New energy power automatic transaction method, device, equipment and product

Abstract

The invention discloses a new energy electric power automatic transaction method, device, equipment and product. The method comprises the following steps: acquiring historical new energy output data, historical electric power spot market clearing price data, historical irradiance data and historical wind speed data; using the first neural network model to predict the current clearing price of the electric power spot market; using the second neural network model to predict the current new energy output; establishing a multi-objective function optimization model; and resolving the multi-objective function optimization model to obtain transaction declaration electric quantity, and automatically declaring the transaction declaration electric quantity in the electric power transaction. According to the invention, the method can determine the automatic declared transaction electric quantity in the power transaction through the prediction of the corresponding future condition based on the two neural network models, has the dynamic future prediction capability, can dynamically adjust the automatic declared transaction electric quantity in the power transaction, and effectively improves the transaction efficiency and benefits.

Description

Technical Field

[0001] This invention belongs to the field of new energy, and in particular relates to a method, device, equipment and product for automatic trading of new energy power. Background Technology

[0002] In current electricity trading related to new energy sources, such as those based on solar and wind power, the traditional electricity trading methods based on fixed prices or long-term contracts have significant limitations due to the inherent difficulty in predicting and the high volatility of new energy production. These limitations include insufficient future forecasting capabilities, a lack of dynamic adjustment capabilities, and issues with the profitability of the trading results, which urgently need to be addressed. Summary of the Invention

[0003] In view of this, the present invention aims to overcome the defects in the prior art and propose a method, device, equipment and product for automatic trading of new energy power.

[0004] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0005] In a first aspect, the present invention discloses an automatic trading method for new energy power, comprising: the method comprising: acquiring historical new energy output data, historical power spot market clearing price data, historical irradiance data, and historical wind speed data;

[0006] Using the first neural network model, based on historical renewable energy output data, historical electricity spot market clearing price data, and historical irradiance data, the current electricity spot market clearing price is predicted;

[0007] Using a second neural network model, the current output of new energy sources is predicted based on historical wind speed data and historical irradiance data.

[0008] A multi-objective function optimization model is established, which includes: maximizing the product of the current electricity spot market clearing price and the transaction bid volume, minus the deviation cost; minimizing the conditional risk value function CvaRa(loss); and the constraint condition is that the transaction bid volume is less than or equal to the current renewable energy output.

[0009] Solve the multi-objective function optimization model to obtain the transaction declaration electricity volume, and automatically declare the transaction declaration electricity volume in power trading.

[0010] In one implementation of the present invention, the deviation cost includes: obtaining the current deviation assessment coefficient, wherein the deviation cost is the absolute value of the difference between the transaction declaration electricity and the current new energy output multiplied by the deviation assessment coefficient.

[0011] In one implementation of the present invention, the constraint condition is that the electricity declared for the transaction is less than or equal to the current renewable energy output, and further includes: the constraint condition is that the electricity declared for the transaction is less than or equal to the current renewable energy output multiplied by a coefficient, wherein the coefficient is the sum of 1 and the absorption rate tolerance threshold.

[0012] In one implementation of the present invention, the method further includes: calculating the revenue deviation; when the revenue deviation is greater than a set threshold, triggering the first neural network model and the second neural network model to be trained again, wherein the revenue deviation is the difference between the product of the actual electricity spot market clearing price and the actual renewable energy output and the product of the predicted current electricity spot market clearing price and the current renewable energy output.

[0013] In one implementation of the present invention, before predicting the current electricity spot market clearing price using a first neural network model based on historical renewable energy output data, historical electricity spot market clearing price data, and historical irradiance data, the method further includes: performing wavelet noise reduction processing on the historical renewable energy output data.

[0014] In one implementation of the present invention, the present invention further includes: an attention module connecting a first neural network model and a second neural network model, wherein the attention module is used to concatenate the latent features of the first neural network model and the latent features of the second neural network model into a joint feature vector, and to establish a trainable weight matrix corresponding to the joint feature vector.

[0015] In one implementation of the present invention, establishing a trainable weight matrix corresponding to the joint feature vector further includes: normalizing the trainable weight matrix using the softmax function to obtain the weight coefficients corresponding to the features.

[0016] Secondly, this invention discloses a new energy power automatic trading device, the device comprising:

[0017] The acquisition module is used to acquire historical renewable energy output data, historical electricity spot market clearing price data, historical irradiance data, and historical wind speed data.

[0018] The first prediction module is used to predict the current electricity spot market clearing price by using the first neural network model based on historical new energy output data, historical electricity spot market clearing price data, and historical irradiance data.

[0019] The second prediction module is used to predict the current output of new energy sources based on historical wind speed data and historical irradiance data using the second neural network model.

[0020] A module for establishing objective function optimization models is used to establish multi-objective function optimization models. The multi-objective function optimization models include: maximizing the product of the current electricity spot market clearing price and the transaction bid volume, minus the deviation cost; minimizing the conditional risk value function CvaRa(loss); and the constraint condition is that the transaction bid volume is less than or equal to the current renewable energy output.

[0021] The automatic trading module is used to solve the multi-objective function optimization model, obtain the transaction declaration electricity volume, and automatically declare the transaction declaration electricity volume in power trading.

[0022] Thirdly, the present invention discloses an electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the above-described method.

[0023] Fourthly, the present invention discloses a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0024] Compared with the prior art, the present invention has the following advantages:

[0025] This invention discloses a method, device, equipment, and product for automatic trading of new energy power, including: acquiring historical new energy output data, historical power spot market clearing price data, historical irradiance data, and historical wind speed data; using a first neural network model to predict the current power spot market clearing price; using a second neural network model to predict the current new energy output; establishing a multi-objective function optimization model; solving the multi-objective function optimization model to obtain the transaction declaration electricity volume, and automatically declaring the transaction declaration electricity volume in power trading. This invention discloses a method, device, equipment, and product for automatic trading of new energy power, which can determine the automatically declared transaction declaration electricity volume in power trading by based on two neural network models corresponding to future situation predictions. It has dynamic future prediction capabilities and can dynamically adjust the automatically declared transaction declaration electricity volume in power trading, effectively improving trading efficiency and benefits. Attached Figure Description

[0026] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0027] In the attached diagram:

[0028] Figure 1 This is a schematic diagram of an embodiment of the present invention and a method for automatic trading of new energy power;

[0029] Figure 2This is a schematic diagram of the attention module in an automatic trading method for new energy power according to an embodiment of the present invention;

[0030] Figure 3 This is a schematic diagram of an automatic trading device for new energy power according to an embodiment of the present invention;

[0031] Figure 4 This is a schematic diagram of an electronic device for automatic trading of new energy power according to an embodiment of the present invention. Detailed Implementation

[0032] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0033] In the description of this invention, it should be understood that the terms "center", "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0034] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0035] In the description of this invention, it should be further noted that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0036] In existing technologies, traditional electricity trading methods based on fixed prices or long-term contracts have significant limitations in new energy electricity trading, including insufficient future forecasting capabilities, a lack of dynamic adjustment capabilities, and limited benefits from trading outcomes, all of which urgently need improvement. This invention discloses a method, device, equipment, and product for automatic trading of new energy electricity. By using two neural network models to predict future conditions, it automatically determines the electricity volume to be submitted for trading, possessing dynamic future forecasting capabilities and the ability to dynamically adjust the automatically submitted trading volume, effectively improving trading efficiency and benefits.

[0037] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0038] In one embodiment of the present invention, such as Figure 1 As shown, this invention discloses an automatic trading method for new energy power, comprising:

[0039] Step S101: Obtain historical renewable energy output data, historical electricity spot market clearing price data, historical irradiance data, and historical wind speed data;

[0040] In this embodiment, every 15 minutes, with a granularity of 1 hour, historical renewable energy output data, historical electricity spot market clearing price data, historical irradiance data, and historical wind speed data for the previous 24 hours are collected.

[0041] For example, historical renewable energy output data is represented by X, where X represents the following:

[0042] X = {x1, x2, ..., x} 24};

[0043] The historical renewable energy output data were processed using wavelet denoising, as follows:

[0044] Using the wavelet basis function db4, the historical renewable energy output data X is divided into 3 layers, and wavelet decomposition at the 1-hour granularity is performed. X is decomposed into approximate component a. j (k) and detail component d j (k), where j represents the j-th layer and k represents the k-th sample;

[0045] Using the thresholding method, the detail components d j (k) Perform soft thresholding to obtain The formula is as follows:

[0046]

[0047] Where λ is the threshold, σ is d j The root mean square of (k), where N is the sample length;

[0048] Furthermore, signal reconstruction is performed to obtain 24 denoised data points, as follows:

[0049] Using the lowest level approximation a j (k) and all processed d j (k) Reverse synthesis reconstruction It is expressed as follows:

[0050]

[0051] Wherein, IDWT is the inverse discrete wavelet transform;

[0052] Step S102: Using the first neural network model, based on historical new energy output data, historical electricity spot market clearing price data, and historical irradiance data, the current electricity spot market clearing price is predicted.

[0053] Step S103: Using the second neural network model, based on historical wind speed data and historical irradiance data, predict the current output of new energy sources;

[0054] In this embodiment, the first neural network model is a Long Short-Term Memory (LSTM) neural network model, and the second neural network model is a Random Forest neural network model.

[0055] In another embodiment, the first neural network model and the second neural network model can also be other neural network models, as long as they can achieve the functions of the Long Short-Term Memory (LSTM) neural network model and the Random Forest neural network model.

[0056] When training the Long Short-Term Memory (LSTM) neural network model, the inputs are the historical renewable energy output, historical electricity spot market clearing price, and historical irradiance of the previous 24 hours with a step size of 1 hour, which are used to predict the current electricity spot market clearing price.

[0057] When training the random forest neural network model, the inputs are historical irradiance and wind speed, as well as the corresponding new energy output.

[0058] Step S104: Establish a multi-objective function optimization model, which includes: maximizing the product of the current electricity spot market clearing price and the transaction declaration amount, minus the deviation cost; minimizing the conditional risk value function CvaRa(loss); the constraint condition is that the transaction declaration amount is less than or equal to the current renewable energy output.

[0059] Step S105: Solve the multi-objective function optimization model to obtain the transaction declaration electricity volume, and automatically declare the transaction declaration electricity volume in the power trading.

[0060] This embodiment can determine the automatically declared electricity volume in electricity trading by using two neural network models to predict future situations. It has dynamic future prediction capabilities and can dynamically adjust the automatically declared electricity volume in electricity trading, effectively improving trading efficiency and benefits.

[0061] Based on the previous embodiment, in another embodiment of the present invention, the deviation cost includes: obtaining the current deviation assessment coefficient, wherein the deviation cost is the absolute value of the difference between the transaction declaration electricity and the current new energy output multiplied by the deviation assessment coefficient.

[0062] Based on the previous embodiment, in another embodiment of the present invention, the method further includes: the constraint condition is that the transaction declaration electricity is less than or equal to the current new energy output multiplied by a coefficient, wherein the coefficient is the sum of 1 and the absorption rate tolerance threshold.

[0063] For example, a multi-objective function optimization model is represented as follows:

[0064]

[0065] The current electricity spot market clearing price is represented as P5, the transaction volume is represented as Q5, and the current renewable energy output is represented as... The tolerance threshold for the absorption rate is represented by ∈, and the deviation assessment coefficient is represented by λ; T represents time, which can be 24.

[0066] For example, λ takes the value 90%, and ∈ takes the value 95%;

[0067] In this embodiment, for example, the NSGA-III algorithm is used to solve the multi-objective function optimization model to determine the value of the transaction declaration electricity Q5;

[0068] In this embodiment, the establishment of a multi-objective function optimization model can obtain the value of the transaction declaration electricity Q5 by optimizing multiple objective functions. Based on the transaction declaration electricity Q5, the electricity is automatically declared in the power transaction, which effectively improves the transaction efficiency and benefits, and has dynamic future prediction capabilities.

[0069] Based on the previous embodiment, in another embodiment of the present invention, the method further includes: calculating the revenue deviation; when the revenue deviation is greater than a set threshold, triggering the first neural network model and the second neural network model to be trained again, wherein the revenue deviation is the difference between the product of the actual electricity spot market clearing price and the actual renewable energy output and the product of the predicted current electricity spot market clearing price and the current renewable energy output.

[0070] For example, the profit deviation is represented by ΔR, and the calculation process is as follows:

[0071] ΔR=∑(P J$KL *Q J$KL -P MJ$d *Q MJ$d );

[0072] Among them, P J$KL Q represents the actual clearing price in the electricity spot market. J$KL P represents the actual output of new energy sources. MJ$d Q represents the projected current electricity spot market clearing price. MJ$d This indicates the projected current output of new energy sources;

[0073] The threshold is set as Rthreshold, which uses ±10% of the prediction deviation as the exemption interval. Rthreshold is expressed as follows:

[0074]

[0075] Among them, P MJ$d,5 This represents the current electricity spot market clearing price corresponding to the predicted time t, where T represents time and can be 24.

[0076] In this embodiment, when the profit deviation exceeds a set threshold, the first neural network model and the second neural network model are retrained, which enables the method to have dynamic future prediction capabilities and maintain the accuracy of the prediction.

[0077] Based on the previous embodiment, in another embodiment of the present invention, such as Figure 2 As shown, the method further includes: a first neural network model and a second neural network model connected to an attention module, wherein the attention module is used to concatenate the latent features of the first neural network model and the latent features of the second neural network model into a joint feature vector, and to establish a trainable weight matrix corresponding to the joint feature vector.

[0078] For example, the latent features of the first neural network model are represented as h. MJiS$ The hidden features of the second neural network model are represented as h M％T$J The joint eigenvector is represented as [h MJiS$ ,h M％T$J The trainable weight matrix is ​​represented as W. K ;

[0079] In this embodiment, a joint feature vector is established, and different features are matched with different weight coefficients. The joint training of the two neural network models improves the accuracy of model prediction.

[0080] Based on the previous embodiment, in another embodiment of the present invention, establishing a trainable weight matrix corresponding to the joint feature vector further includes: normalizing the trainable weight matrix using the softmax function to obtain the weight coefficients corresponding to the features.

[0081] For example, the weighting coefficient is represented as a5, as follows:

[0082] a5 = softmax(W K .[h MJiS$ ,h M％T$J ]);

[0083] like Figure 3 As shown, the present invention also discloses a new energy power automatic trading device, comprising:

[0084] The acquisition module 301 is used to acquire historical new energy output data, historical electricity spot market clearing price data, historical irradiance data, and historical wind speed data.

[0085] The first prediction module 302 is used to predict the current electricity spot market clearing price by using the first neural network model based on historical new energy output data, historical electricity spot market clearing price data and historical irradiance data.

[0086] The second prediction module 303 is used to predict the current output of new energy sources based on historical wind speed data and historical irradiance data using a second neural network model.

[0087] Module 304 for establishing an objective function optimization model is used to establish a multi-objective function optimization model. The multi-objective function optimization model includes: maximizing the product of the current electricity spot market clearing price and the transaction bid volume, minus the deviation cost; minimizing the conditional risk value function CvaRa(loss); the constraint condition is that the transaction bid volume is less than or equal to the current renewable energy output.

[0088] The automatic trading module 305 is used to solve the multi-objective function optimization model, obtain the transaction declaration electricity volume, and automatically declare the transaction declaration electricity volume in power trading.

[0089] The present invention also discloses an electronic device, such as Figure 4 The diagram shows a block diagram of an embodiment of an electronic device applicable to the above-mentioned automatic trading of new energy power.

[0090] The electronic device 40 of this embodiment includes a processor 401, which can perform various appropriate actions and processes according to a program stored in ROM 402 or a program loaded from storage portion 408 into RAM 403. The processor 401 may include, for example, a general-purpose microprocessor, an instruction set processor and / or related chipsets and / or dedicated microprocessors, etc. The processor 401 may also include onboard memory for caching purposes. The processor 401 may include a single processing unit or multiple processing units for performing different actions of the method flow according to embodiments of the present invention.

[0091] RAM 403 stores various programs and data required for the operation of electronic device 40. Processor 401, ROM 402, and RAM 403 are interconnected via bus 404. Processor 401 executes various operations of the method flow according to embodiments of the present invention by executing programs in ROM 402 and / or RAM 403. It should be noted that programs may also be stored in one or more memories other than ROM 402 and RAM 403, and processor 401 may also execute various operations of the method flow according to embodiments of the present invention by executing programs stored in one or more memories.

[0092] According to an embodiment of the present invention, the electronic device 40 may further include an I / O interface 405, which is also connected to the bus 404. The electronic device 40 may also include one or more of the following components connected to the I / O interface 405: an input section 406 including a keyboard, mouse, etc.; an output section 407 including a cathode ray tube, liquid crystal display, and speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN card or modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 4010 is also connected to the I / O interface 405 as needed. A removable medium 4011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 4010 as needed so that computer programs read from it can be installed into the storage section 408 as needed.

[0093] The present invention also provides a computer-readable storage medium.

[0094] The computer-readable storage medium may be included in the electronic device / apparatus system described in the above embodiments; or it may exist independently and not assembled into the electronic device / apparatus. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of the present invention.

[0095] According to embodiments of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0096] Embodiments of the present invention also include a computer program product.

[0097] The computer program product includes a computer program containing program code for performing the methods provided in the embodiments of the present invention. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the methods provided in the embodiments of the present invention.

[0098] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0099] According to embodiments of the present invention, program code for executing the computer programs provided in the embodiments of the present invention can be written using any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages. Programming languages ​​include, but are not limited to, Java, C++, Python, C, or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device.

[0100] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features recited in the various embodiments and / or claims of the present invention can be combined and / or combined in various ways, even if such combinations or combinations are not expressly stated in the present invention. In particular, the features described in the various embodiments and / or claims of this invention can be combined and / or combined in various ways without departing from the spirit and teachings of this invention. All such combinations and / or combinations fall within the scope of this invention.

[0101] The embodiments of the present invention have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of the invention. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of the invention is defined by the appended claims and their equivalents. Without departing from the scope of the invention, various substitutions and modifications can be made by those skilled in the art, and all such substitutions and modifications should fall within the scope of the invention.

Claims

1. A method for automatic trading of new energy power, characterized in that, include: The method includes: acquiring historical renewable energy output data, historical electricity spot market clearing price data, historical irradiance data, and historical wind speed data; Using the first neural network model, based on the historical renewable energy output data, the historical electricity spot market clearing price data, and the historical irradiance data, the current electricity spot market clearing price is predicted; Using a second neural network model, the current output of new energy sources is predicted based on the historical wind speed data and the historical irradiance data. A multi-objective function optimization model is established, wherein the multi-objective function optimization model includes: maximizing the product of the current electricity spot market clearing price and the transaction declaration electricity volume, minus the deviation cost; minimizing the conditional risk value function CvaRa(loss); the constraint condition is that the transaction declaration electricity volume is less than or equal to the current renewable energy output. Solve the multi-objective function optimization model to obtain the transaction declaration electricity volume, and automatically declare the transaction declaration electricity volume in the power transaction.

2. The automatic trading method for new energy power according to claim 1, characterized in that, The deviation cost includes: obtaining the current deviation assessment coefficient, wherein the deviation cost is the absolute value of the difference between the declared electricity volume and the current renewable energy output multiplied by the deviation assessment coefficient.

3. The automatic trading method for new energy power according to claim 1, characterized in that, The constraint condition is that the declared electricity volume for the transaction is less than or equal to the current renewable energy output, and also includes: the constraint condition is that the declared electricity volume for the transaction is less than or equal to the current renewable energy output multiplied by a coefficient, wherein the coefficient is the sum of 1 and the absorption rate tolerance threshold.

4. The automatic trading method for new energy power according to claim 1, characterized in that, The method further includes: calculating the revenue deviation; when the revenue deviation is greater than a set threshold, triggering the first neural network model and the second neural network model to be trained again, wherein the revenue deviation is the difference between the product of the actual electricity spot market clearing price and the actual renewable energy output and the product of the predicted current electricity spot market clearing price and the current renewable energy output.

5. The automatic trading method for new energy power according to claim 1, characterized in that, Before using the first neural network model to predict the current electricity spot market clearing price based on the historical renewable energy output data, the historical electricity spot market clearing price data, and the historical irradiance data, the method further includes: performing wavelet noise reduction processing on the historical renewable energy output data.

6. The automatic trading method for new energy power according to claim 1, characterized in that, The method further includes: connecting the first neural network model and the second neural network model to an attention module, wherein the attention module is used to concatenate the latent features of the first neural network model and the latent features of the second neural network model into a joint feature vector, and to establish a trainable weight matrix corresponding to the joint feature vector.

7. The automatic trading method for new energy power according to claim 6, characterized in that, The step of establishing the trainable weight matrix corresponding to the joint feature vector further includes: normalizing the trainable weight matrix using the softmax function to obtain the weight coefficients corresponding to the features.

8. A new energy power automatic trading device, characterized in that: The device includes: The acquisition module is used to acquire historical renewable energy output data, historical electricity spot market clearing price data, historical irradiance data, and historical wind speed data. The first prediction module is used to predict the current electricity spot market clearing price by using a first neural network model based on the historical new energy output data, the historical electricity spot market clearing price data, and the historical irradiance data. The second prediction module is used to predict the current output of new energy sources based on the historical wind speed data and the historical irradiance data using a second neural network model. A module for establishing an objective function optimization model is used to establish a multi-objective function optimization model, wherein the multi-objective function optimization model includes: maximizing the product of the current electricity spot market clearing price and the transaction declaration electricity volume, minus the deviation cost; minimizing the conditional risk value function CvaRa(loss); the constraint condition is that the transaction declaration electricity volume is less than or equal to the current renewable energy output. An automatic trading module is used to solve the multi-objective function optimization model, obtain the declared electricity volume for trading, and automatically declare the declared electricity volume for trading in electricity trading.

9. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method of any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 7.

Images