Power transaction decision-making method, device and equipment and storage medium

Abstract

The invention provides a power transaction decision-making method, device and equipment and a storage medium, and the method comprises the steps: obtaining time sequence operation data of a power market; inputting the time sequence operation data into a pre-trained transaction decision model, and obtaining predicted transaction data output by the transaction decision model; wherein the transaction decision-making model comprises a price prediction sub-model and a transaction decision-making sub-model, the price prediction sub-model obtains the capability of predicting and obtaining price prediction data based on time sequence operation data through pre-training, and the transaction decision-making sub-model obtains the capability of generating a transaction decision based on the price prediction data through pre-training; the activation function of the transaction decision model is composed of a plurality of nonlinear activation functions, and the transaction decision model is obtained by training with the prediction error of the power price and the prediction error of the decision income as loss functions. According to the technical scheme, support can be provided for transaction decision-making of the electricity market.

Description

Technical Field

[0001] The present application relates to the field of power trading technology, and in particular to a power trading decision-making method, device, equipment and storage medium. Background Art

[0002] In related technologies, due to the high volatility and nonlinear characteristics of the current electricity trading market price, traditional forecasting methods (e.g., ARIMA, BP neural network) have the following limitations when dealing with long-term dependencies and complex patterns: (1) Traditional activation functions cannot dynamically adapt to different time series characteristics; (2) They can only predict electricity trading prices; (3) They have poor adaptability to dynamic changes in the electricity trading market. Summary of the Invention

[0003] The present application aims to solve one of the technical problems in the related art at least to a certain extent.

[0004] In the first aspect, the present application proposes a method for electricity trading decision-making, which includes: obtaining time-series operation data of the electricity market; inputting the time-series operation data into a pre-trained trading decision model to obtain predicted trading data output by the trading decision model; wherein the trading decision model includes a price prediction sub-model and a trading decision sub-model, the price prediction sub-model obtains the ability to predict price prediction data based on the time-series operation data through pre-training, and the trading decision sub-model obtains the ability to generate trading decisions based on the price prediction data through pre-training, the activation function of the trading decision model is composed of multiple nonlinear activation functions, and the trading decision model is trained with the prediction error of electricity price and the prediction error of decision benefit as the loss function.

[0005] In one implementation, the operating data includes at least one of the following: historical electricity trading data; load demand; new energy output; market supply-demand ratio; and weather data.

[0006] In one implementation, the above loss function can be expressed as follows:

[0007] L=λ1L price +λ2L decision

[0008]

[0009] Among them, L is the total loss function, L price is the loss function corresponding to the price prediction sub-model, L decision is the loss function corresponding to the transaction decision sub-model, λ1 and λ2 are dynamic weight values, T ′ is the total prediction duration, is the predicted electricity value at time t, pt is the actual electricity price at time t, q t is the trading volume at time t, c(q t ) is the transaction cost at time t, and γ is the risk discount factor at time t.

[0010] In one implementation, the nonlinear activation function includes a Swish function and a GELU function. The activation function can be expressed as follows:

[0011] f(x)=α·Swish(x)+(1-α)GELU(x)

[0012] Among them, α is the weight coefficient.

[0013] In one implementation, the method further includes: obtaining actual transaction data corresponding to the predicted transaction data; and optimizing parameters of the transaction decision model based on the predicted transaction data and the actual transaction results.

[0014] In the second aspect, the present application proposes an electricity trading decision-making device, which includes: an acquisition module for acquiring time-series operation data of the electricity market; a processing module for inputting the time-series operation data into a pre-trained trading decision model to obtain predicted trading data output by the trading decision model; wherein, the trading decision model includes a price prediction sub-model and a trading decision sub-model, the price prediction sub-model obtains the ability to predict price prediction data based on the time-series operation data through pre-training, and the trading decision sub-model obtains the ability to generate trading decisions based on the price prediction data through pre-training, the activation function of the trading decision model is composed of multiple nonlinear activation functions, and the trading decision model is trained with the prediction error of electricity price and the prediction error of decision benefit as the loss function.

[0015] In one implementation, the operating data includes at least one of the following: historical electricity trading data; load demand; new energy output; market supply-demand ratio; and weather data.

[0016] In one implementation, the above loss function can be expressed as follows:

[0017] L=λ1L price +λ2L decision

[0018]

[0019] Among them, λ1 and λ2 are dynamic weight values, T ′ is the total prediction duration, is the predicted electricity value at time t, p t is the actual electricity price at time t, q tis the trading volume at time t, c(q t ) is the transaction cost at time t, and γ is the risk discount factor at time t.

[0020] In one implementation, the nonlinear activation function includes a Swish function and a GELU function. The activation function can be expressed as follows:

[0021] f(x)=α·Swish(x)+(1-α)GELU(x)

[0022] Among them, α is the weight coefficient.

[0023] In one implementation, the device may further include a model optimization module, configured to: obtain actual transaction data corresponding to the predicted transaction data; and optimize parameters of the transaction decision model based on the predicted transaction data and the actual transaction results.

[0024] In a third aspect, the present application proposes an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the electricity trading decision-making method as described in the first aspect.

[0025] In a fourth aspect, the present application proposes a computer-readable storage medium for storing instructions, which, when executed, enables the method described in the first aspect to be implemented.

[0026] In a fifth aspect, the present application proposes a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the power trading decision-making method as described in the first aspect.

[0027] The power trading decision-making method, apparatus, device, and storage medium provided in this application can input acquired power market time series operation data into a pre-trained trading decision model to obtain intersection decision data output by the trading decision model. This method can make power market trading decisions based on a pre-trained neural network model, providing support for power market trading decisions.

[0028] 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 practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] 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:

[0030] Figure 1This is a flow chart of a power trading decision-making method provided in an embodiment of the present application;

[0031] Figure 2 This is a schematic diagram of the structure of a power trading decision-making device provided in an embodiment of the present application;

[0032] Figure 3 This is a schematic diagram of the structure of another power trading decision-making device provided in an embodiment of the present application;

[0033] Figure 4 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] The following describes in detail embodiments of the present application, examples of which 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.

[0035] The following describes the power trading decision-making method and device according to the embodiments of the present application with reference to the accompanying drawings.

[0036] Figure 1 This is a flow chart of a power trading decision-making method provided by an embodiment of the present application. Figure 1 As shown, the method may include but is not limited to the following steps:

[0037] Step S101: Acquire time-series operation data of the power market.

[0038] For example, a set of operation data of the power market may be periodically acquired at preset time intervals, and the acquired multiple sets of operation data may be used as the above-mentioned time series operation data.

[0039] For example, the decision preference data may include but is not limited to:

[0040] In an embodiment of the present application, the above-mentioned operating data includes at least one of the following: historical data on electricity transactions; load demand; new energy output; market supply-demand ratio; and weather data.

[0041] For example, the electricity transaction history data may include electricity price-related data and corresponding timestamps, wherein the electricity price-related data may include but is not limited to: the highest price, the lowest price, and the average price.

[0042] For example, the load demand data may include power demand data and a corresponding timestamp.

[0043] Exemplarily, the renewable energy output data may include renewable energy power generation in each cycle and a corresponding timestamp.

[0044] Exemplarily, the market supply-demand ratio data may include the supply-demand ratio of the electricity market and a corresponding timestamp.

[0045] For example, the weather data may include weather events, occurrence times of weather events, and impact levels of weather events on the power system.

[0046] Step S102: input the time series operation data into the pre-trained transaction decision model to obtain the predicted transaction data output by the transaction decision model.

[0047] In an embodiment of the present application, the above-mentioned predicted transaction data includes at least one of the following: predicted transaction time, predicted transaction volume, and predicted transaction amount.

[0048] Among them, the trading decision model includes a price prediction sub-model and a trading decision sub-model. The price prediction sub-model obtains the ability to obtain price prediction data based on time series operation data through pre-training, and the trading decision sub-model obtains the ability to generate trading decisions based on price prediction data through pre-training.

[0049] In some embodiments, the transaction decision model may be a spatiotemporal Transformer model.

[0050] In some embodiments, the transaction decision model is trained using the prediction error of electricity price and the prediction error of decision benefit as loss functions.

[0051] In an optional implementation, the above loss function can be expressed as follows:

[0052] L=λ1L price +λ2L decision

[0053]

[0054]

[0055] Among them, L is the total loss function, L price is the loss function corresponding to the price prediction sub-model, L decision is the loss function corresponding to the transaction decision sub-model, λ1 and λ2 are dynamic weight values, T ′ is the total prediction duration, is the predicted electricity value at time t, p t is the actual electricity price at time t, q t is the trading volume at time t, c(q t ) is the transaction cost at time t, and γ is the risk discount factor at time t.

[0056] Illustratively, the risk discount factor may be obtained by using any one of a risk premium method, a risk-adjusted discount rate method, and a historical volatility method.

[0057] In some embodiments, the activation function of the transaction decision model is composed of multiple nonlinear activation functions.

[0058] In one implementation, the nonlinear activation function may include a Swish activation function and a GELU (Gaussian Error Linear Unit) activation function. The activation function may be expressed as follows:

[0059] f(x)=α·Swish(x)+(1-α)GELU(x)

[0060] Among them, α is the dynamic weight value.

[0061] It is understandable that the above dynamic weight value can be adjusted based on gradient information.

[0062] By implementing the embodiments of the present application, the acquired time series operation data of the power market can be input into a pre-trained trading decision model to obtain the intersection decision data output by the trading decision model. The ability to make trading decisions in the power market based on a pre-trained neural network model can provide support for trading decisions in the power market.

[0063] In some embodiments, the above-mentioned trading decision model is pre-trained through the following steps: obtaining historical operating data of the electricity market and performing normalization processing and data cleaning, performing feature extraction on the processed historical operating data to obtain time series feature data, dividing the time series feature data into a training set, a validation set, and a test set, using the training set in combination with the aforementioned loss function to train the initial model, and using hyperparameters for tuning, using the test set model for evaluation, and obtaining a trading decision model.

[0064] In some embodiments of the present application, the above method may further include: obtaining actual transaction data corresponding to the predicted transaction data; and optimizing parameters of the transaction decision model based on the predicted transaction data and the actual transaction results.

[0065] Exemplarily, actual transaction data corresponding to the predicted transaction data is obtained every preset time period (for example, 24 hours) to calculate the prediction deviation and decision profit loss based on the predicted transaction data and the actual transaction data, and the model parameters of the transaction decision model are updated using the gradient descent update method based on the prediction deviation and decision profit loss.

[0066] In some embodiments, the user's transaction preferences may also be obtained, and the transaction decision model may be optimized based on the transaction preferences.

[0067] Exemplarily, the user's historical transaction data is obtained and standardized, and feature extraction is performed on the processed data to obtain feature data. During the process of training the model, the feature data is used as the input of the model to optimize the model parameters.

[0068] See Figure 2 , Figure 2 This is a schematic diagram of the structure of a power trading decision-making device provided in an embodiment of the present application. Figure 2 As shown, the device 200 includes: an acquisition module 201, which is used to obtain time series operation data of the electricity market; a processing module 202, which is used to input the time series operation data into a pre-trained trading decision model to obtain predicted trading data output by the trading decision model; wherein the trading decision model includes a price prediction sub-model and a trading decision sub-model, the price prediction sub-model obtains the ability to obtain price prediction data based on the time series operation data through pre-training, and the trading decision sub-model obtains the ability to generate trading decisions based on the price prediction data through pre-training, the activation function of the trading decision model is composed of multiple nonlinear activation functions, and the trading decision model is trained with the prediction error of electricity price and the prediction error of decision benefit as the loss function.

[0069] In one implementation, the operating data includes at least one of the following: historical electricity trading data; load demand; new energy output; market supply-demand ratio; and weather data.

[0070] In one implementation, the above loss function can be expressed as follows:

[0071] L=λ1L price +λ2L decision

[0072]

[0073] Among them, L is the total loss function, L price is the loss function corresponding to the price prediction sub-model, L decision is the loss function corresponding to the transaction decision sub-model, λ1 and λ2 are dynamic weight values, T ′ is the total prediction duration, is the predicted electricity value at time t, p t is the actual electricity price at time t, q t is the trading volume at time t, c(q t ) is the transaction cost at time t, and γ is the risk discount factor at time t.

[0074] In one implementation, the nonlinear activation function includes a Swish function and a GELU function. The activation function can be expressed as follows:

[0075] f(x)=α·Swish(x)+(1-α)GELU(x)

[0076] Among them, α is the weight coefficient.

[0077] In one implementation, the above device may further include a model optimization module. As an example, see Figure 3 , Figure 3 This is a schematic diagram of the structure of another power transaction decision-making device provided in an embodiment of the present application. Figure 3 As shown, the device 300 also includes a model optimization module for: obtaining actual transaction data corresponding to the predicted transaction data; and optimizing parameters of the transaction decision model based on the predicted transaction data and the actual transaction results. Figure 3 Modules 301 and 302 in Figure 2 Modules 201 and 202 have the same structure and function.

[0078] Through the apparatus of the present invention, the acquired time series operation data of the power market can be input into a pre-trained trading decision model to obtain the intersection decision data output by the trading decision model. The ability to make trading decisions in the power market based on a pre-trained neural network model can provide support for trading decisions in the power market.

[0079] It should be noted that the above explanations of the embodiment of the power trading decision-making method are also applicable to the power trading decision-making device of this embodiment and will not be repeated here.

[0080] In order to implement the above embodiment, the present application also proposes an electronic device. Figure 4 , Figure 4 Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 4 As shown, the electronic device 400 includes: a processor 401, and a memory 402 communicatively connected to the processor 401; the memory 402 stores computer-executable instructions; the processor 401 executes the computer-executable instructions stored in the memory to implement the method provided in the aforementioned embodiment.

[0081] 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.

[0082] 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.

[0083] 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.

[0084] In the description of this application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is only a way to describe the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0085] It is important to note that personal information collected from users should be used for legitimate and reasonable purposes and should not be shared or sold beyond these legitimate uses. Furthermore, such collection / sharing should be conducted only after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign an agreement / authorization that includes the relevant user information before using the feature. Furthermore, any necessary steps must be taken to safeguard and secure access to such personal information and ensure that others with access to personal information comply with its privacy policy and procedures.

[0086] This application contemplates providing implementation options for users to selectively block the use or access of personal information data. Specifically, this application contemplates providing hardware and / or software to prevent or block access to such personal information data. Risks can be minimized by limiting data collection and deleting data once it is no longer needed. Furthermore, where applicable, such personal information can be de-identified to protect user privacy.

[0087] In the descriptions of the foregoing embodiments, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean 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 expressions 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, unless they are mutually inconsistent.

[0088] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0089] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment 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 be performed out of 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 those skilled in the art to which the embodiments of the present application belong.

[0090] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, 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 it in another suitable manner if necessary, and then storing it in a computer memory.

[0091] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0092] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0093] In addition, the functional units in the various embodiments 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 a 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.

[0094] 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 is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A power trading decision-making method, characterized in that: include: Obtaining time-series operation data of the power market; The time series operation data is input into a pre-trained transaction decision model to obtain predicted transaction data output by the transaction decision model; wherein, the transaction decision model includes a price prediction sub-model and a transaction decision sub-model, the price prediction sub-model is pre-trained to obtain the ability to predict price prediction data based on the time series operation data, and the transaction decision sub-model is pre-trained to obtain the ability to generate transaction decisions based on the price prediction data, the activation function of the transaction decision model is composed of multiple nonlinear activation functions, and the transaction decision model is trained with the prediction error of electricity price and the prediction error of decision benefit as the loss function.

2. The method according to claim 1, wherein The operation data includes at least one of the following: Historical data on electricity transactions; load demand; New energy output; Market supply-demand ratio; Weather data.

3. The method according to claim 1, wherein The above loss function can be expressed as follows: L=λ1L price +λ2L decision Among them, L is the total loss function, L price is the loss function corresponding to the price prediction sub-model, L decision is the loss function corresponding to the transaction decision sub-model, λ1 and λ2 are dynamic weight values, T ′ is the total prediction duration, is the predicted electricity value at time t, p t is the actual electricity price at time t, q t is the trading volume at time t, c(q t ) is the transaction cost at time t, and γ is the risk discount factor at time t.

4. The method according to claim 1, wherein The nonlinear activation function includes the Swish function and the GELU function. The activation function can be expressed as follows: f(x)=α·Swish(x)+(1-α)GELU(x) Among them, α is the weight coefficient.

5. The method according to claim 1, wherein The method further comprises: Acquiring actual transaction data corresponding to the predicted transaction data; Parameters of the transaction decision model are optimized based on the predicted transaction data and the actual transaction results.

6. An electric power trading decision-making device, characterized in that: include: Acquisition module, used to obtain time series operation data of the power market; A processing module is used to input the time series operation data into a pre-trained transaction decision model to obtain predicted transaction data output by the transaction decision model; wherein, the transaction decision model includes a price prediction sub-model and a transaction decision sub-model, the price prediction sub-model is pre-trained to obtain the ability to predict price prediction data based on the time series operation data, and the transaction decision sub-model is pre-trained to obtain the ability to generate transaction decisions based on the price prediction data, the activation function of the transaction decision model is composed of multiple nonlinear activation functions, and the transaction decision model is trained with the prediction error of electricity price and the prediction error of decision benefit as the loss function.

7. The device according to claim 6, characterized in that The operation data includes at least one of the following: Historical data on electricity transactions; load demand; New energy output; Market supply-demand ratio; Extreme weather events.

8. An electronic device, characterized in that: include: 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 according to any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 5 when executed by a processor.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 5 when being executed by a processor.

Images