Ultra-short-term electricity price prediction method and device for electricity market

Abstract

The invention provides an ultra-short-term electricity price prediction method and device for an electricity market, and relates to the technical field of data processing, and the method comprises the steps: obtaining historical electricity price data of one or more analysis periods, and carrying out the time stamp extraction of the historical electricity price data, and obtaining time feature data; obtaining supplementary prediction data, and determining a first prediction value according to the supplementary prediction data, the time characteristic data and the historical electricity price data; performing feature extraction on the historical electricity price data of each period to obtain corresponding feature electricity price data, and determining a second prediction value according to the supplementary prediction data, the time feature data and the feature electricity price data; and obtaining electricity price prediction data based on the first prediction value and the second prediction value, thereby solving the technical problem of inaccurate ultra-short-term electricity price prediction in the prior art.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for predicting ultra-short-term electricity prices in an electricity market. Background Art

[0002] In electricity market transactions, accurately predicting short-term and ultra-short-term electricity prices is a key task. Short-term and ultra-short-term electricity prices are the basis for formulating quotations, transactions and electricity management strategies. Accurate electricity price forecasts can not only optimize power dispatch and resource allocation, but also significantly improve profit margins. By accurately predicting electricity prices, some companies can store electricity when prices are low and sell it when prices are high, thereby implementing a strategy of buying low and selling high to maximize profits. In addition, accurate electricity price forecasts can also reduce companies' trading risks in the spot market and avoid economic losses due to market price fluctuations.

[0003] At present, a lot of research has been carried out on the short-term electricity price forecasting methods in the power market, mainly focusing on time series models and neural networks. Since the reduction of time intervals will increase the uncertainty of input data, there is a lack of high-accuracy models for ultra-short-term electricity price forecasts, and the forecast results are unsatisfactory. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the first purpose of this application is to propose an ultra-short-term electricity price forecasting method for the power market to achieve accurate ultra-short-term electricity price forecasting.

[0006] The second objective of the present application is to propose an ultra-short-term electricity price prediction device for the electricity market.

[0007] The third objective of the present application is to provide an electronic device.

[0008] A fourth objective of the present application is to provide a computer-readable storage medium.

[0009] A fifth object of the present application is to provide a computer program product.

[0010] To achieve the above objectives, the first embodiment of the present application proposes an ultra-short-term electricity price forecasting method for the electricity market, comprising:

[0011] Acquire historical electricity price data of one or more analysis periods, and extract timestamps from the historical electricity price data to obtain time characteristic data;

[0012] Acquire supplementary forecast data, and determine a first forecast value according to the supplementary forecast data, the time characteristic data and the historical electricity price data, wherein the supplementary forecast data at least includes load data and meteorological data;

[0013] Extracting features from the historical electricity price data of each period to obtain corresponding characteristic electricity price data, and determining a second prediction value based on the supplementary prediction data, the time characteristic data and the characteristic electricity price data;

[0014] Based on the first prediction value and the second prediction value, electricity price prediction data is obtained.

[0015] To achieve the above-mentioned purpose, the second embodiment of the present application proposes an ultra-short-term electricity price prediction device for the electricity market, comprising:

[0016] A first acquisition module is used to acquire historical electricity price data of one or more analysis periods, and extract timestamps from the historical electricity price data to obtain time characteristic data;

[0017] a second acquisition module, configured to acquire supplementary forecast data, and determine a first forecast value according to the supplementary forecast data, the time characteristic data and the historical electricity price data, wherein the supplementary forecast data at least includes load data and meteorological data;

[0018] a third acquisition module, configured to extract features from the historical electricity price data of each period to obtain corresponding characteristic electricity price data, and determine a second prediction value according to the supplementary prediction data, the time characteristic data and the characteristic electricity price data;

[0019] A prediction module is used to obtain electricity price prediction data based on the first prediction value and the second prediction value.

[0020] To achieve the above-mentioned purpose, the third aspect of the present application provides an electronic device, including:

[0021] A processor, and a memory communicatively connected to the processor;

[0022] The memory stores computer-executable instructions;

[0023] The processor executes the computer-executable instructions stored in the memory to implement the method described in the embodiment of the first aspect.

[0024] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present application proposes a computer-readable storage medium, in which computer-readable storage medium is stored computer execution instructions, and when the computer execution instructions are executed by a processor, they are used to implement the method described in the first aspect embodiment.

[0025] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0026] The ultra-short-term electricity price forecasting method and device for the power market provided by the present application obtain a first forecast value through historical electricity price data, time characteristic data and supplementary forecast data, extract characteristic values ​​from the historical electricity price data to obtain characteristic electricity price data, obtain a second forecast value from the characteristic electricity price data, time characteristic data and supplementary forecast data, and finally obtain electricity price forecast data based on the first forecast value and the second forecast value, comprehensively consider the time characteristics and supplementary forecast data, and obtain the forecast electricity price in combination with the second forecast value corresponding to the characteristic electricity price, thereby solving the current problem of inaccurate ultra-short-term electricity price forecasting, refining the forecast electricity price corresponding to each time characteristic, and improving the accuracy of ultra-short-term electricity price forecasting.

[0027] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0029] Figure 1 A flow chart of a method for predicting ultra-short-term electricity prices in an electricity market provided in an embodiment of the present application;

[0030] Figure 2 A schematic diagram of a process for obtaining time characteristic data provided by an embodiment of the present application;

[0031] Figure 3 A schematic diagram of a process for obtaining characteristic electricity price data provided in an embodiment of the present application;

[0032] Figure 4 A flow chart of another method for predicting ultra-short-term electricity prices in the electricity market provided in an embodiment of the present application;

[0033] Figure 5 A logic block diagram of an ultra-short-term electricity price prediction method for the power market provided in an embodiment of the present application;

[0034] Figure 6 A schematic diagram of the structure of an ultra-short-term electricity price forecasting device for an electricity market provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0036] The following describes the ultra-short-term electricity price prediction method and device for the power market in an embodiment of the present application with reference to the accompanying drawings.

[0037] Figure 1 This is a flow chart of a method for predicting ultra-short-term electricity prices in the power market provided by an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0038] S101, obtaining historical electricity price data of one or more analysis periods, and extracting timestamps from the historical electricity price data to obtain time characteristic data.

[0039] Optionally, the analysis period may be a time period such as one day, one week or one month. In this embodiment, one day is taken as one analysis period, and obtaining historical electricity price data for one or more analysis periods means obtaining historical electricity price data for one or more days.

[0040] In some implementations, historical electricity price data can be obtained from the power trading center in the analysis area; taking into account the fluctuation of electricity prices, this embodiment obtains historical electricity prices in a time period of every 15 minutes, that is, obtains historical electricity prices every 15 minutes in a day to obtain corresponding historical electricity price data.

[0041] It is understandable that historical electricity price data for multiple days can be obtained from the power trading center. The time interval between each two historical electricity prices in the daily historical electricity price data is 15 minutes, so 96 historical electricity prices can be obtained in one day. The timestamps corresponding to the historical electricity prices are 0:15, 0:30, 0:45, ..., 23:45, and 24:00 (0:00).

[0042] In some implementations, each timestamp can be marked in sequence according to the time, for example, 0:00-0:15 is marked as 0, 0:15-0:30 is marked as 1, 0:30-0:45 is marked as 2, and so on, 23:45-24:00 is marked as 95, and the corresponding time characteristic data is obtained by the marking value of each timestamp, or the marking value is mapped according to the mapping rule to obtain the corresponding time characteristic data, and the time characteristic data is used to characterize the time characteristics of each electricity price data to improve the accuracy of electricity price prediction.

[0043] S102, obtaining supplementary prediction data, and determining a first prediction value according to the supplementary prediction data, time characteristic data and historical electricity price data.

[0044] The supplementary forecast data includes at least load data and meteorological data. It can be understood that the supplementary forecast data is data used to assist in electricity price forecasting, which is data obtained from load and environmental forecasts for the next analysis period. Electricity price forecasting is based on the supplementary forecast data, which comprehensively considers the impact of other factors on electricity prices, thereby improving the accuracy of electricity price forecasting.

[0045] In some implementations, load data may include civil load, commercial load, or network-wide load data, and meteorological data may include temperature, wind speed, rainfall, and other data, without specific limitation; the data collection interval in the supplementary forecast data is consistent with the time interval of historical electricity price data, that is, the supplementary forecast data is obtained with a collection interval of 15 minutes, and the supplementary forecast data can be obtained from the local power trading center and related databases.

[0046] Optionally, the first prediction value can be obtained based on a pre-trained first prediction model, that is, the supplementary prediction data, time characteristic data and historical electricity price data are input into the pre-trained first prediction model, and the first prediction value of the electricity price is output; the pre-trained first prediction model can be a prediction model commonly used in the electricity price prediction market.

[0047] S103, extracting features from the historical electricity price data of each period to obtain corresponding characteristic electricity price data, and determining a second prediction value based on the supplementary prediction data, the time characteristic data and the characteristic electricity price data.

[0048] It is understandable that the historical electricity price data of each cycle includes historical electricity prices with 96 timestamps. The characteristic values ​​in the historical electricity price data can be obtained, and the historical electricity prices can be processed based on the characteristic values ​​to obtain the corresponding characteristic electricity prices, thereby generating characteristic electricity price data of the historical electricity price data.

[0049] As an example, 96 timestamps are equally divided into multiple time groups, and the mean or median of the historical electricity prices in each time group is calculated as the characteristic value of the corresponding time group. Each historical electricity price is processed according to the characteristic value, for example, the ratio of the historical electricity price to the characteristic value is calculated as the corresponding characteristic electricity price, so as to obtain the characteristic electricity price data of all historical electricity price data.

[0050] Furthermore, a pre-trained second prediction model can be obtained, and the supplementary prediction data, time characteristic data and characteristic electricity price data can be input into the pre-trained second prediction model to output a second prediction value of the electricity price. The second prediction model can be a pre-trained prediction model of the characteristic electricity price, that is, the second prediction model is a prediction of the characteristic electricity price, and the obtained second prediction value is the characteristic electricity price.

[0051] S104, obtaining electricity price prediction data based on the first prediction value and the second prediction value.

[0052] Optionally, the product of the first prediction value and the second prediction value may be calculated as prediction data. In other embodiments, the first prediction value and the second prediction value may be fused in other ways, which will not be described in detail.

[0053] In this embodiment, time characteristic data of historical electricity price data and supplementary forecast data are obtained, and auxiliary forecasts are performed with the supplementary forecast data and time characteristic data to obtain a first forecast value. The first forecast value takes into account the influence of time characteristics and other supplementary forecast data, and has a better forecast effect. Further feature extraction is performed on the historical electricity price data to determine the corresponding characteristic electricity price data, and a secondary forecast is performed with the characteristic electricity price data, time characteristic data and supplementary forecast data to obtain a second forecast value. The final electricity price forecast data is determined by combining the first forecast value and the second forecast value. This embodiment can refine the predicted electricity price corresponding to each time characteristic, and improve the accuracy of ultra-short-term electricity price forecasts.

[0054] Based on the above embodiments, Figure 2 A schematic diagram of a process for obtaining time characteristic data provided by an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0055] S201, assigning values ​​to each time stamp of historical electricity price data of an analysis period to obtain a value of each time stamp.

[0056] It can be understood that, in this embodiment, one analysis cycle is 1 day, and the historical electricity price data within 1 day is data obtained with a timestamp at intervals of 15 minutes.

[0057] Optionally, this embodiment assigns values ​​to each timestamp in chronological order, wherein 0:15 is assigned a value of 0, 0:30 is assigned a value of 1, 0:45 is assigned a value of 2, and so on, and 24:00 is assigned a value of 95, thereby obtaining a value corresponding to each timestamp.

[0058] S202, mapping the value of each timestamp according to a mapping rule to obtain a corresponding mapping value.

[0059] Optionally, in this embodiment, the mapping rule may be to equidistantly map the values ​​of each timestamp to an interval of [0, 2π) to obtain a corresponding mapping value. For example, the mapping rule may be expressed as: Wherein, A is the value set of the timestamp, A = {0, 1, ..., 95}, a is any value in the A set; C represents the set of corresponding mapping values, c is the mapping value in C, R is a real number set, and N is a natural number set.

[0060] S203, processing the mapping value based on the preset first function and second function to obtain a first function value and a second function value, and determining the time feature data according to the first function value and the second function value.

[0061] Optionally, the first function may be a cosine function, and the second function may be a sinine function, so the first function value and the second function value may be respectively expressed as:

[0062]

[0063]

[0064] Among them, f1 represents the first function value; f2 represents the second function value. The value range of the first function value and the second function value is in [-1,1], and the 24-hour cyclicity is retained, which is convenient for capturing data patterns in units of days. According to the first function value and the second function value of all mapping values, time characteristic data is constructed for auxiliary prediction, thereby improving the accuracy of electricity price prediction.

[0065] In this embodiment, each timestamp of the historical electricity price data is mapped to obtain a corresponding mapping value, and the mapping value is processed to obtain the time characteristic data corresponding to each timestamp, thereby retaining the 24-hour cyclicity, facilitating the capture of data patterns in units of days, and increasing the accuracy of electricity price predictions at each time point based on the time characteristic data.

[0066] Based on the above embodiments, Figure 3 A schematic diagram of a process for obtaining characteristic electricity price data provided by an embodiment of the present application. Figure 3 As shown, the method comprises the following steps:

[0067] S301, grouping historical electricity price data within a fixed time range to obtain one or more electricity price data groups.

[0068] For example, this embodiment groups historical electricity price data using 1 hour as a fixed time range, that is, the historical electricity price data collected at 15-minute intervals are divided into 24 electricity price data groups based on 1 hour, and each electricity price data group includes 4 historical electricity price data.

[0069] S302, obtaining the average electricity price corresponding to each electricity price data group.

[0070] The average value of the four electricity price data in each electricity price data group is calculated to obtain the average electricity price corresponding to the electricity price data group. The calculation of the average electricity price can be expressed as:

[0071]

[0072] ...

[0074]

[0075] in, represents the average electricity price of the first electricity price data group in the i-th analysis period. In this embodiment, one analysis period is one day; represents the mean electricity price of the second electricity price data group in the i-th analysis period; p i,1 represents the first electricity price data in the i-th analysis period; and so on, p i,96 represents the 96th electricity price data in the i-th analysis period; Represents the mean electricity price of the 24th electricity price data group in the i-th analysis period.

[0076] S303, determining the characteristic electricity price of each electricity price data in the corresponding electricity price data group according to the average electricity price, so as to obtain the characteristic electricity price data of the historical electricity price data.

[0077] Optionally, the ratio of each electricity price data in the electricity price data group to the average electricity price corresponding to the electricity price data group can be obtained, and the ratio is used as the characteristic electricity price of the electricity price data; the characteristic electricity price can be expressed as:

[0078]

[0079] Among them, u i,j represents the jth characteristic electricity price in the i-th analysis period; p i,j represents the jth electricity price data of the i-th analysis period; represents the mean electricity price of the mth electricity price data group in the ith analysis period, To round up, for example, when * is 2.3, the result of rounding up is 3.

[0080] After determining the characteristic electricity price corresponding to each historical electricity price data, the characteristic electricity price data corresponding to the historical electricity price data is obtained.

[0081] In this embodiment, the historical electricity price data is grouped every hour, and the characteristic electricity price is obtained for each timestamp based on the average electricity price in the electricity price data, so as to refine the electricity price at each time point, and realize ultra-short-term electricity price forecast in combination with the characteristic electricity price, thereby reducing the deviation caused by direct prediction through historical electricity price data and improving the accuracy of ultra-short-term electricity price forecast.

[0082] Based on the above embodiments, Figure 4 This is a flow chart of another method for predicting ultra-short-term electricity prices in the power market provided by the embodiment of the present application. Figure 4 As shown, the method comprises the following steps:

[0083] S401, obtaining historical electricity price data of one or more analysis periods.

[0084] In the embodiment of the present application, the implementation method of step S401 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0085] S402, assigning values ​​to each time stamp of the historical electricity price data of an analysis period to obtain the value of each time stamp.

[0086] In the embodiment of the present application, the implementation method of step S402 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0087] S403, mapping the values ​​of each time stamp according to a mapping rule to obtain the values ​​of each time point.

[0088] In the embodiment of the present application, the implementation method of step S403 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0089] S404, processing the mapping value based on the preset first function and second function to obtain a first function value and a second function value, and determining time feature data according to the first function value and the second function value.

[0090] In the embodiment of the present application, the implementation method of step S404 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0091] S405, obtaining supplementary prediction data, and determining a first prediction value according to the supplementary prediction data, time characteristic data and historical electricity price data.

[0092] In some implementations, considering that the meteorological data in the supplementary forecast data may not be collected every 15 minutes, in order to ensure the output effect of the model, the meteorological data in the supplementary forecast data is upsampled to obtain the upsampled supplementary forecast data, wherein the specific method of data upsampling is not limited. In this embodiment, linear interpolation is used for filling, and other interpolation methods may be used in other embodiments, which can be expressed as:

[0093]

[0094] Where x is the index value in [x i ,x i+1 ] Any interpolation point in the interval, the coordinates of the point are (x, y), [x i ,x i+1 ]The endpoints of the interval are (xi ,y i ) and (x i+1 ,y i+1 ); add indexes to the data in chronological order for digitization, (x j , j∈N) is the interpolation index, (y j , j∈N) is the data point to be upsampled.

[0095] The meteorological data is interpolated for data gaps using a linear interpolation method. It is understandable that meteorological data may include data of different dimensions such as temperature and precipitation, and the data interpolation methods of each dimension may be the same or different, but the interpolation method used in actual prediction is consistent with the preprocessing interpolation method.

[0096] Optionally, a pre-trained first prediction model may be obtained; the supplementary prediction data, time characteristic data, and historical electricity price data are input into the first prediction model to obtain a first prediction value. In some implementations, the first prediction model is a time-of-use electricity price prediction model commonly used in the existing electricity market, and the supplementary prediction data, event characteristic data, and historical electricity price data are input into the first prediction model to achieve a relatively accurate electricity price prediction and obtain a first prediction value.

[0097] It can be understood that the first prediction value is the time-sharing predicted electricity price output by the time-sharing electricity price prediction model, that is, the electricity price prediction result per hour, and the electricity price prediction result per hour is the average value of all predicted electricity prices within the corresponding hour.

[0098] In the embodiment of the present application, the implementation method of step S405 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0099] S406, grouping the historical electricity price data in a fixed time range to obtain one or more electricity price data groups.

[0100] In the embodiment of the present application, the implementation method of step S406 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0101] S407, obtaining the average electricity price corresponding to each electricity price data group.

[0102] In the embodiment of the present application, the implementation method of step S407 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0103] S408, determining the characteristic electricity price of each electricity price data in the corresponding electricity price data group according to the average electricity price, so as to obtain the characteristic electricity price data of the historical electricity price data.

[0104] In the embodiment of the present application, the implementation method of step S408 can be implemented by any of the methods in the embodiments of the present disclosure, which is not limited here and will not be repeated.

[0105] S409, determining a second prediction value according to the supplementary prediction data, the time characteristic data and the characteristic electricity price data.

[0106] Optionally, a pre-trained second prediction model may be obtained; the supplementary prediction data, time characteristic data and characteristic electricity price data are input into the second prediction model to obtain a second prediction value; it is understandable that the second prediction value is the predicted characteristic electricity price.

[0107] In some implementations, a pre-trained first prediction model and sample characteristic electricity price data may be obtained; the first prediction model may be optimized based on the sample characteristic electricity price data to obtain a pre-trained second prediction model.

[0108] In some implementations, the sample characteristic electricity price data can be calculated from the sample electricity data obtained from the power trading center. Therefore, the pre-trained first prediction model can be optimized according to the sample characteristic electricity price data, and the model parameters are adjusted by back-propagating the difference between the true value and the predicted value in the sample characteristic data to obtain the trained first prediction model, that is, the second prediction model, which outputs the predicted characteristic electricity price.

[0109] In some implementations, when inputting data, correlation analysis and random forest methods can be used to filter data dimensions to reduce the complexity of the input data, and the characteristic electricity price can be used as the per-unit value to participate in electricity price prediction, avoiding trend deviations that may occur in direct predictions, and improving the accuracy of electricity price predictions.

[0110] S410, obtaining electricity price prediction data based on the first prediction value and the second prediction value.

[0111] Optionally, the product of the first prediction value and the second prediction value may be calculated, and the product may be used as the corresponding electricity price prediction data; the electricity price prediction data may be expressed as: in is the electricity price forecast data corresponding to the jth timestamp; represents the first predicted value, and the jth timestamp is within the time period corresponding to the first predicted value; It represents the second predicted value, and so on, to obtain the predicted electricity price at each time stamp in the next analysis cycle, that is, the predicted electricity price every 15 minutes in the next day, to achieve ultra-short-term electricity price prediction.

[0112] In this embodiment, each time stamp of the historical electricity price data is mapped to obtain a corresponding mapping value, and the mapping value is processed to obtain time characteristic data corresponding to each time stamp, so as to capture the data regularity in units of days, increase the accuracy of the electricity price prediction at each time point based on the time characteristic data, obtain supplementary prediction data, and perform auxiliary prediction with the supplementary prediction data and the time characteristic data to obtain a first prediction value, which takes into account the influence of the time characteristics and other supplementary prediction data, and has a better prediction effect; further, the historical electricity price data is grouped every hour, and the characteristic electricity price is obtained for each time stamp according to the average electricity price in the electricity price data, so as to refine the electricity price at each time point and obtain a second prediction value, and determine the final electricity price prediction data by combining the first prediction value and the second prediction value, so as to reduce the deviation caused by direct prediction through historical electricity price data, refine the predicted electricity price corresponding to each time characteristic, and improve the accuracy of ultra-short-term electricity price prediction.

[0113] Figure 5 A logic block diagram of an ultra-short-term electricity price forecasting method for an electricity market provided in an embodiment of the present application; both supplementary forecast data and historical electricity price data are taken as input, and the historical electricity price data and the supplementary forecast data are preprocessed respectively, that is, time feature extraction and data upsampling are performed, and the sample electricity price data are subjected to the same processing for model training, and the characteristic electricity price is used to participate in model training, and a second forecasting model, that is, a characteristic electricity price forecasting model, is trained on the basis of a time-of-use electricity price forecasting model (first forecasting model), and finally the time-of-use electricity price forecasting result based on the first forecasting model is fused with the characteristic electricity price forecasting result of the characteristic electricity price forecasting model to obtain the electricity price forecasting result for the next analysis period, thereby improving the accuracy of electricity price forecasting.

[0114] In order to implement the above embodiment, the present application also proposes an electricity price prediction device.

[0115] Figure 6 This is a schematic diagram of the structure of an ultra-short-term electricity price prediction device for the power market provided in an embodiment of the present application. Figure 6 As shown, the ultra-short-term electricity price forecasting device 600 of the electricity market includes:

[0116] The first acquisition module 601 is used to acquire historical electricity price data of one or more analysis periods, and extract timestamps from the historical electricity price data to obtain time characteristic data;

[0117] A second acquisition module 602 is used to acquire supplementary forecast data and determine a first forecast value according to the supplementary forecast data, time characteristic data and historical electricity price data, wherein the supplementary forecast data at least includes load data and meteorological data;

[0118] The third acquisition module 603 is used to extract features from the historical electricity price data of each period to obtain corresponding characteristic electricity price data, and determine the second prediction value according to the supplementary prediction data, the time characteristic data and the characteristic electricity price data;

[0119] The prediction module 604 is used to obtain electricity price prediction data based on the first prediction value and the second prediction value.

[0120] Furthermore, in a possible implementation of the embodiment of the present application, the first acquisition module 601 includes:

[0121] Assign values ​​to each time stamp of the historical electricity price data of an analysis period to obtain the value of each time stamp;

[0122] Map the values ​​of each timestamp according to the mapping rules to obtain the corresponding mapping value;

[0123] The mapping value is processed based on the preset first function and the second function to obtain the first function value and the second function value, and the time feature data is determined according to the first function value and the second function value.

[0124] Furthermore, in a possible implementation of the embodiment of the present application, the second acquisition module 602 includes:

[0125] Obtain a pre-trained first prediction model;

[0126] The supplementary prediction data, time characteristic data and historical electricity price data are input into the first prediction model to obtain a first prediction value.

[0127] Further, in a possible implementation of the embodiment of the present application, the third acquisition module 603 includes:

[0128] Grouping historical electricity price data in a fixed time range to obtain one or more electricity price data groups;

[0129] Obtain the average electricity price corresponding to each electricity price data group;

[0130] The characteristic electricity price of each electricity price data in the corresponding electricity price data group is determined according to the average electricity price to obtain the characteristic electricity price data of the historical electricity price data.

[0131] Further, in a possible implementation of the embodiment of the present application, the third acquisition module 603 includes:

[0132] The ratio of each electricity price data in the electricity price data group to the corresponding electricity price average of the electricity price data group is obtained, and the ratio is used as the characteristic electricity price of the electricity price data.

[0133] Further, in a possible implementation of the embodiment of the present application, the third acquisition module 603 includes:

[0134] Obtain a pre-trained second prediction model;

[0135] The supplementary prediction data, time characteristic data and characteristic electricity price data are input into the second prediction model to obtain a second prediction value.

[0136] Further, in a possible implementation of the embodiment of the present application, the third acquisition module 603 includes:

[0137] Obtaining a pre-trained first prediction model and sample characteristic electricity price data;

[0138] The first prediction model is optimized based on the sample characteristic electricity price data to obtain a pre-trained second prediction model.

[0139] Furthermore, in a possible implementation of the embodiment of the present application, the prediction module 604 includes:

[0140] The product of the first prediction value and the second prediction value is calculated, and the product is used as the corresponding electricity price prediction data.

[0141] Furthermore, in a possible implementation of the embodiment of the present application, the apparatus 600 further includes:

[0142] Upsampling the meteorological data in the supplementary forecast data to obtain upsampled supplementary forecast data;

[0143] The first prediction value and the second prediction value are acquired according to the up-sampled supplementary prediction data.

[0144] It should be noted that the above explanation of the embodiment of the method for ultra-short-term electricity price forecasting in the electricity market is also applicable to the ultra-short-term electricity price forecasting device in the electricity market of this embodiment, and will not be repeated here.

[0145] In an embodiment of the present application, each time stamp of the historical electricity price data is mapped to obtain a corresponding mapping value, and the mapping value is processed to obtain time characteristic data corresponding to each time stamp, so as to capture the data regularity in units of days, increase the accuracy of the electricity price prediction at each time point based on the time characteristic data, obtain supplementary prediction data, and use the supplementary prediction data and the time characteristic data for auxiliary prediction to obtain a first prediction value, which takes into account the influence of time characteristics and other supplementary prediction data, and has a better prediction effect; further, the historical electricity price data is grouped every hour, and the characteristic electricity price is obtained for each time stamp according to the average electricity price in the electricity price data, so as to refine the electricity price at each time point and obtain a second prediction value, and determine the final electricity price prediction data by combining the first prediction value and the second prediction value, thereby reducing the deviation caused by direct prediction through historical electricity price data, refining the predicted electricity price corresponding to each time characteristic, and improving the accuracy of ultra-short-term electricity price prediction.

[0146] In order to implement the above embodiments, the present application also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.

[0147] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0148] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0149] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0150] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0151] The present application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, the present disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0152] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0153] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0154] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0155] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0156] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0157] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0158] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0159] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for predicting ultra-short-term electricity prices in the power market, characterized in that: The following steps are involved: Acquire historical electricity price data of one or more analysis periods, and extract timestamps from the historical electricity price data to obtain time characteristic data; Acquire supplementary forecast data, and determine a first forecast value according to the supplementary forecast data, the time characteristic data and the historical electricity price data, wherein the supplementary forecast data at least includes load data and meteorological data; Extracting features from the historical electricity price data of each period to obtain corresponding characteristic electricity price data, and determining a second prediction value based on the supplementary prediction data, the time characteristic data and the characteristic electricity price data; Based on the first prediction value and the second prediction value, electricity price prediction data is obtained.

2. The method according to claim 1, characterized in that The extracting of timestamps from the historical electricity price data to obtain time characteristic data includes: Assign values ​​to each time stamp of the historical electricity price data of an analysis period to obtain the value of each time stamp; Map the values ​​of each timestamp according to the mapping rules to obtain the corresponding mapping value; The mapping value is processed based on a preset first function and a second function to obtain a first function value and a second function value, and time feature data is determined according to the first function value and the second function value.

3. The method according to claim 1 or 2, characterized in that: The determining of the first prediction value according to the supplementary prediction data, the time characteristic data and the historical electricity price data comprises: Obtain a pre-trained first prediction model; The supplementary prediction data, the time characteristic data and the historical electricity price data are input into the first prediction model to obtain the first prediction value.

4. The method according to claim 1, characterized in that: The feature extraction of the historical electricity price data of each cycle to obtain corresponding feature electricity price data includes: Grouping the historical electricity price data in a fixed time range to obtain one or more electricity price data groups; Obtaining the average electricity price corresponding to each of the electricity price data groups; The characteristic electricity price corresponding to each electricity price data in the electricity price data group is determined according to the electricity price average value to obtain the characteristic electricity price data of the historical electricity price data.

5. The method according to claim 4, characterized in that The step of determining the characteristic electricity price corresponding to each electricity price data in the electricity price data group according to the electricity price average includes: The ratio of each electricity price data in the electricity price data group to the electricity price average corresponding to the electricity price data group is obtained, and the ratio is used as the characteristic electricity price of the electricity price data.

6. The method according to claim 4, characterized in that The determining of the second prediction value according to the supplementary prediction data, the time characteristic data and the characteristic electricity price data comprises: Obtain a pre-trained second prediction model; The supplementary prediction data, the time characteristic data and the characteristic electricity price data are input into the second prediction model to obtain the second prediction value.

7. The method according to claim 6, characterized in that The obtaining of the pre-trained second prediction model comprises: Obtaining a pre-trained first prediction model and sample characteristic electricity price data; The first prediction model is optimized based on the sample characteristic electricity price data to obtain a pre-trained second prediction model.

8. The method according to claim 1, characterized in that The obtaining electricity price prediction data based on the first prediction value and the second prediction value includes: Calculate the product of the first prediction value and the second prediction value, and use the product as the corresponding electricity price prediction data.

9. The method according to claim 1, characterized in that: The method further comprises: Upsampling the meteorological data in the supplementary forecast data to obtain upsampled supplementary forecast data; The first prediction value and the second prediction value are acquired according to the up-sampled supplementary prediction data.

10. An ultra-short-term electricity price forecasting device for the power market, characterized in that: include: A first acquisition module is used to acquire historical electricity price data of one or more analysis periods, and extract timestamps from the historical electricity price data to obtain time characteristic data; a second acquisition module, configured to acquire supplementary forecast data, and determine a first forecast value according to the supplementary forecast data, the time characteristic data and the historical electricity price data, wherein the supplementary forecast data at least includes load data and meteorological data; a third acquisition module, configured to extract features from the historical electricity price data of each period to obtain corresponding characteristic electricity price data, and determine a second prediction value according to the supplementary prediction data, the time characteristic data and the characteristic electricity price data; A prediction module is used to obtain electricity price prediction data based on the first prediction value and the second prediction value.

Images