtain daily historical electricity price data, daily historical wind power data, daily historical bidding space data, daily simulated wind power data, and daily simulated bidding space data. The probability calculation module clusters historical daily electricity price data and classifies typical electricity price scenarios based on the clustering results; it calculates the prior probability of each typical electricity price scenario based on the historical daily electricity price data; it clusters historical daily wind power data under different typical electricity price scenarios, classifies typical wind power scenarios based on the clustering results, and constructs a mapping relationship between typical electricity price scenarios and typical wind power scenarios; it calculates the conditional probability of typical wind power scenarios based on historical daily wind power data; and it calculates the joint probability of typical electricity price scenarios and typical wind power scenarios based on the prior probability and conditional probability. The electricity price calculation module obtains the posterior probability of the simulated wind power daily data belonging to different typical electricity price scenarios based on the simulated wind power daily data and joint probability; and performs weighted calculation based on the posterior probability of each typical electricity price scenario to obtain the original simulated expected electricity price. The electricity price correction module corrects the original simulated expected electricity price based on historical bidding space daily data and simulated bidding space daily data to obtain the simulated expected electricity price.

[0015] Furthermore, this invention also provides a method for optimizing the scheduling of wind-storage power stations based on the expected electricity price simulated in wind power simulation, the method comprising: Acquire simulation data, including simulated expected electricity price and simulated wind power output data; A wind power plant scheduling optimization model is constructed based on simulation data. The wind power plant scheduling optimization model includes an optimization objective function constructed with the goal of maximizing the total revenue of the wind power plant within the scheduling period, and constraint conditions constructed with preset constraint conditions. Solve the wind-storage power station scheduling optimization model to obtain the charging and discharging plan of the energy storage system and the power output allocation scheme of the wind turbine in each time period; The wind power station dispatch and operation instructions are generated based on the charging and discharging plans of the energy storage system and the power output allocation scheme of the wind turbines for each time period.

[0016] The present invention has the following beneficial effects: 1. This invention constructs a mapping relationship between typical electricity price scenarios and typical wind power scenarios through two-layer clustering, and explicitly models the coupling pattern between electricity price and wind power in historical data. This enables simulated daily wind power data to be mapped back to the electricity price space through the mapping relationship, realizing the probabilistic reconstruction from wind power data to electricity price data. This achieves the purpose of distinguishing different operating conditions in the power market, improving the precision of operating condition identification, enhancing the effectiveness of data modeling, and strengthening the model's ability to capture differentiated electricity price patterns. 2. This invention utilizes the principle of joint probability calculation combined with the prior probabilities of two types of scenarios to solve the joint probability of scenarios, thereby achieving the purpose of quantifying the correlation between wind power operating conditions and typical electricity price scenarios, improving the accuracy of scenario coupling relationship representation, enhancing the modeling level of wind power and electricity price correlation, and strengthening the characterization effect of multivariate coupling laws. 3. This invention uses the neural network prediction results of the bidding space to adaptively correct the original electricity price, and dynamically adjusts the correction weight based on Mahalanobis distance to match the corrected electricity price with the supply and demand boundary conditions of the simulation scenario, thereby improving the effectiveness of the simulation electricity price calculation results and enhancing the practical value of the simulation electricity price data in supporting power simulation and analysis. Attached Figure Description

[0017] Figure 1 This is a detailed flowchart of an embodiment of the present invention; Figure 2 This is a standardized data diagram of an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the clustering effect of historical electricity price daily data in an embodiment of the present invention; Figure 4 This is a schematic diagram of the clustering effect of a typical wind power scenario under the first typical electricity price scenario of the present invention. Figure 5 This is a schematic diagram of the clustering effect of a typical wind power scenario under the second typical electricity price scenario of the present invention; Figure 6 This is a schematic diagram of the clustering effect of a typical wind power scenario under the third typical electricity price scenario of the present invention; Figure 7This is a multilayer perceptron regressive neural network according to an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0020] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0022] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0023] Example 1: See Figure 1 This invention provides a method for generating electricity prices from wind power simulation data based on scene mapping, the method comprising: S1. Obtain historical electricity price data, historical wind power data, historical bidding space data, simulated wind power data, and simulated bidding space data; All the acquired data are time-aligned data sequences; S2. Reorganize the acquired data to obtain historical electricity price daily data, historical wind power daily data, historical bidding space daily data, simulated wind power daily data, and simulated bidding space daily data; The reorganization refers to segmenting the standardized data according to a 24-hour calendar day to obtain daily data, for example, the first... Heavenly Historical wind power data at that time; This also includes standardizing the acquired data before reorganizing it, using the Z-score standardization calculation method as follows: ; In the formula, This represents the standardized data; This represents the input data, i.e., all the data obtained; This represents the mean of the input data; This represents the standard deviation of the input data; S3. Cluster the historical daily electricity price data and classify typical electricity price scenarios based on the clustering results; The elbow rule is used to determine the optimal number of clusters for historical daily electricity price data. ; The K-means algorithm was used to cluster historical daily electricity price data to obtain... A typical electricity pricing scenario; S4. Calculate the prior probability of each typical electricity price scenario based on historical daily electricity price data, specifically based on the... The ratio of the number of historical daily electricity price data points to the total number of historical daily electricity price data points for a typical electricity price scenario is used to calculate the first... The prior probabilities of a typical electricity pricing scenario are calculated as follows: ; In the formula, Indicates the first Prior probabilities of a typical electricity price scenario; Indicates the first Number of historical daily electricity price data for a typical electricity price scenario; This indicates the total number of historical daily electricity price data. S5. Cluster the historical daily wind power data under different typical electricity price scenarios, divide typical wind power scenarios according to the clustering results, and construct the mapping relationship between typical electricity price scenarios and typical wind power scenarios. Clustering labels are extracted from typical electricity price scenarios. These clustering labels are used to identify which electricity price scenario the historical electricity price daily data belongs to. For example, if the historical electricity price daily data of a certain day belongs to the second electricity price scenario, then its clustering label is 2. After mapping to the historical wind power daily data, the historical wind power daily data of that day is divided into the wind power data subset under electricity price scenario 2. Based on the clustering label mapping to historical wind power daily data, extract the historical wind power data subset corresponding to each typical electricity price scenario in the historical wind power daily data. That is, the number of typical electricity price scenarios and historical wind power data subsets are the same. The elbow rule is used to determine the optimal number of wind power clusters for a subset of historical wind power data. ; The K-means algorithm was used to cluster a subset of historical wind power data to obtain... A typical wind power scenario; S6. Calculate the conditional probability of typical wind power scenarios based on historical daily wind power data, specifically: based on the... The subset of historical wind power data belongs to the first The number of historical daily wind power data for a typical wind power scenario and the number of... The ratio of the total number of historical wind power daily data points in the first subset of historical wind power data is used to calculate the first... Under a typical electricity pricing scenario, the first The conditional probability of a typical wind power scenario is calculated as follows: ; In the formula, Indicates the first Under a typical electricity pricing scenario, the first Conditional probabilities of a typical wind power scenario; Indicates the first The subset of historical wind power data belongs to the first The number of historical daily wind power data for a typical wind power scenario; Indicates the first The total number of historical wind power daily data in each historical wind power data subset; The mapping relationship between the typical electricity price scenario and the typical wind power scenario is as follows: each typical electricity price scenario corresponds to one or more typical wind power scenarios, and each typical wind power scenario belongs to only one typical electricity price scenario, forming a directed conditional mapping from the typical electricity price scenario to the typical wind power scenario. S7. Calculate the joint probability of typical electricity price scenarios and typical wind power scenarios based on prior probability and conditional probability. The calculation method is as follows: ; In the formula, Denotes the joint probability, i.e., the first... A typical electricity pricing scenario With the Under a typical electricity pricing scenario, the first A typical wind power scenario The probability of them happening simultaneously; S8. Based on simulated wind power daily data and joint probability, obtain the posterior probability that simulated wind power daily data belongs to different typical electricity price scenarios; S81. Calculate the Euclidean distance between simulated daily wind power data and typical wind power scenarios to obtain the joint probability. The calculation method is as follows: ; In the formula, Indicates the first Simulated daily wind power data and the first Under a typical electricity pricing scenario, the first A typical wind power scenario The Euclidean distance between them; Indicates the first Simulated daily wind power data for the day; The wind power scenario here refers to the data within this wind power scenario cluster; S82. The Euclidean distance is converted into corresponding matching weights using a Gaussian kernel function. The calculation method is as follows: ; In the formula, Indicates the first Simulated daily wind power data and the first Under a typical electricity pricing scenario, the first A typical wind power scenario Matching weights between them; Represents an exponential function; The hyperparameters representing the Gaussian kernel function; S83. Sort all matching weights in descending order and select the top... Each matching weight is used as the scene matching weight; S84. Calculate the posterior probability that the simulated daily wind power data belongs to a typical electricity price scenario based on scenario matching weights and joint probability. The calculation method is as follows: ; In the formula, Indicates the first The simulated wind power daily data belongs to the first day. A typical electricity pricing scenario The posterior probability; Indicates the first Simulated daily wind power data and the first Under a typical electricity pricing scenario, the first A typical wind power scenario Scene matching weights between them; S9. The original simulated expected electricity price is obtained by weighting the posterior probabilities of each typical electricity price scenario. The calculation method is as follows: ; In the formula, Indicates the first Heavenly The original simulated expected electricity price at that time; S10. Based on historical bidding space daily data and simulated bidding space daily data, the original simulated expected electricity price is corrected to obtain the simulated expected electricity price; S1001. Input the simulated daily bidding space data into the trained multilayer perceptron regression neural network to obtain the predicted daily price. The calculation method is as follows: ; In the formula, Indicates the first The daily average electricity price predicted by the heaven; This represents a multilayer perceptron regressive neural network that has completed training. Indicates the first Daily data on simulated bidding space; S1002, Calculate the historical bidding center for daily historical bidding space data. The calculation method is as follows: , ; ; In the formula, Indicates the historical bidding center; Indicates the first Heavenly The average of daily historical bidding space data; This represents the total number of days in the historical wind power data. Indicates the first Historical bidding space daily data; Indicates the first Heavenly Historical bidding space daily data; It also includes training a multilayer perceptron regression neural network based on historical daily bidding space data; The multilayer perceptron regressive neural network includes a 24-node input layer, a 64-node first hidden layer, a 32-node second hidden layer, and a 1-node output layer. The hidden layer uses ReLU activation and mean squared error (MSE) loss function. Early stopping is used to prevent overfitting. S1003. Calculate the Mahalanobis distance between the simulated daily bidding space data and the historical bidding center. The calculation method is as follows: ; In the formula, Indicates the first The Mahalanobis distance between the simulated daily bidding space vector and the historical bidding center; Indicates the transpose operation; The inverse matrix of the covariance matrix; S1004. Calculate the adaptive correction weights based on the Mahalanobis distance; Set Mahalanobis distance threshold ; If the Mahalanobis distance is less than or equal to the Mahalanobis distance threshold, the Mahalanobis distance is input into an exponential function with a decay coefficient to obtain the decayed Mahalanobis distance; the adaptive correction weight is obtained based on the product of the decayed Mahalanobis distance and the preset maximum correction coefficient. Otherwise, the adaptive adjustment weight is 0; The adaptive correction weights are calculated as follows: ; In the formula, Indicates the first The adaptive adjustment weights of the day; Indicates the maximum correction factor; Represents an exponential function; Indicates the attenuation coefficient; S1005. Calculate the daily average of the simulated expected electricity price based on the adaptively adjusted weights, the predicted daily electricity price, and the original simulated expected electricity price. The calculation method is as follows: ; In the formula, Indicates the first The average daily expected electricity price for the day; Indicates the first The average daily value of the original simulated expected electricity price for the day; in, , Indicates the first Heavenly The original simulation expected electricity price; S1006. The original simulated expected electricity price is corrected based on the daily average value of the simulated expected electricity price to obtain the simulated expected electricity price. The calculation method is as follows: ; In the formula, Indicates the first Heavenly Simulated expected electricity price; In one specific embodiment, a proposed wind farm with an installed capacity of 200MW is constructed. Historical hourly electricity prices, wind power output, and bidding space data of the region in recent years are collected. Simulated wind power output data of the proposed wind farm for one year generated by wind power simulation software and simulated bidding space data of the region for the next year generated by other means are also collected. The data sampling interval is 1 hour, with 24 points per day. Historical wind power data and simulated daily wind power data are standardized and recombined. Taking a certain day as an example, the standardized data is as follows: Figure 2 As shown; The K-means algorithm was used to cluster historical daily electricity price data; the elbow rule was used to determine the optimal number of clusters K=3, resulting in three typical electricity price scenarios after clustering. , and The clustering results are as follows Figure 3 As shown, the prior probabilities for each typical electricity price scenario are: , and ; Extract the corresponding historical wind power data subsets for each typical electricity price scenario from the historical wind power data to obtain 3 historical wind power data subsets; The K-means algorithm was used to cluster each subset of historical wind power data. The elbow rule was used to determine the optimal number of clusters as 4, 4, and 5. Specifically, there were 4 typical wind power scenarios under the first typical electricity price scenario, 4 typical wind power scenarios under the second typical electricity price scenario, and 5 typical wind power scenarios under the third typical electricity price scenario. Figure 4-6 As shown; at this time, the corresponding conditional probability is , , , , , , , , , , , , ; Calculate the joint probability based on the prior probabilities of typical electricity price scenarios and typical wind power scenarios; The hyperparameter of the Gaussian kernel function is set to 0.05, and the first three matching weights are selected as scene matching weights. The specific matching weight calculation process is shown in Table 1. Table 1. Example of matching weight calculation

[0024] Calculate the posterior probability of simulated daily wind power data belonging to a typical electricity price scenario. , and ; The original expected electricity price is calculated as shown in Table 2; Table 2 Example of Calculating Original Expected Electricity Price

[0025] Multilayer perceptron regressive neural network, such as Figure 5 As shown, the training set and validation set are divided in a 9:1 ratio, the maximum number of training rounds is 500, and the early stopping condition is set to the loss function not decreasing for 10 consecutive rounds. Input the simulated bidding space daily data to obtain the predicted daily electricity price Yuan / MWh; The Mahalanobis distance between the simulated daily bidding space data and the historical bidding center is calculated as follows: ; Choosing a maximum correction coefficient of 0.7 and a decay coefficient of 0.1, the adaptive correction weight is obtained as follows: ; Calculate the average daily value of the simulated expected electricity price Yuan / MWh; The original simulated expected electricity price was corrected based on the daily average of the simulated expected electricity price, and the simulated expected electricity price was obtained, as shown in Table 3. Table 3. Example of Calculating Expected Electricity Price in Simulation

[0026] Example 2: This embodiment provides a wind power simulation data electricity price generation system based on scene mapping. The system includes: The data acquisition module acquires historical electricity price data, historical wind power data, historical bidding space data, simulated wind power data, and simulated bidding space data; it then reorganizes the acquired data to obtain daily historical electricity price data, daily historical wind power data, daily historical bidding space data, daily simulated wind power data, and daily simulated bidding space data. The probability calculation module clusters historical daily electricity price data and classifies typical electricity price scenarios based on the clustering results; it calculates the prior probability of each typical electricity price scenario based on the historical daily electricity price data; it clusters historical daily wind power data under different typical electricity price scenarios, classifies typical wind power scenarios based on the clustering results, and constructs a mapping relationship between typical electricity price scenarios and typical wind power scenarios; it calculates the conditional probability of typical wind power scenarios based on historical daily wind power data; and it calculates the joint probability of typical electricity price scenarios and typical wind power scenarios based on the prior probability and conditional probability. The electricity price calculation module obtains the posterior probability of the simulated wind power daily data belonging to different typical electricity price scenarios based on the simulated wind power daily data and joint probability; and performs weighted calculation based on the posterior probability of each typical electricity price scenario to obtain the original simulated expected electricity price. The electricity price correction module corrects the original simulated expected electricity price based on historical bidding space daily data and simulated bidding space daily data to obtain the simulated expected electricity price.

[0027] Example 3: This embodiment provides a wind-storage power station scheduling optimization method based on the expected electricity price simulated in wind power simulation. The method includes: T1. Obtain simulation data, including simulated expected electricity price and simulated wind power output data; The simulated expected electricity price is used to characterize the market electricity price signal of the power system in each time period, and the simulated wind power output data is used to characterize the wind power generation capacity in each time period. T2. Construct a wind power station scheduling optimization model based on simulation data. The wind power station scheduling optimization model includes an optimization objective function constructed with the goal of maximizing the total revenue of the wind power station within the scheduling period, and constraint conditions constructed with preset constraint conditions. The total revenue is the sum of the revenue from selling electricity from wind turbines and the revenue from selling electricity from energy storage systems, minus the operating costs of the energy storage systems. The constraints include at least one of the following: The power balance constraint is that the sum of the output of the wind turbine and the net output of the energy storage system should be equal to the total power transmitted to the grid by the wind-storage power station minus the power curtailed by the wind turbine. The state of charge (SOC) constraint of the energy storage system is that the SOC of the energy storage system at each time period should be between the preset minimum SOC and maximum SOC, and the recursive relationship of the SOC satisfies the charging and discharging efficiency constraint. The charging and discharging power constraints of the energy storage system are as follows: the charging power and discharging power of the energy storage system at each time period should not exceed its maximum charging power and maximum discharging power, respectively. The curtailment rate constraint for wind turbines is that the curtailment rate of wind turbines throughout the entire dispatch cycle should not exceed the preset curtailment rate. T3. Solve the wind-storage power station scheduling optimization model to obtain the charging and discharging plan of the energy storage system and the power distribution scheme of the wind turbine in each time period; When the wind-storage power station scheduling optimization model is in linear or mixed-integer linear form, the mixed-integer linear programming method is used for solving it; when the wind-storage power station scheduling optimization model is in nonlinear form, the interior-point method is used for solving it. T4. Generate wind-storage power station dispatch and operation instructions based on the charging and discharging plans of the energy storage system and the power distribution scheme of the wind turbine in each time period; Based on the charging and discharging plans of the energy storage system and the power output allocation scheme of the wind turbine, a wind-storage power station scheduling and operation instruction is generated; the wind-storage power station scheduling and operation instruction is issued to the energy storage system and the wind turbine for execution to guide the scheduling and operation of the wind-storage power station; the wind-storage power station scheduling and operation instruction includes the charging / discharging power instruction of the energy storage system for each time period and the instructions of the grid-connected power, abandoned power, and power transmitted to the energy storage of the wind turbine for each time period.

[0028] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0029] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0030] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0031] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, 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 this application. 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.

[0032] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for generating electricity prices from wind power simulation data based on scene mapping, characterized in that, The method includes: Acquire historical electricity price data, historical wind power data, historical bidding space data, simulated wind power data, and simulated bidding space data; reorganize the acquired data to obtain historical daily electricity price data, historical daily wind power data, historical daily bidding space data, simulated daily wind power data, and simulated daily bidding space data. Cluster historical daily electricity price data and classify typical electricity price scenarios based on the clustering results; calculate the prior probability of each typical electricity price scenario based on historical daily electricity price data; Historical daily wind power data under different typical electricity price scenarios are clustered, and typical wind power scenarios are divided according to the clustering results. A mapping relationship between typical electricity price scenarios and typical wind power scenarios is constructed. The conditional probability of typical wind power scenarios is calculated based on historical daily wind power data. The joint probability of typical electricity price scenarios and typical wind power scenarios is calculated based on prior probability and conditional probability. Based on simulated daily wind power data and joint probability, the posterior probability of simulated daily wind power data belonging to different typical electricity price scenarios is obtained; the original simulated expected electricity price is obtained by weighted calculation based on the posterior probability of each typical electricity price scenario. The original simulated expected electricity price is corrected based on historical bidding space daily data and simulated bidding space daily data to obtain the simulated expected electricity price.

2. The method for generating electricity prices from wind power simulation data based on scene mapping according to claim 1, characterized in that, The acquired data is reorganized, and the specific steps are as follows: The acquired data is segmented according to a 24-hour calendar day to obtain the corresponding daily data.

3. The method for generating electricity prices from wind power simulation data based on scene mapping according to claim 2, characterized in that, Clustering of historical daily electricity price data, specifically: The elbow rule is used to determine the optimal number of clusters for historical daily electricity price data. ; The K-means algorithm was used to cluster historical daily electricity price data to obtain... A typical electricity pricing scenario; Based on the The ratio of the number of historical daily electricity price data points for a typical electricity price scenario to the total number of historical daily electricity price data points is used to calculate the first... Prior probabilities for a typical electricity price scenario.

4. The method for generating electricity prices from wind power simulation data based on scene mapping according to claim 3, characterized in that, Clustering of historical daily wind power data under different typical electricity price scenarios is performed as follows: Extract cluster labels from typical electricity pricing scenarios; Based on the clustering label mapping to historical wind power daily data, extract the corresponding historical wind power data subset for each typical electricity price scenario in the historical wind power daily data; The elbow rule is used to determine the optimal number of wind power clusters for a subset of historical wind power data. ; The K-means algorithm was used to cluster a subset of historical wind power data to obtain... A typical wind power scenario; Based on the The subset of historical wind power data belongs to the first The number of historical daily wind power data for a typical wind power scenario and the number of... The ratio of the total number of historical wind power daily data points in the first subset of historical wind power data is used to calculate the first... Under a typical electricity pricing scenario, the first Conditional probabilities for a typical wind power scenario.

5. The method for generating electricity prices from wind power simulation data based on scene mapping according to claim 1, characterized in that, The mapping relationship between the typical electricity price scenario and the typical wind power scenario is as follows: each typical electricity price scenario corresponds to one or more typical wind power scenarios, and each typical wind power scenario belongs to only one typical electricity price scenario.

6. The method for generating electricity prices from wind power simulation data based on scene mapping according to claim 1, characterized in that, The posterior probability of simulated daily wind power data belonging to different typical electricity price scenarios is obtained based on simulated daily wind power data and joint probability. The specific steps are as follows: The Euclidean distance between simulated daily wind power data and typical wind power scenarios is calculated, and the Euclidean distance is converted into corresponding matching weights using a Gaussian kernel function. Sort all matching weights in descending order and select the top [weights]. Each matching weight is used as the scene matching weight; Based on scenario matching weights and joint probability calculations, the posterior probability of simulated daily wind power data belonging to a typical electricity price scenario is obtained.

7. The method for generating electricity prices from wind power simulation data based on scene mapping according to claim 6, characterized in that, The original simulated expected electricity price is corrected based on historical bidding space daily data and simulated bidding space daily data. The specific steps are as follows: The simulated daily bidding space data is input into the trained multilayer perceptron regression neural network to obtain the predicted daily price. Calculate the historical bid center for daily historical bid space data; Calculate the Mahalanobis distance between the simulated daily bidding space data and the historical bidding center; The adaptive correction weights are calculated based on the Mahalanobis distance. The daily average of the simulated expected electricity price is calculated based on the daily average of the adaptively adjusted weights, the predicted daily electricity price, and the original simulated expected electricity price. The original simulated expected electricity price is corrected based on the daily average value of the simulated expected electricity price to obtain the simulated expected electricity price.

8. The method for generating electricity prices from wind power simulation data based on scene mapping according to claim 7, characterized in that, The adaptive correction weights are calculated based on the Mahalanobis distance, specifically as follows: Set the Mahalanobis distance threshold; If the Mahalanobis distance is less than or equal to the Mahalanobis distance threshold, the Mahalanobis distance is input into an exponential function with a decay coefficient to obtain the decayed Mahalanobis distance; the adaptive correction weight is obtained based on the product of the decayed Mahalanobis distance and the preset maximum correction coefficient. Otherwise, the adaptive adjustment weight is 0.

9. A wind power simulation data electricity price generation system based on scene mapping, characterized in that, The system includes: The data acquisition module acquires historical electricity price data, historical wind power data, historical bidding space data, simulated wind power data, and simulated bidding space data; it then reorganizes the acquired data to obtain daily historical electricity price data, daily historical wind power data, daily historical bidding space data, daily simulated wind power data, and daily simulated bidding space data. The probability calculation module clusters historical daily electricity price data and classifies typical electricity price scenarios based on the clustering results; it calculates the prior probability of each typical electricity price scenario based on the historical daily electricity price data; it clusters historical daily wind power data under different typical electricity price scenarios, classifies typical wind power scenarios based on the clustering results, and constructs a mapping relationship between typical electricity price scenarios and typical wind power scenarios; it calculates the conditional probability of typical wind power scenarios based on historical daily wind power data; and it calculates the joint probability of typical electricity price scenarios and typical wind power scenarios based on the prior probability and conditional probability. The electricity price calculation module obtains the posterior probability of the simulated wind power daily data belonging to different typical electricity price scenarios based on the simulated wind power daily data and joint probability; and performs weighted calculation based on the posterior probability of each typical electricity price scenario to obtain the original simulated expected electricity price. The electricity price correction module corrects the original simulated expected electricity price based on historical bidding space daily data and simulated bidding space daily data to obtain the simulated expected electricity price.

10. A wind-storage power station scheduling optimization method based on simulated expected electricity price, characterized in that, The method includes: Acquire simulation data, including simulated expected electricity price and simulated wind power output data; A wind power plant scheduling optimization model is constructed based on simulation data. The wind power plant scheduling optimization model includes an optimization objective function constructed with the goal of maximizing the total revenue of the wind power plant within the scheduling period, and constraint conditions constructed with preset constraint conditions. Solve the wind-storage power station scheduling optimization model to obtain the charging and discharging plan of the energy storage system and the power output allocation scheme of the wind turbine in each time period; The wind power station dispatch and operation instructions are generated based on the charging and discharging plans of the energy storage system and the power output allocation scheme of the wind turbines for each time period.

Images