Electricity market monthly time-of-use transaction average price prediction method and device and storage medium

Abstract

The present application relates to the technical field of electricity transaction average price prediction, in particular to a kind of electricity market monthly time-sharing transaction average price prediction method, device, equipment and computer storage medium.The present application realizes the accurate and reliable prediction of monthly time-sharing transaction average price by establishing a comprehensive prediction framework that integrates the monthly probability characteristics of new energy and the spot market mechanism.The present application first deeply couples the monthly output probability prediction of new energy, the cumulative probability density modeling and the formation mechanism of spot price, uses beta distribution fitting and clustering analysis to improve the stability of probability characteristics, introduces multi-dimensional market characteristics and linear regression model to enhance the adaptability of prediction to actual market conditions, and finally quantifies the influence of wind and light uncertainty on average price through integrated calculation and error calibration, thereby significantly improving the accuracy and practicality of the prediction results, providing a reliable decision basis for market participants to avoid medium and long-term transaction risks and optimize their inventory strategies.

Description

Technical Field

[0001] This invention relates to the field of electricity trading average price prediction technology, and in particular to a method, apparatus, equipment, and computer storage medium for predicting monthly time-segmented average trading prices in the electricity market. Background Technology

[0002] With the continuous expansion of new energy capacity, the balance between power supply and demand faces challenges, leading to significant fluctuations in spot market prices. Especially in extreme scenarios such as prolonged periods of no wind or solar power generation or periods of high power generation from new energy sources, accurately predicting the monthly time-of-use average transaction price becomes a critical issue. Existing research mainly focuses on long-term power generation forecasting for new energy sources and time-series modeling of spot market prices. However, it generally lacks in-depth analysis of the probabilistic characteristics of monthly new energy output and fails to establish an effective correlation between the cumulative probability density of new energy and the average spot clearing price. This results in insufficient adaptability and limited reliability of the forecast results in actual market transactions, failing to effectively support market participants in risk aversion and profit optimization in medium- and long-term transactions. Therefore, there is an urgent need for a forecasting method that integrates the monthly probabilistic characteristics of new energy and the mechanism of spot price formation to improve the accuracy and practicality of monthly time-of-use average transaction price prediction. Summary of the Invention

[0003] Therefore, the technical problem to be solved by the present invention is to overcome the problem that the existing technology fails to effectively combine the monthly output probability characteristics of new energy with the formation mechanism of the average price in the spot market, resulting in insufficient accuracy and limited adaptability of the monthly time-sharing average transaction price prediction.

[0004] To address the aforementioned technical problems, this invention provides a method for predicting the monthly time-of-use average transaction price in the electricity market, comprising:

[0005] Based on historical renewable energy output data and system operation constraints, a monthly renewable energy output probability prediction model is generated, and its cumulative probability density function is fitted by beta distribution. Based on the cumulative probability density function of new energy, historical beta distribution parameters are introduced, and cluster analysis is used to determine the typical values ​​of the distribution parameters, thus establishing a quantitative conversion relationship between the cumulative probability density of new energy and the average price of the spot market. Based on the quantitative conversion relationship between the cumulative probability density of new energy and the average price in the spot market, and combined with the statistical characteristics of historical power output fluctuations, the average power output of new energy in each period of the month is calculated to form a time-segmented average power output sequence. Based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output and system operating capacity, key influencing factors are screened through feature importance analysis, and a linear regression spot clearing electricity price prediction model is established to generate spot clearing time-of-use electricity price prediction values. Based on the monthly time-of-day average output sequence of new energy sources, the spot clearing price prediction model, and the system operation status reconstruction input, the average spot price prediction considering the monthly fluctuation of new energy output is obtained through integrated calculation. The result is then calibrated by combining the prediction error probability threshold to form the final monthly spot clearing average price prediction output that takes into account the uncertainty of wind and solar power.

[0006] Preferably, the step of generating a monthly renewable energy output probability prediction model based on historical renewable energy output data and system operation constraints, and fitting its cumulative probability density function using a beta distribution, includes: Based on historical renewable energy output data and system operation constraints including renewable energy installed capacity constraints, probability distribution constraints, and wind and solar resource constraints, a monthly renewable energy output probability density function is constructed to generate a monthly renewable energy output probability prediction model. Integrating the monthly output probability density function of new energy sources yields the cumulative distribution function, which is then fitted using a beta distribution to obtain the fitted cumulative probability density function.

[0007] Preferably, the new energy installed capacity constraint variables include the new energy installed capacity in time period t and the maximum simultaneous rate of new energy in the region or station; the probability distribution constraint variables include the standard deviation of the logarithm of the new energy output variable, the mean of the logarithm of the new energy output probability density variable, the standard deviation of the logarithm of the new energy output probability density variable, the maximum and minimum output of new energy during normal operation, and the lower limit of the new energy output probability density distribution; the wind and solar resource constraint variables include the total area of ​​the solar photovoltaic array, the actual irradiance, the efficiency of the solar photovoltaic system, the effective wind energy density, the air density, and the wind speed.

[0008] Preferably, the step of establishing a quantitative conversion relationship between the cumulative probability density function of new energy sources, introducing historical beta distribution parameters, using cluster analysis to determine typical values ​​of the distribution parameters, and establishing a quantitative conversion relationship between the cumulative probability density function of new energy sources and the average spot market price includes: Historical beta distribution parameters are introduced as candidate variables, and combined with the current monthly new energy resource status data, a parameter candidate set is constructed to generate a beta distribution parameter candidate set; Based on the candidate set of beta distribution parameters, the K-means clustering method is used to classify the candidate parameters. After removing outliers, the mean of each class of parameters is taken as the typical value of the parameter, thus generating typical values ​​of the beta distribution parameters. Based on the typical values ​​of beta distribution parameters and the cumulative probability density function of new energy sources, a quantitative conversion relationship between the cumulative probability density of new energy sources and the average price in the spot market is established.

[0009] Preferably, the step of calculating the average output of new energy sources in each period of the month, based on the quantitative conversion relationship between the cumulative probability density of new energy sources and the average price in the spot market, and combining the historical output fluctuation statistical characteristics, to form a time-segmented average output sequence includes: Based on the historical monthly time-period output probability distribution standard deviation mean data, calculate the historical time-period output fluctuation standard deviation mean; Based on the quantitative conversion relationship between the cumulative probability density of new energy sources and the average price in the spot market, and the standard deviation of historical time-sharing power output fluctuations, the average power output of new energy sources at each moment within a month is calculated, generating a time-sharing power output average sequence.

[0010] Preferably, the step of establishing a linear regression spot clearing electricity price prediction model based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output, and system operating capacity, and selecting key influencing factors through feature importance analysis, to generate the spot clearing time-of-use electricity price prediction value includes: Based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, new energy output and system operating capacity, the feature importance analysis function in the linear regression model is used to screen key features that have a significant impact on the spot clearing average price. Based on key features, a multiple linear regression model is established to fit the relationship between the average spot clearing price and each feature, thereby generating a spot clearing electricity price prediction model. Real-time feature data is input into the spot clearing electricity price prediction model to calculate the time-of-use electricity price prediction value for spot clearing.

[0011] Preferably, the step of integrating the monthly time-segmented average output sequence of new energy sources, the spot clearing price prediction model, and the system operation status reconstruction input to obtain the average spot price prediction considering monthly new energy output fluctuations, and then calibrating the results by combining the prediction error probability threshold, to form the final monthly spot clearing average price prediction output taking into account the uncertainties of wind and solar power, includes: Based on the system operation status reconstruction input features, including new energy installed capacity, load data, external or incoming data, and system operating capacity, a reconstructed system operation status feature set is obtained. The monthly average output of new energy sources by time period and the reconstructed system operation status feature set are input into the spot clearing electricity price prediction model to calculate the monthly spot forecast average considering the fluctuation of new energy output. The monthly spot price forecast average is calibrated based on the prediction error probability threshold to obtain the final monthly spot clearing average price forecast output that takes into account the uncertainty of wind and solar power.

[0012] The present invention also provides a device for predicting the monthly time-of-use average transaction price in the electricity market, comprising: The monthly renewable energy output probability prediction model building module is used to generate a monthly renewable energy output probability prediction model based on historical renewable energy output data and system operation constraints, and fit its cumulative probability density function through beta distribution. The coupling relationship construction module is used to establish a quantitative conversion relationship between the cumulative probability density of new energy and the average price of the spot market, based on the cumulative probability density function of new energy, by introducing historical beta distribution parameters, using cluster analysis to determine typical values ​​of the distribution parameters; The time-segmented average output sequence construction module is used to calculate the average output of new energy in each time period within a month, based on the quantitative conversion relationship between the cumulative probability density of new energy and the average price in the spot market, combined with the statistical characteristics of historical output fluctuations, to form a time-segmented average output sequence. The spot clearing time-of-use electricity price prediction model construction module is used to establish a linear regression spot clearing electricity price prediction model based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output and system operating capacity, by screening key influencing factors through feature importance analysis, and generating the spot clearing time-of-use electricity price prediction value. The spot clearing average price prediction module is used to calculate the average spot price prediction that takes into account the monthly fluctuations in new energy output based on the monthly time-segmented average output sequence of new energy, the spot clearing electricity price prediction model, and the system operation status reconstruction input. The module then combines the prediction error probability threshold to calibrate the results and form the final monthly spot clearing average price prediction output that takes into account the uncertainties of wind and solar power.

[0013] This invention also provides a device for predicting the monthly time-of-use average transaction price in the electricity market, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the steps of the above-described method for predicting the monthly time-of-use average transaction price in the electricity market.

[0014] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for predicting the average monthly time-of-use transaction price in the electricity market.

[0015] The technical solution of the present invention has the following advantages over the prior art: The method for predicting monthly time-of-use average transaction prices in the electricity market, as described in this invention, achieves accurate and reliable predictions of monthly time-of-use average transaction prices by establishing a comprehensive prediction framework that integrates the monthly probabilistic characteristics of new energy sources with the spot market mechanism. This invention is the first to deeply couple the monthly output probability prediction of new energy sources, cumulative probability density modeling, and the spot price formation mechanism, utilizing beta distribution fitting and cluster analysis to improve the stability of the probabilistic characteristics. By introducing multi-dimensional market characteristics and a linear regression model, the adaptability of the prediction to actual market conditions is enhanced. Finally, through integrated calculation and error calibration, the impact of wind and solar uncertainties on the average price is effectively quantified, thereby significantly improving the accuracy and practicality of the prediction results and providing a reliable decision-making basis for market participants to avoid medium- and long-term trading risks and optimize their holding strategies. Attached Figure Description

[0016] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein: Figure 1 This is a flowchart illustrating the implementation of a method for predicting the average monthly transaction price in the electricity market based on time periods, as provided by this invention. Figure 2 This is a structural block diagram of a monthly time-of-use average transaction price prediction device for the electricity market provided in an embodiment of the present invention. Detailed Implementation

[0017] The core of this invention is to provide a method, apparatus, equipment, and computer storage medium for predicting the monthly time-of-use average transaction price in the electricity market, which effectively achieves accurate and reliable prediction of the monthly time-of-use average transaction price.

[0018] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely 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] Please refer to Figure 1. Figure 1 The flowchart illustrates the implementation of a method for predicting the monthly time-of-use average transaction price in the electricity market provided by this invention; the specific operation steps are as follows: S101: Based on historical renewable energy output data and system operation constraints, a monthly renewable energy output probability prediction model is generated, and its cumulative probability density function is fitted by beta distribution. S102: Based on the cumulative probability density function of new energy, historical beta distribution parameters are introduced, and cluster analysis is used to determine the typical values ​​of the distribution parameters, and a quantitative conversion relationship between the cumulative probability density of new energy and the average price of the spot market is established. S103: Based on the quantitative conversion relationship between the cumulative probability density of new energy and the average price in the spot market, and combined with the statistical characteristics of historical power output fluctuations, calculate the average power output of new energy in each period of the month to form a time-segmented power output average sequence. S104: Based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output and system operating capacity, key influencing factors are screened through feature importance analysis, a linear regression spot clearing electricity price prediction model is established, and spot clearing time-of-use electricity price prediction values ​​are generated. S105: Based on the monthly time-segmented average output sequence of new energy sources, the spot clearing price prediction model, and the system operation status reconstruction input, the average spot price prediction considering the monthly fluctuation of new energy output is obtained through integrated calculation. The result is then calibrated by combining the prediction error probability threshold to form the final monthly spot clearing average price prediction output that takes into account the uncertainty of wind and solar power.

[0020] It should be noted that "wind and solar" refers to wind power and photovoltaic power.

[0021] Based on the above embodiments, this embodiment will provide a detailed description of step S101: In some embodiments, a monthly renewable energy output probability prediction model is generated based on historical renewable energy output data and system operation constraints, and its cumulative probability density function is fitted using a beta distribution, including: Based on historical renewable energy output data and system operation constraints including renewable energy installed capacity constraints, probability distribution constraints, and wind and solar resource constraints, a monthly renewable energy output probability density function is constructed to generate a monthly renewable energy output probability prediction model. • Integrate the probability density function of monthly output of new energy sources to obtain the cumulative distribution function, and then fit it with a beta distribution to obtain the fitted cumulative probability density function.

[0022] In other embodiments, the renewable energy installed capacity constraint includes the renewable energy installed capacity in time period t and the maximum simultaneous renewable energy rate of a region or power station; the probability distribution constraint includes the standard deviation of the logarithm of the renewable energy output variable, the mean of the logarithm of the renewable energy output probability density variable, the standard deviation of the logarithm of the renewable energy output probability density variable, the maximum and minimum output of renewable energy during normal operation, and the lower limit of the renewable energy output probability density distribution; the wind and solar resource constraint includes the total area of ​​the solar photovoltaic array, the actual irradiance, the efficiency of the solar photovoltaic system, the effective wind energy density, the air density, and the wind speed.

[0023] Specifically, the expressions for installed capacity constraints, probability distribution constraints, and wind and solar resource constraints are as follows: (1) (2) (3) In the formula: To contribute to the monthly power generation schedule of new energy sources for New energy installed capacity during the period This refers to the maximum simultaneous rate of new energy in the region or at the facility. The standard deviation of the logarithm of the output variable of new energy sources during this cycle. It is the probability density variable of new energy output. The logarithm mean, yes The standard deviation of the logarithm, Variables Appeared in -0.5 to The probability density within the range of +0.5 represents the maximum and minimum power output of new energy sources during normal operation. This is the lower bound of the probability density distribution of new energy power output; considering practical considerations, it should be taken as 0. for The standard deviation of ; A is the total area of ​​the solar photovoltaic array; G is the actual solar irradiance in a certain location over a period of time. For solar photovoltaic conversion efficiency, For the efficiency of solar photovoltaic systems, For effective wind energy density, air density, Let be the wind speed with the i-th value between 3 and 25, and n be the number of wind speed sampling points.

[0024] Based on the above embodiments, this embodiment will provide a detailed description of step S102: In some embodiments, based on the cumulative probability density function of new energy sources, historical beta distribution parameters are introduced, and cluster analysis is used to determine typical values ​​of the distribution parameters, establishing a quantitative conversion relationship between the cumulative probability density of new energy sources and the average spot market price, including: • Historical beta distribution parameters (α and β values) are introduced as candidate variables, and combined with the current monthly new energy resource status data, a parameter candidate set is constructed to generate a beta distribution parameter candidate set; • Based on the candidate set of beta distribution parameters, the K-means clustering method is used to classify the candidate parameters. After removing outliers, the mean of each class of parameters is taken as the typical value of the parameter to generate typical values ​​of beta distribution parameters. • Based on the typical values ​​of beta distribution parameters and the cumulative probability density function of new energy sources, a quantitative conversion relationship between the cumulative probability density of new energy sources and the average price in the spot market is established.

[0025] Specifically, the cumulative probability density function of new energy reveals the coupling mechanism between the cumulative probability density of new energy and the average price in the spot market, and its expression is as follows: (4) (5) (6) (7) In the formula, For regularized incomplete beta functions, and These are the incomplete beta function and the complete beta function, respectively, where t is the integration variable. It's a gamma function. To fit a regularized incomplete beta function, the most crucial step is determining the values ​​of the shape parameters α and β of the beta distribution. Considering the similarity of new energy resources between adjacent months, and given that the fitting result of the cumulative probability density function for the previous month is known when predicting monthly cumulative probability density functions, the beta distribution parameters for the previous month—α and β values—are highly valuable variables. Therefore, similar parameters from the previous month are also considered as candidate variables. The relationship between these α and β values ​​and similar parameters from the previous month, as well as the status of new energy resources, is observed. The variable most conducive to selecting typical values ​​is selected. After manually removing obvious outliers, K-means clustering analysis is used to divide this variable into several typical classes. If no obvious simple functional relationship exists, the mean of the parameter (α or β) in each class is taken as the typical value of that parameter. This represents the conversion relationship between the monthly cumulative probability density of new energy and the average spot clearing price. This represents the average price in the spot market.

[0026] Based on the above embodiments, this embodiment will provide a detailed description of step S103: In some embodiments, based on the quantitative conversion relationship between the cumulative probability density of new energy sources and the average spot market price, and combined with the statistical characteristics of historical power output fluctuations, the average power output of new energy sources in each period of the month is calculated to form a time-segmented power output average sequence, including: • Calculate the mean standard deviation of historical time-period power output fluctuations based on the historical monthly time-period power output probability distribution standard deviation data; Based on the quantitative conversion relationship between the cumulative probability density of new energy and the average price in the spot market, and the mean standard deviation of historical time-sharing power output fluctuations, the average power output of new energy at each moment within a month is calculated, generating a time-sharing power output average sequence.

[0027] Specifically, the monthly average output of new energy sources across different time periods is expressed as follows: (8) In the formula: Indicates monthly output by time period Time-to-time average; The mean standard deviation of the probability distribution of monthly output at different times over the past 10 years.

[0028] Based on the above embodiments, this embodiment will provide a detailed description of step S104: In some embodiments, based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output, and system operating capacity, key influencing factors are screened through feature importance analysis, and a linear regression spot clearing electricity price prediction model is established to generate spot clearing time-of-use electricity price prediction values, including: • Based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, new energy output and system operating capacity, the feature importance analysis function in the linear regression model is used to screen key features that have a significant impact on the spot clearing average price. • Based on key features, a multiple linear regression model is established to fit the relationship between the average spot clearing price and each feature, thereby generating a spot clearing electricity price prediction model; • Input real-time feature data into the spot clearing electricity price prediction model to calculate the time-of-use electricity price prediction value for spot clearing.

[0029] Specifically, the expression for the time-of-use electricity price forecasting model based on linear regression for spot clearing is as follows: (9) (10) In the formula: , , , The monthly spot market average includes the bidding range, regional market load, regional power transmission or reception, and new energy output. Regression-predicted prices and These are the lower and upper limits of the spot market price. For system boot capacity, and , respectively, are the bidding space thresholds that trigger the lower and upper limits of the price, and a and b are linear regression coefficients.

[0030] Based on the above embodiments, this embodiment will provide a detailed description of step S105: In some embodiments, based on the monthly time-of-use average output sequence of new energy sources, the spot clearing price prediction model, and the system operation status reconstruction input, the average spot price prediction considering monthly new energy output fluctuations is obtained through integrated calculation. The result is then calibrated using a prediction error probability threshold to form the final monthly spot clearing average price prediction output that takes into account the uncertainties of wind and solar power. • Based on the system operation status reconstruction input features, including new energy installed capacity, load data, external or incoming data, and system operating capacity, a reconstructed system operation status feature set is obtained; • Input the monthly time-segmented average output sequence of new energy sources and the reconstructed system operation status feature set into the spot clearing electricity price prediction model to calculate the monthly spot forecast average considering the fluctuation of new energy output; • The monthly spot forecast mean is calibrated based on the prediction error probability threshold to obtain the final monthly spot clearing average price forecast output that takes into account the uncertainty of wind and solar power.

[0031] Specifically, the formula for calculating the average spot clearing price, taking into account the uncertainty of monthly weather conditions, is as follows: (11) (12) In the formula: P(·) represents probability, This represents the probability threshold for monthly new energy forecasting error. This represents the monthly average spot price forecast.

[0032] Please refer to Figure 2 , Figure 2 This invention provides a structural block diagram of a device for predicting the monthly time-of-use average transaction price in the electricity market; the specific device may include: The monthly renewable energy output probability prediction model building module 100 is used to generate a monthly renewable energy output probability prediction model based on historical renewable energy output data and system operation constraints, and fit its cumulative probability density function through beta distribution. The coupling relationship construction module 200 is used to establish a quantitative conversion relationship between the cumulative probability density of new energy and the average price of the spot market by introducing historical beta distribution parameters based on the cumulative probability density function of new energy, using cluster analysis to determine the typical values ​​of the distribution parameters; The time-segmented average output sequence construction module 300 is used to calculate the average output of new energy in each time period within a month, based on the quantitative conversion relationship between the cumulative probability density of new energy and the average price in the spot market, combined with the statistical characteristics of historical output fluctuations, to form a time-segmented average output sequence. The 400 module for building a spot clearing time-of-use electricity price prediction model is used to establish a linear regression spot clearing electricity price prediction model based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output and system operating capacity, by screening key influencing factors through feature importance analysis, and generating the predicted value of spot clearing time-of-use electricity price. The Spot Clearing Average Price Forecasting Module 500 is used to calculate the average spot price forecast that takes into account the monthly fluctuations in new energy output based on the monthly time-segmented average output sequence of new energy, the spot clearing electricity price forecasting model, and the system operation status reconstruction input. The module then combines the forecast error probability threshold to calibrate the results and form the final monthly spot clearing average price forecast output that takes into account the uncertainties of wind and solar power.

[0033] The electricity market monthly time-of-use transaction average price prediction device of this embodiment is used to implement the aforementioned electricity market monthly time-of-use transaction average price prediction method. Therefore, the specific implementation of the electricity market monthly time-of-use transaction average price prediction device can be found in the previous embodiment section of the electricity market monthly time-of-use transaction average price prediction method. For example, the monthly new energy output probability prediction model construction module 100, the coupling relationship construction module 200, the time-of-use output average value sequence construction module 300, the spot clearing time-of-use electricity price prediction model construction module 400, and the spot clearing average price prediction module 500 are respectively used to implement steps S101, S102, S103, S104, and S105 in the above-mentioned electricity market monthly time-of-use transaction average price prediction method. Therefore, its specific implementation can be referred to the description of the corresponding embodiments, and will not be repeated here.

[0034] A specific embodiment of the present invention also provides a device for predicting the monthly time-of-use average transaction price in the electricity market, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-mentioned method for predicting the monthly time-of-use average transaction price in the electricity market.

[0035] A specific embodiment of the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for predicting the average monthly time-of-use transaction price in the electricity market.

[0036] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0037] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0038] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0039] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0040] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for predicting the monthly time-of-use average transaction price in the electricity market, characterized in that, include: Based on historical renewable energy output data and system operation constraints, a monthly renewable energy output probability prediction model is generated, and its cumulative probability density function is fitted by beta distribution. Based on the cumulative probability density function of new energy, historical beta distribution parameters are introduced, and cluster analysis is used to determine the typical values ​​of the distribution parameters, thus establishing a quantitative conversion relationship between the cumulative probability density of new energy and the average price of the spot market. Based on the quantitative conversion relationship between the cumulative probability density of new energy and the average price in the spot market, and combined with the statistical characteristics of historical power output fluctuations, the average power output of new energy in each period of the month is calculated to form a time-segmented average power output sequence. Based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output and system operating capacity, key influencing factors are screened through feature importance analysis, and a linear regression spot clearing electricity price prediction model is established to generate spot clearing time-of-use electricity price prediction values. Based on the monthly time-of-day average output sequence of new energy sources, the spot clearing price prediction model, and the system operation status reconstruction input, the average spot price prediction considering the monthly fluctuation of new energy output is obtained through integrated calculation. The result is then calibrated by combining the prediction error probability threshold to form the final monthly spot clearing average price prediction output that takes into account the uncertainty of wind and solar power.

2. The method for predicting the monthly time-of-use average transaction price in the electricity market according to claim 1, characterized in that, The process of generating a monthly renewable energy output probability prediction model based on historical renewable energy output data and system operation constraints, and fitting its cumulative probability density function using a beta distribution, includes: Based on historical renewable energy output data and system operation constraints including renewable energy installed capacity constraints, probability distribution constraints, and wind and solar resource constraints, a monthly renewable energy output probability density function is constructed to generate a monthly renewable energy output probability prediction model. Integrating the monthly output probability density function of new energy sources yields the cumulative distribution function, which is then fitted using a beta distribution to obtain the fitted cumulative probability density function.

3. The method for predicting the monthly time-of-use average transaction price in the electricity market according to claim 2, characterized in that, The new energy installed capacity constraint variables include the new energy installed capacity in time period t and the maximum simultaneous rate of new energy in the region or station; the probability distribution constraint variables include the standard deviation of the logarithm of the new energy output variable, the mean of the logarithm of the new energy output probability density variable, the standard deviation of the logarithm of the new energy output probability density variable, the maximum and minimum output of new energy during normal operation, and the lower limit of the new energy output probability density distribution; the wind and solar resource constraint variables include the total area of ​​the solar photovoltaic array, the actual irradiance, the efficiency of the solar photovoltaic system, the effective wind energy density, the air density, and the wind speed.

4. The method for predicting the monthly time-of-use average transaction price in the electricity market according to claim 1, characterized in that, The process of establishing a quantitative conversion relationship between the cumulative probability density function of new energy sources, introducing historical beta distribution parameters, using cluster analysis to determine typical values ​​of the distribution parameters, and establishing the quantitative conversion relationship between the cumulative probability density function of new energy sources and the average spot market price includes: Historical beta distribution parameters are introduced as candidate variables, and combined with the current monthly new energy resource status data, a parameter candidate set is constructed to generate a beta distribution parameter candidate set; Based on the candidate set of beta distribution parameters, the K-means clustering method is used to classify the candidate parameters. After removing outliers, the mean of each class of parameters is taken as the typical value of the parameter, thus generating typical values ​​of the beta distribution parameters. Based on the typical values ​​of beta distribution parameters and the cumulative probability density function of new energy sources, a quantitative conversion relationship between the cumulative probability density of new energy sources and the average price in the spot market is established.

5. The method for predicting the monthly time-of-use average transaction price in the electricity market according to claim 1, characterized in that, The quantitative conversion relationship between the cumulative probability density of new energy sources and the average price in the spot market, combined with the statistical characteristics of historical power output fluctuations, calculates the average power output of new energy sources in each period of the month, forming a time-segmented average power output sequence, including: Based on the historical monthly time-period output probability distribution standard deviation mean data, calculate the historical time-period output fluctuation standard deviation mean; Based on the quantitative conversion relationship between the cumulative probability density of new energy sources and the average price in the spot market, and the standard deviation of historical time-sharing power output fluctuations, the average power output of new energy sources at each moment within a month is calculated, generating a time-sharing power output average sequence.

6. The method for predicting the monthly time-of-use average transaction price in the electricity market according to claim 1, characterized in that, Based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output, and system operating capacity, a linear regression spot clearing electricity price prediction model is established by screening key influencing factors through feature importance analysis, generating time-of-use (TOU) spot clearing electricity price predictions, including: Based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, new energy output and system operating capacity, the feature importance analysis function in the linear regression model is used to screen key features that have a significant impact on the spot clearing average price. Based on key features, a multiple linear regression model is established to fit the relationship between the average spot clearing price and each feature, thereby generating a spot clearing electricity price prediction model. Real-time feature data is input into the spot clearing electricity price prediction model to calculate the time-of-use electricity price prediction value for spot clearing.

7. The method for predicting the monthly time-of-use average transaction price in the electricity market according to claim 1, characterized in that, The method, based on the monthly time-of-day average output sequence of new energy sources, the spot clearing electricity price prediction model, and the system operation status reconstruction input, integrates calculations to obtain the average spot price prediction considering monthly new energy output fluctuations. The result is then calibrated using a prediction error probability threshold to form the final monthly spot clearing average price prediction output that takes into account the uncertainties of wind and solar power. Based on the system operation status reconstruction input features, including new energy installed capacity, load data, external or incoming data, and system operating capacity, a reconstructed system operation status feature set is obtained. The monthly average output of new energy sources by time period and the reconstructed system operation status feature set are input into the spot clearing electricity price prediction model to calculate the monthly spot forecast average considering the fluctuation of new energy output. The monthly spot price forecast average is calibrated based on the prediction error probability threshold to obtain the final monthly spot clearing average price forecast output that takes into account the uncertainty of wind and solar power.

8. A device for predicting the monthly time-of-use average transaction price in the electricity market, characterized in that, include: The monthly renewable energy output probability prediction model building module is used to generate a monthly renewable energy output probability prediction model based on historical renewable energy output data and system operation constraints, and fit its cumulative probability density function through beta distribution. The coupling relationship construction module is used to establish a quantitative conversion relationship between the cumulative probability density of new energy and the average price of the spot market, based on the cumulative probability density function of new energy, by introducing historical beta distribution parameters, using cluster analysis to determine typical values ​​of the distribution parameters; The time-segmented average output sequence construction module is used to calculate the average output of new energy in each time period within a month, based on the quantitative conversion relationship between the cumulative probability density of new energy and the average price in the spot market, combined with the statistical characteristics of historical output fluctuations, to form a time-segmented average output sequence. The spot clearing time-of-use electricity price prediction model construction module is used to establish a linear regression spot clearing electricity price prediction model based on multi-dimensional features including historical clearing average price, bidding space, load, external transmission, renewable energy output and system operating capacity, by screening key influencing factors through feature importance analysis, and generating the spot clearing time-of-use electricity price prediction value. The spot clearing average price prediction module is used to calculate the average spot price prediction that takes into account the monthly fluctuations in new energy output based on the monthly time-segmented average output sequence of new energy, the spot clearing electricity price prediction model, and the system operation status reconstruction input. The module then combines the prediction error probability threshold to calibrate the results and form the final monthly spot clearing average price prediction output that takes into account the uncertainties of wind and solar power.

9. A device for predicting the monthly time-of-use average transaction price in the electricity market, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the method for predicting the monthly time-of-use average transaction price in the electricity market as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method for predicting the monthly time-of-use average transaction price in the electricity market as described in any one of claims 1 to 7.

Images