Short-term electricity price prediction method and device based on per unit, and electronic equipment
By acquiring and analyzing historical electricity price data, calculating the electricity price benchmark value and standard value, and combining load data to predict, the problem of low accuracy of short-term electricity price prediction is solved, the accuracy and efficiency of electricity price prediction is improved, and the power scheduling and Resource configuration.
Patent Information
- Application Number
- CN202411166207.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-12-27
AI Technical Summary
The accuracy of the short-term electricity price prediction results in the prior art is low, which affects the efficiency of power scheduling and resource allocation.
By obtaining the electricity price data within n days of history, calculating the electricity price benchmark value and unitary value, and making predictions based on these data and the load data to be predicted, obtaining the electricity price prediction benchmark value and unitary value, and finally determining the electricity price target prediction value.
It improves the accuracy and efficiency of electricity price target prediction, optimizes power scheduling and resource allocation, and reduces economic losses caused by market price fluctuations.
Smart Images
Figure CN119205189A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of electricity price prediction, and in particular to a method, device and electronic equipment for short-term electricity price prediction based on per-unit normalization. Background Art
[0002] In electricity market transactions, obtaining accurate short-term electricity price forecast results has become the basis for formulating quotation, trading and electricity management strategies. Accurate electricity price forecast results can not only optimize power dispatch and resource allocation, but also significantly improve profit margins. In related technologies, short-term electricity price forecasts are mainly achieved through time series models and artificial neural networks. However, the above methods often obtain short-term electricity price forecast results with low accuracy. Summary of the invention
[0003] The present disclosure provides a short-term electricity price prediction method, device and electronic device based on normalization, so as to at least solve the problem of low accuracy of short-term electricity price prediction results in the related art.
[0004] The technical solution of the present disclosure is as follows:
[0005] According to a first aspect of an embodiment of the present disclosure, a short-term electricity price forecasting method based on per-unit is provided, the method comprising: obtaining electricity price data within a historical period of n days, wherein n is a positive integer; obtaining an electricity price benchmark value within a historical period of n days according to the electricity price data, and obtaining a per-unit value of electricity price for each hour within the historical period of n days according to the electricity price data and the electricity price benchmark value within the historical period of n days; forecasting the electricity price benchmark value for the n+1th day based on the electricity price benchmark value and first load data for the n+1th day to be forecasted, so as to obtain a forecast electricity price benchmark value for the n+1th day; forecasting the electricity price per-unit value for the n+1th day based on the electricity price per-unit value and second load data for each hour within the n+1th day, so as to obtain a forecast electricity price per-unit value for each hour within the n+1th day; determining a target forecast electricity price value for the tth hour within the n+1th day according to the forecast electricity price benchmark value for the n+1th day and the forecast electricity price per-unit value for each hour within the n+1th day, wherein 0≤t≤24, and t is a positive integer.
[0006] According to a second aspect of an embodiment of the present disclosure, a short-term electricity price forecasting device based on per unit is provided, comprising: a first acquisition module, used to acquire electricity price data within n historical days, wherein n is a positive integer; a second acquisition module, used to acquire an electricity price benchmark value within n historical days according to the electricity price data, and to acquire a per unit value of electricity price per hour within n historical days according to the electricity price data and the electricity price benchmark value within n historical days; a first prediction module, used to predict the electricity price benchmark value and the first load data of the n+1 day to be predicted for the n+1 day; The first prediction module is used to predict the electricity price benchmark value to obtain the electricity price forecast benchmark value for the n+1th day; the second prediction module is used to predict the electricity price benchmark value for the n+1th day based on the electricity price benchmark value and the second load data for each hour in the n+1th day, so as to obtain the electricity price forecast benchmark value for each hour in the n+1th day; the determination module is used to determine the electricity price target forecast value for the tth hour in the n+1th day according to the electricity price forecast benchmark value for the n+1th day and the electricity price forecast benchmark value for each hour in the n+1th day, wherein 0≤t≤24 and t is a positive integer.
[0007] According to the fourth aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the instructions to implement the normalized short-term electricity price prediction method as described in the first aspect of the embodiment of the present disclosure.
[0008] According to the sixth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the short-term electricity price prediction method based on normalization as described in the first aspect of the embodiment of the present disclosure.
[0009] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:
[0010] In the embodiment of the present disclosure, by obtaining electricity price data within n historical days, where n is a positive integer, obtaining the electricity price benchmark value within n historical days according to the electricity price data, and obtaining the electricity price per unit value for each hour within n historical days according to the electricity price data and the electricity price benchmark value within n historical days, predicting the electricity price benchmark value for the n+1 day based on the electricity price benchmark value and the first load data for the n+1 day to be predicted, so as to obtain the electricity price prediction benchmark value for the n+1 day, predicting the electricity price per unit value for the n+1 day based on the electricity price per unit value and the second load data for each hour within the n+1 day, and The electricity price forecast per unit value is measured to obtain the hourly electricity price forecast value within the n+1th day; the electricity price target forecast value for the tth hour within the n+1th day is determined according to the electricity price forecast reference value within the n+1th day and the electricity price forecast per unit value within the n+1th day, wherein 0≤t≤24 and t is a positive integer. Thus, the present invention improves the accuracy and efficiency of obtaining the electricity price target forecast value by capturing the deep correlation between electricity price data and load data and obtaining the electricity price target forecast value based on the electricity price forecast reference value and the electricity price forecast per unit value, which is beneficial to optimizing power dispatching and resource allocation.
[0011] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.
[0013] Figure 1 The figure is a flow chart showing a method for short-term electricity price prediction based on per unitization according to an exemplary embodiment.
[0014] Figure 2 is a flowchart of a method for short-term electricity price prediction based on per-unit normalization according to another exemplary embodiment.
[0015] Figure 3 is a flowchart of a method for short-term electricity price prediction based on per-unit normalization according to another exemplary embodiment.
[0016] Figure 4 is a flowchart of a method for short-term electricity price prediction based on per-unit normalization according to another exemplary embodiment.
[0017] Figure 5 The figure is a block diagram of a short-term electricity price prediction device based on per-unit according to an exemplary embodiment.
[0018] Figure 6 is a block diagram of an electronic device according to another exemplary embodiment. DETAILED DESCRIPTION
[0019] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.
[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.
[0021] The following embodiments describe in detail the short-term electricity price prediction method, device and electronic device based on per-unit normalization proposed in the present disclosure.
[0022] Figure 1 A flowchart of a short-term electricity price prediction method based on normalization is provided in an embodiment of the present disclosure.
[0023] like Figure 1 As shown, the short-term electricity price prediction method based on normalization proposed in this embodiment includes the following steps:
[0024] S101, obtaining electricity price data within n historical days, where n is a positive integer.
[0025] It should be noted that the present disclosure does not limit the specific method for obtaining the electricity price data within the historical n days, and it can be selected according to the actual situation.
[0026] Optionally, the electricity price data within the past n days may be queried from the electricity trading platform to obtain the electricity price data within the past n days.
[0027] S102, obtaining a base value of electricity prices within n historical days based on electricity price data, and obtaining a per-unit value of electricity prices per hour within n historical days based on the electricity price data and the base value of electricity prices within n historical days.
[0028] In the embodiment of the present disclosure, after obtaining the electricity price data, the historical maximum daily electricity price and the historical daily average electricity price can be determined from the electricity price data within the historical n days, and the electricity price benchmark value within the historical n days can be obtained based on the historical maximum daily electricity price or the historical daily average electricity price and a preset benchmark coefficient.
[0029] It should be noted that the present disclosure does not limit the setting of the reference coefficient co, which can be set according to the actual situation. For example, when the reference coefficient satisfies 0 < co < 1, the floating range of the reference value can be lengthened and the difference of the per-unit value can be reduced; when the reference coefficient satisfies co = 1, the reference value is the highest daily electricity price and there is no additional impact; when the reference coefficient satisfies co > 1, the floating range of the reference value can be shortened and the difference of the per-unit value can be increased.
[0030] It should be noted that by introducing the reference coefficient co, the electricity price reference value within the historical n days and the per-unit value of the electricity price per hour within the historical n days can be affected. By setting different reference coefficients, it is beneficial to improve the accuracy of the subsequent electricity price prediction reference value and the per-unit value of the electricity price prediction.
[0031] Optionally, for the highest daily electricity price in history, the electricity price of each hour of each day in history can be obtained, and the electricity prices of each hour in history can be sorted in descending order, and the electricity price at the top can be selected as the highest daily electricity price.
[0032] For example, for the highest daily electricity price p on the i-th day i = max(p i,1 , p i,2 ,..., p i,24 ), where 1 ≤ i ≤ n, p i,1 is the electricity price of the first hour on the i-th day, p i,2 is the electricity price of the second hour on the i-th day, p i,24 is the electricity price of the 24th hour on the i-th day.
[0033] In the embodiment of the present disclosure, after obtaining the highest daily electricity prices p1, p2,..., p n in history, the electricity price reference value within the historical n days can be obtained by using the following formula: where 1 ≤ i ≤ n, b i is the electricity price reference value of the i-th day in history. By performing the above process, the electricity price reference values b1, b2,..., b n within the historical n days can be obtained, where b1 is the electricity price reference value of the first day in history, b2 is the electricity price reference value of the second day in history, and b n is the electricity price reference value of the n-th day in history.
[0034] Optionally, for the average daily electricity price in history, the electricity price of each hour of each day in history can be obtained, and the average of the electricity prices of each hour in history can be calculated, and the average value can be used as the average daily electricity price.
[0035] For example, for the average daily electricity price on the i-th day where 1 ≤ i ≤ n, p i,1 is the electricity price of the first hour on the i-th day, p i,2is the electricity price in the second hour of the i-th day, p i,24 is the electricity price at the 24th hour of the i-th day.
[0036] In the embodiment of the present disclosure, after obtaining the historical daily average electricity price After that, the following formula can be used to obtain the benchmark value of electricity prices within the past n days: Among them, 1≤i≤n, is the electricity price benchmark value for the i-th day in history. By executing the above process, the electricity price benchmark value for the n-th day in history can be obtained. in, The base value of electricity price on the first day of history. The base value of electricity price for the second day in history. It is the benchmark value of electricity price on the nth day in history.
[0037] In the embodiment of the present disclosure, the daily time-of-use electricity price is obtained, and the per-unit electricity price value for each hour in the historical n days is obtained based on the daily time-of-use electricity price and the electricity price benchmark value in the historical n days.
[0038] Among them, time-of-use electricity price refers to the different electricity prices at different times of the 24 hours in a day.
[0039] Optionally, the electricity price benchmark value for the jth day can be obtained from the electricity price benchmark values within the historical n days, where j≤n and j is a positive integer. The electricity price at the mth hour on the jth day can be obtained from the daily time-of-use electricity price, where 0≤m≤24 and m is a positive integer. The quotient between the electricity price at the mth hour on the jth day and the electricity price benchmark value of the jth day is obtained, and the quotient is used as the denominator of the electricity price at the mth hour on the jth day.
[0040] For example, based on the historical daily maximum electricity price and the preset benchmark coefficient, the electricity price benchmark value within n days can be obtained, and the following formula can be used to obtain the electricity price standard value at the mth time on the jth day: Among them, u j,m is the per unit value of the electricity price at the mth hour in the jth day, p j,m is the electricity price at the mth hour in the jth day, b j is the base value of electricity price on the jth day. By executing the above process, the per-unit electricity price value (u 1,1 ,u 1,2 ,...,u 1,24 ,...,u n,1 ,u n,2 ,...,u n,24 ).
[0041] It should be noted that regarding the specific process of obtaining the electricity price benchmark value within n historical days by combining the historical daily average electricity price with the preset benchmark coefficient, and obtaining the electricity price per unit value per hour within n historical days, please refer to the above embodiment and will not be repeated here.
[0042] S103 , based on the electricity price reference value and the first load data of the n+1 day to be predicted, predict the electricity price reference value of the n+1 day to obtain the electricity price prediction reference value of the n+1 day.
[0043] Optionally, for the first load data on the n+1th day to be predicted, the load data on the n+1th day may be queried from the power trading platform to obtain the first load data on the n+1th day to be predicted.
[0044] The first load data is the highest load forecast data of the n+1th day L = [l1,l2,...,l q ], where there are q types of first load data, such as wind power, photovoltaic power, etc.
[0045] It should be noted that the present disclosure does not limit the specific method for predicting the electricity price benchmark value for the n+1 day based on the electricity price benchmark value and the first load data for the n+1 day to be predicted to obtain the electricity price forecast benchmark value for the n+1 day, and it can be selected according to actual conditions.
[0046] Optionally, a pre-trained electricity price benchmark prediction model can be obtained, the electricity price benchmark value and the first load data can be input into the electricity price benchmark prediction model, and the electricity price benchmark value for the n+1th day can be predicted through the electricity price benchmark prediction model to obtain the electricity price prediction benchmark value for the n+1th day.
[0047] S104, based on the electricity price per unit value and the second load data for each hour on the n+1th day, predict the electricity price per unit value for the n+1th day to obtain the predicted electricity price per unit value for each hour on the n+1th day.
[0048] Optionally, for the second load data per hour on the n+1 day, the load data per hour on the n+1 day can be queried from the power trading platform to obtain the second load data per hour on the n+1 day. t =[l 1,t ,l 2,t ,...,l q,t ], where there are q types of second load data, such as wind power, photovoltaic power, etc.
[0049] It should be noted that the present invention does not limit the specific method for predicting the per-unit value of the electricity price on the n+1th day based on the per-unit value of the electricity price and the second load data every hour on the n+1th day to obtain the predicted per-unit value of the electricity price every hour on the n+1th day, and it can be selected according to actual conditions.
[0050] Optionally, a pre-trained electricity price per unit value prediction model can be obtained, the electricity price per unit value and the second load data can be input into the electricity price per unit value prediction model, and the electricity price per unit value of the n+1th day can be predicted through the electricity price per unit value prediction model to obtain the predicted electricity price per unit value for each hour in the n+1th day.
[0051] S105, determining the target electricity price forecast value for the tth hour in the n+1th day based on the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour in the n+1th day, where 0≤t≤24 and t is a positive integer.
[0052] In the embodiment of the present disclosure, after obtaining the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour in the n+1th day, the electricity price target forecast per unit value for the tth hour in the n+1th day can be determined from the electricity price forecast per unit value for each hour in the n+1th day, and the product of the electricity price forecast reference value for the n+1th day and the target forecast per unit value can be obtained, and the product can be used as the electricity price target forecast value for the tth hour in the n+1th day.
[0053] In summary, the short-term electricity price prediction method based on per unit provided by the embodiment of the present disclosure obtains electricity price data within the past n days, obtains the electricity price benchmark value within the past n days according to the electricity price data, obtains the per unit value of the electricity price for each hour within the past n days according to the electricity price data and the electricity price benchmark value within the past n days, predicts the electricity price benchmark value for the n+1 day based on the electricity price benchmark value and the first load data for the n+1 day to be predicted, so as to obtain the electricity price prediction benchmark value for the n+1 day, and predicts the electricity price benchmark value for the n+1 day based on the electricity price benchmark value and the second load data for each hour within the n+1 day. The electricity price per unit value of the n+1th day is predicted to obtain the predicted electricity price per unit value for each hour in the n+1th day; the target electricity price prediction value for the tth hour in the n+1th day is determined according to the predicted electricity price reference value of the n+1th day and the predicted electricity price per unit value for each hour in the n+1th day. Thus, the present invention improves the accuracy and efficiency of obtaining the target electricity price prediction value by capturing the deep correlation between electricity price data and load data, and obtaining the target electricity price prediction value based on the predicted electricity price reference value and the predicted electricity price per unit value, which is beneficial to optimizing power dispatching and resource allocation.
[0054] As a possible implementation, Figure 2As shown, on the basis of the above embodiment, based on the electricity price reference value and the first load data of the n+1 day to be predicted, the electricity price reference value of the n+1 day is predicted to obtain the specific process of the electricity price prediction reference value of the n+1 day, including the following steps:
[0055] S201, obtaining a pre-trained electricity price benchmark value prediction model.
[0056] It should be noted that the present disclosure does not limit the model type of the electricity price benchmark value prediction model, and it can be selected according to actual conditions.
[0057] Optionally, the electricity price benchmark value prediction model may be an artificial neural network (ANN) model, wherein the ANN model includes an input layer, a hidden layer and an output layer.
[0058] For example, a first training sample can be obtained, which includes a sample electricity price benchmark value and a first sample load data, as well as a true electricity price benchmark value corresponding to the sample electricity price benchmark value and the first sample load data. According to the training sample, an initial electricity price benchmark value prediction model is trained to obtain a trained electricity price benchmark value prediction model.
[0059] S202: Inputting the electricity price reference value and the first load data into an electricity price reference value prediction model to obtain an electricity price prediction reference value for the (n+1)th day.
[0060] For example, the electricity price reference value B = [b1, b2, ..., b n ] and the first load data L=[l1,l2,...,l q ] is input into the electricity price benchmark value prediction model to obtain the electricity price forecast benchmark value for the n+1th day
[0061]
[0062] Among them, g is the final result activation function, the mapping output is to the [0,1] interval, a and b are the number of neurons used for the electricity price benchmark value B and the first load data respectively, and the hidden layers h and k can be expanded to neural networks with different architectures.
[0063] As a possible implementation, Figure 3 As shown, on the basis of the above embodiment, based on the electricity price per unit value and the second load data per hour on the n+1 day, the electricity price per unit value on the n+1 day is predicted to obtain the specific process of the electricity price prediction per unit value per hour on the n+1 day, including the following steps:
[0064] S301, obtaining a pre-trained electricity price per unit value prediction model.
[0065] It should be noted that the present disclosure does not limit the model type of the electricity price per unit value prediction model, and it can be selected according to actual conditions.
[0066] Optionally, the electricity price per unit value prediction model may be an ANN model, wherein the ANN model comprises an input layer, a hidden layer and an output layer.
[0067] For example, a second training sample can be obtained, which includes the sample electricity price per unit value and the second sample load data and the actual per unit value of the electricity price corresponding to the sample electricity price per unit value and the second sample load data. According to the second training sample, the initial electricity price per unit value prediction model is trained to obtain a trained electricity price per unit value prediction model.
[0068] S302, inputting the per-unit value of the electricity price and the second load data into the per-unit value prediction model of the electricity price to obtain the predicted per-unit value of the electricity price for each hour on the (n+1)th day.
[0069] For example, the electricity price can be expressed as U t =[u 1,t ,u 2,t ,...,u n,t ] and the second load data L t =[l 1,t ,l 2,t ,...,l q,t ] is input into the electricity price per unit value prediction model to obtain the electricity price prediction per unit value for each hour on the n+1th day:
[0070]
[0071] Among them, g is the final result activation function, the mapping output is in the interval [0,1], x and y are the unit value of the electricity price U t and the second load data L t The number of neurons, hidden layers h, k used can be extended to neural networks with different architectures.
[0072] In summary, the short-term electricity price forecasting method based on standardization provided by the embodiment of the present disclosure obtains a pre-trained electricity price benchmark prediction model, inputs the electricity price benchmark value and the first load data into the electricity price benchmark prediction model to obtain the electricity price forecast benchmark value for the n+1th day, and obtains a pre-trained electricity price standard value prediction model, inputs the electricity price standard value and the second load data into the electricity price standard value prediction model to obtain the electricity price forecast standard value for each hour in the n+1th day. Therefore, the present disclosure can capture the deep correlation between electricity price data and load data through the pre-trained electricity price benchmark prediction model and the electricity price standard value prediction model, thereby improving the accuracy and reliability of obtaining the electricity price forecast benchmark value and the electricity price forecast standard value, and laying a foundation for the subsequent accurate acquisition of short-term electricity price forecast results.
[0073] In the embodiment of the present disclosure, after obtaining the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour in the n+1th day, the electricity price target forecast value for the tth hour in the n+1th day can be determined based on the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour in the n+1th day.
[0074] As a possible implementation, Figure 4 As shown, based on the above embodiment, according to the electricity price forecast reference value of the n+1th day and the electricity price forecast per unit value of each hour in the n+1th day, the specific process of determining the electricity price target forecast value of the tth hour in the n+1th day includes the following steps:
[0075] S401, determining the target predicted per-unit value of the electricity price at the tth hour on the n+1th day from the predicted per-unit value of the electricity price per hour on the n+1th day.
[0076] In the embodiment of the present disclosure, the target predicted per unit value of the electricity price at the tth hour on the n+1th day can be determined from the predicted per unit value of the electricity price per hour on the n+1th day.
[0077] S402, obtaining the product of the electricity price forecast reference value and the target forecast per unit value on the n+1th day, and using the product as the electricity price target forecast value for the tth hour on the n+1th day.
[0078] In the embodiment of the present disclosure, after obtaining the electricity price forecast reference value for the n+1th day, and the target predicted per unit value After that, the target electricity price forecast value for the tth hour on the n+1th day can be obtained using the following formula:
[0079]
[0080] in, is the target electricity price forecast value for the tth hour on the n+1th day, is the base value of electricity price on day n+1 and The per-unit value of the electricity price target forecast for the t-th hour on the n+1th day.
[0081] In the embodiment of the present disclosure, the above process can be used to obtain the target electricity price forecast value for each hour in the n+1 day, sort the target electricity price forecast value for each hour in the n+1 day, and determine the highest electricity price forecast value in the n+1 day according to the sorting result.
[0082] In an embodiment of the present disclosure, a user's electricity price query request can be received, wherein the query request includes a target query time and a target query area. Through the normalized short-term electricity price prediction method provided by an embodiment of the present disclosure, the electricity price at the target query time can be predicted, the electricity price prediction value can be output, and the electricity price prediction value can be fed back to the client.
[0083] Furthermore, after obtaining the electricity price forecast value, the electricity price forecast value can be stored and displayed according to different regions and times, thereby realizing the visualization of the electricity price forecast value and improving the efficiency of obtaining the electricity price forecast value.
[0084] In summary, the short-term electricity price forecasting method based on per-unit standardization provided by the embodiment of the present disclosure determines the target electricity price forecast per-unit value for the t-th hour in the n+1-th day from the electricity price forecast per-unit value of each hour in the n+1-th day, obtains the product between the electricity price forecast reference value and the target forecast per-unit value for the n+1-th day, and uses the product as the target electricity price forecast value for the t-th hour in the n+1-th day. The present disclosure determines the target electricity price forecast value based on the electricity price forecast reference value and the target forecast per-unit value, thereby improving the accuracy of obtaining the short-term electricity price forecast value based on per-unit standardization, improving the efficiency of the short-term electricity price forecasting process based on per-unit standardization, being able to capture the small fluctuations of the electricity price in the power market, and being based on the electricity price target forecast value, being conducive to optimizing power dispatching and resource allocation, improving the profit margin, and reducing the transaction risk of enterprises in the spot market, thereby avoiding the economic losses caused by market price fluctuations.
[0085] Figure 5 FIG. 1 is a block diagram of a short-term electricity price prediction device based on per unitization according to an exemplary embodiment. Figure 5 As shown, the short-term electricity price prediction device 500 based on normalization according to an embodiment of the present disclosure may specifically include: a first acquisition module 501 , a second acquisition module 502 , a first prediction module 503 , a second prediction module 504 and a determination module 505 .
[0086] The first acquisition module 501 is used to acquire electricity price data within n historical days, where n is a positive integer;
[0087] The second acquisition module 502 is used to acquire the electricity price benchmark value within the historical n days according to the electricity price data, and acquire the electricity price per unit value per hour within the historical n days according to the electricity price data and the electricity price benchmark value within the historical n days;
[0088] The first prediction module 503 is used to predict the electricity price reference value of the n+1th day based on the electricity price reference value and the first load data of the n+1th day to be predicted, so as to obtain the electricity price prediction reference value of the n+1th day;
[0089] The second prediction module 504 is used to predict the electricity price per unit value on the n+1th day based on the electricity price per unit value and the second load data for each hour on the n+1th day, so as to obtain the predicted electricity price per unit value for each hour on the n+1th day;
[0090] The determination module 505 is used to determine the target electricity price forecast value for the tth hour in the n+1th day based on the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour in the n+1th day, where 0≤t≤24 and t is a positive integer.
[0091] In one embodiment of the present disclosure, the determination module 505 is further used to: determine the target predicted per-unit value of electricity price for the tth hour in the n+1th day from the predicted per-unit value of electricity price per hour in the n+1th day; obtain the product of the predicted reference value of electricity price for the n+1th day and the target predicted per-unit value, and use the product as the target predicted value of electricity price for the tth hour in the n+1th day.
[0092] In one embodiment of the present disclosure, the device 500 is also used to: obtain the target electricity price forecast value for each hour in the n+1th day; sort the target electricity price forecast value for each hour in the n+1th day, and determine the highest electricity price forecast value in the n+1th day based on the sorting result.
[0093] In one embodiment of the present disclosure, the first prediction module 503 is further used to: obtain a pre-trained electricity price benchmark prediction model; input the electricity price benchmark value and the first load data into the electricity price benchmark prediction model to obtain the electricity price prediction benchmark value for the n+1th day.
[0094] In one embodiment of the present disclosure, the second prediction module 504 is further used to: obtain a pre-trained electricity price per unit value prediction model; input the electricity price per unit value and the second load data into the electricity price per unit value prediction model to obtain the predicted electricity price per unit value for each hour in the n+1th day.
[0095] In one embodiment of the present disclosure, the second acquisition module 502 is further used to: determine the historical maximum daily electricity price and the historical daily average electricity price from the electricity price data within the historical n days; and obtain the electricity price reference value within the historical n days based on the historical maximum daily electricity price or the historical daily average electricity price and a preset reference coefficient.
[0096] In one embodiment of the present disclosure, the second acquisition module 502 is further used to: obtain the daily time-of-use electricity price in history; and obtain the per-unit electricity price value for each hour in the historical n days based on the daily time-of-use electricity price and the electricity price benchmark value in the historical n days.
[0097] In one embodiment of the present disclosure, the second acquisition module 502 is further used to: obtain the electricity price reference value of the jth day from the electricity price reference values within the historical n days, wherein j≤n, and j is a positive integer; obtain the electricity price at the mth hour of the jth day from the daily time-of-use electricity price, wherein 0≤m≤24, and m is a positive integer; obtain the quotient between the electricity price at the mth hour of the jth day and the electricity price reference value of the i-th day, and use the quotient as the denominator of the electricity price at the mth hour of the jth day.
[0098] In the embodiment of the present disclosure, the specific manner in which each module in the short-term electricity price prediction device based on normalization performs operations in the above-mentioned embodiment has been described in detail in the embodiment of the short-term electricity price prediction method based on normalization, and will not be repeated here.
[0099] In summary, the short-term electricity price prediction device based on per unit provided by the embodiment of the present disclosure obtains electricity price data within the past n days, obtains the electricity price benchmark value within the past n days according to the electricity price data, obtains the per unit value of the electricity price for each hour within the past n days according to the electricity price data and the electricity price benchmark value within the past n days, predicts the electricity price benchmark value for the n+1 day based on the electricity price benchmark value and the first load data for the n+1 day to be predicted, so as to obtain the electricity price prediction benchmark value for the n+1 day, and predicts the electricity price benchmark value for the n+1 day based on the electricity price benchmark value and the second load data for each hour within the n+1 day. The electricity price per unit value of the n+1th day is predicted to obtain the predicted electricity price per unit value for each hour in the n+1th day; the target electricity price prediction value for the tth hour in the n+1th day is determined according to the predicted electricity price reference value of the n+1th day and the predicted electricity price per unit value for each hour in the n+1th day. Thus, the present invention improves the accuracy and efficiency of obtaining the target electricity price prediction value by capturing the deep correlation between electricity price data and load data, and obtaining the target electricity price prediction value based on the predicted electricity price reference value and the predicted electricity price per unit value, which is beneficial to optimizing power dispatching and resource allocation.
[0100] Figure 6 is a block diagram of an electronic device 600 according to an exemplary embodiment.
[0101] like Figure 6 As shown, the electronic device 600 includes:
[0102] The memory 601 and the processor 602, a bus 603 connecting different components (including the memory 601 and the processor 602), the memory 601 stores a computer program, and when the processor 602 executes the program, the short-term electricity price prediction method based on normalization according to the embodiment of the present disclosure is implemented.
[0103] Bus 603 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.
[0104] The electronic device 600 typically includes a variety of electronic device readable media. These media can be any available media that can be accessed by the electronic device 600, including volatile and non-volatile media, removable and non-removable media.
[0105] The memory 601 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 604 and / or cache memory 605. The electronic device 600 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 606 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 603 via one or more data medium interfaces. The memory 601 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present disclosure.
[0106] A program / utility 608 having a set (at least one) of program modules 607 may be stored, for example, in the memory 601, such program modules 607 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 607 generally perform the functions and / or methods of the embodiments described in the present disclosure.
[0107] The electronic device 600 may also communicate with one or more external devices 609 (e.g., keyboard, pointing device, display 611, etc.), one or more devices that enable a user to interact with the electronic device 600, and / or any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 612. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) through a network adapter 613. Figure 6 As shown, the network adapter 613 communicates with other modules of the electronic device 600 via the bus 603. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0108] The processor 602 executes various functional applications and data processing by running the programs stored in the memory 601 .
[0109] It should be noted that the implementation process and technical principles of the electronic device of this embodiment refer to the aforementioned explanation of the short-term electricity price prediction method based on normalization in the embodiment of the present disclosure, and will not be repeated here.
[0110] The electronic device provided by the embodiment of the present disclosure obtains electricity price data within n historical days, obtains the electricity price benchmark value within n historical days according to the electricity price data, obtains the electricity price per unit value of each hour within n historical days according to the electricity price data and the electricity price benchmark value within n historical days, predicts the electricity price benchmark value of the n+1 day based on the electricity price benchmark value and the first load data of the n+1 day to be predicted, so as to obtain the electricity price prediction benchmark value of the n+1 day, and predicts the electricity price per unit value of the n+1 day based on the electricity price benchmark value and the second load data of each hour within the n+1 day. The electricity price per unit value is predicted to obtain the electricity price forecast per unit value for each hour in the n+1th day; the electricity price target forecast value for the tth hour in the n+1th day is determined according to the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour in the n+1th day. Thus, the present invention improves the accuracy and efficiency of obtaining the electricity price target forecast value by capturing the deep correlation between electricity price data and load data, and obtaining the electricity price target forecast value based on the electricity price forecast reference value and the electricity price forecast per unit value, which is beneficial to optimizing power dispatching and resource allocation.
[0111] In order to implement the above embodiments, the present disclosure also proposes a computer-readable storage medium.
[0112] When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device can perform the short-term electricity price prediction method based on per-unit as described above. Optionally, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.
[0113] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.
[0114] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A short-term electricity price forecasting method based on per-unit normalization, characterized in that: The method comprises: Obtaining electricity price data within n historical days, where n is a positive integer; According to the electricity price data, obtain the electricity price benchmark value within the historical n days, and according to the electricity price data and the electricity price benchmark value within the historical n days, obtain the electricity price per unit value per hour within the historical n days; Based on the electricity price reference value and the first load data of the n+1 day to be predicted, predicting the electricity price reference value of the n+1 day to obtain the electricity price prediction reference value of the n+1 day; Based on the electricity price per unit value and the second load data for each hour on the n+1 day, predict the electricity price per unit value on the n+1 day to obtain the predicted electricity price per unit value for each hour on the n+1 day; According to the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour within the n+1th day, the electricity price target forecast value for the tth hour within the n+1th day is determined, where 0≤t≤24 and t is a positive integer.
2. The method according to claim 1, characterized in that The step of determining the target electricity price forecast value for the tth hour on the n+1th day according to the electricity price forecast reference value on the n+1th day and the electricity price forecast per unit value for each hour on the n+1th day comprises: Determine the target predicted per unit value of the electricity price at the tth hour on the n+1th day from the predicted per unit value of the electricity price per hour on the n+1th day; The product of the electricity price forecast reference value for the n+1th day and the target forecast per unit value is obtained, and the product is used as the electricity price target forecast value for the tth hour on the n+1th day.
3. The method according to claim 2, characterized in that The method further comprises: Obtain the target electricity price forecast value for each hour on the n+1th day; The target electricity price forecast value for each hour in the n+1th day is sorted, and the highest electricity price forecast value in the n+1th day is determined according to the sorting result.
4. The method according to claim 1, characterized in that: The step of predicting the electricity price reference value on the n+1th day based on the electricity price reference value and the first load data on the n+1th day to be predicted, so as to obtain the electricity price prediction reference value on the n+1th day, includes: Obtain a pre-trained electricity price benchmark prediction model; The electricity price reference value and the first load data are input into the electricity price reference value prediction model to obtain the electricity price prediction reference value for the (n+1)th day.
5. The method according to claim 1, characterized in that The step of predicting the electricity price per unit value on the n+1th day based on the electricity price per unit value and the second load data for each hour on the n+1th day to obtain the predicted electricity price per unit value for each hour on the n+1th day includes: Obtain a pre-trained electricity price per unit value prediction model; The electricity price per unit value and the second load data are input into the electricity price per unit value prediction model to obtain the predicted electricity price per unit value for each hour on the (n+1)th day.
6. The method according to claim 1, characterized in that The step of obtaining the electricity price benchmark value within n historical days according to the electricity price data includes: Determine the historical daily maximum electricity price and the historical daily average electricity price from the electricity price data within the n historical days; The electricity price reference value within the historical n days is obtained according to the historical daily maximum electricity price or the historical daily average electricity price and a preset reference coefficient.
7. The method according to claim 6, characterized in that The step of obtaining the per unit electricity price value per hour within the historical n days according to the electricity price data and the electricity price benchmark value within the historical n days includes: Get the historical daily time-of-use electricity price; According to the daily time-of-use electricity price and the electricity price benchmark value within the historical n days, the electricity price per unit value for each hour within the historical n days is obtained.
8. The method according to claim 7, characterized in that The obtaining the per unit electricity price value per hour for the historical n days according to the daily time-of-use electricity price and the electricity price benchmark value for the historical n days includes: Obtaining the electricity price benchmark value for the jth day from the electricity price benchmark values within the n historical days, wherein j≤n, and j is a positive integer; From the daily time-of-use electricity price, obtain the electricity price at the mth hour in the jth day, where 0≤m≤24, and m is a positive integer; The quotient between the electricity price at the mth time in the jth day and the electricity price reference value on the ith day is obtained, and the quotient is used as the per unit value of the electricity price at the mth time in the jth day.
9. A short-term electricity price prediction device based on per-unit normalization, characterized in that: The device comprises: The first acquisition module is used to acquire electricity price data within n historical days, where n is a positive integer; A second acquisition module is used to acquire a base value of electricity prices within n historical days according to the electricity price data, and to acquire a per unit value of electricity prices per hour within n historical days according to the electricity price data and the base value of electricity prices within n historical days; A first prediction module is used to predict the electricity price reference value on the n+1th day based on the electricity price reference value and the first load data on the n+1th day to be predicted, so as to obtain the electricity price prediction reference value on the n+1th day; A second prediction module is used to predict the electricity price per unit value on the n+1th day based on the electricity price per unit value and the second load data for each hour on the n+1th day, so as to obtain the predicted electricity price per unit value for each hour on the n+1th day; A determination module is used to determine the target electricity price forecast value for the tth hour in the n+1th day based on the electricity price forecast reference value for the n+1th day and the electricity price forecast per unit value for each hour in the n+1th day, wherein 0≤t≤24 and t is a positive integer.
10. An electronic device, characterized in that: include: processor; a memory for storing executable instructions for the processor; The processor is configured to execute the instructions to implement the short-term electricity price prediction method based on per-unit as described in any one of claims 1-8.