Information generation method and device, equipment and storage medium

Abstract

The invention discloses an information generation method, device and equipment and a storage medium, and the method comprises the steps: carrying out the image processing of historical and future market feature data of each time point of an electricity market, obtaining the market image data of each time point, carrying out the reconstruction processing of historical and future meteorological feature data of each time point, and obtaining the market image data of each time point; obtaining meteorological image data of each time point; performing dimension reduction processing on the market image data and the meteorological image data at each time point to obtain one-dimensional market vector data and one-dimensional meteorological vector data at each time point; fusing the two pieces of one-dimensional vector data to obtain target fusion data of each time point; inputting the target fusion data of each time point into a pre-trained prediction model to obtain a predicted electricity price of each time point on a prediction day, an output result of the prediction model being a weighted summation result of at least two sub-prediction models; and generating participation decision information for the participation end of the electricity market based on the predicted electricity price at each time point of the prediction day.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of information generation, and in particular, to an information generation method, apparatus, device and storage medium. Background Art

[0002] In today's digital and intelligent era, various industries have an increasing demand for accurate and efficient information generation methods, especially in the power sector. Accurate decision-making information plays a vital role in the decision-making of participants (such as power companies or users) in various business activities.

[0003] However, when processing market characteristic data and meteorological characteristic data, existing information generation methods usually simply splice or directly superimpose different types of data. This approach not only increases the computational complexity, but also reduces the efficiency of information generation, and cannot meet the current needs of the power industry for high-precision and high-efficiency information generation.

[0004] Therefore, it is urgent to propose a new method to solve the above problems. Summary of the invention

[0005] The present invention provides an information generation method, device, equipment and storage medium, which can improve the efficiency of information generation in the power field.

[0006] In a first aspect, an embodiment of the present invention provides an information generation method, the method comprising:

[0007] Performing image processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, and performing reconstruction processing on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point;

[0008] Performing dimensionality reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and performing dimensionality reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point;

[0009] Fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fused data at each time point;

[0010] Inputting the target fusion data at each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, wherein the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum result of the at least two sub-prediction models;

[0011] Based on the predicted electricity prices at each time point on the predicted day, participation decision information is generated for participants in the electricity market.

[0012] The technical solution of the present invention firstly processes the historical market characteristic data and future market characteristic data of the power market at each time point into images to obtain the market image data at each time point, which not only helps to capture the spatial and temporal relationship of the power market and provide rich characteristic information, thereby improving the quality of market data, but also enhances the learning ability of the model, so that it can more accurately understand and predict market dynamics. The historical meteorological characteristic data and future meteorological characteristic data at each time point are reconstructed to obtain the meteorological image data at each time point, which can avoid mutual interference between different meteorological characteristics, while retaining their original information in different channels, improving the expression ability and quality of meteorological data, and thus improving the reliability of the model. After that, the market image data and meteorological image data at each time point are respectively subjected to dimensionality reduction processing to obtain one-dimensional market vector data and one-dimensional meteorological vector data at each time point, which can significantly reduce the dimension of the data, reduce the complexity of subsequent data analysis and calculation, and improve the processing efficiency and the speed of information generation. Then, the one-dimensional market vector data and one-dimensional meteorological vector data at each time point are fused to obtain the target fusion data at each time point, which improves the comprehensiveness of the data information, provides richer feature information for the prediction model, and helps the model better capture the complexity of market dynamics and meteorological changes, thereby improving the accuracy of the prediction and the reliability of decision-making information. The target fusion data at each time point is input into the pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, which improves the accuracy and efficiency of the determined predicted electricity price. Finally, based on the predicted electricity price at each time point on the prediction day, decision-making information is generated for the participants of the power market, which not only improves the efficiency of information generation, but also improves the scientificity and accuracy of decision-making, thereby improving the satisfaction of the participants in the power market and promoting the healthy development of the power market.

[0013] In a second aspect, an embodiment of the present invention further provides an information generating device, the device comprising:

[0014] An imaging module is used to perform imaging processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, and to perform reconstruction processing on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point;

[0015] A dimension reduction module, used to perform dimension reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and to perform dimension reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point;

[0016] A fusion module, used for fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fusion data at each time point;

[0017] A prediction module, used for inputting the target fusion data of each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, wherein the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum result of the at least two sub-prediction models;

[0018] A generation module is used to generate participation decision information for participants in the power market based on the predicted electricity prices at each time point on the predicted day.

[0019] In a third aspect, an embodiment of the present invention further provides a computer device, the computer device comprising:

[0020] at least one processor; and a memory communicatively coupled to the at least one processor;

[0021] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute any information generation method described in the first aspect.

[0022] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, implement any information generating method described in the first aspect.

[0023] It should be noted that the above computer instructions may be stored in whole or in part on a computer-readable storage medium, wherein the computer-readable storage medium may be packaged together with the processor of the information generating device, or may be packaged separately from the processor of the information generating device, and this application does not limit this.

[0024] The description of the second, third and fourth aspects in this application can refer to the detailed description of the first aspect; and the beneficial effects of the description of the second, third and fourth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.

[0025] In this application, the name of the above-mentioned information generating device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalent technologies.

[0026] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 A flowchart of an information generation method provided by an embodiment of the present invention;

[0029] Figure 2 A flowchart of another information generating method provided by an embodiment of the present invention;

[0030] Figure 3 A schematic diagram of the structure of an information generating device provided by an embodiment of the present invention;

[0031] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0033] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0034] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0035] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0036] It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe various operations (or steps) as sequential processes, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of various operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to methods, functions, procedures, subroutines, subprograms, etc. In addition, the embodiments in the present invention and the features in the embodiments can be combined with each other without conflict.

[0037] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0038] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0039] Figure 1 This is a flowchart of an information generation method provided by an embodiment of the present invention. This embodiment is applicable to situations where the efficiency of generating information in the power field needs to be improved. The method can be executed by an information generation device, and the information generation device can be implemented in hardware / software. The device can be integrated in a computer device, for example, it can be installed in a computer, and this embodiment of the present invention does not limit this. Figure 1 As shown, the specific steps include:

[0040] Step 110: Graphically process the historical market characteristic data and future market characteristic data of the electricity market at each time point to obtain market image data at each time point, and reconstruct the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain meteorological image data at each time point.

[0041] Specifically, historical market characteristic data refers to various market-related data disclosed and recorded by the power market in the past period of time (such as one day, one month, etc.). Future market characteristic data refers to various market-related data disclosed and recorded by the power market in the future period of time (such as one day). For example: market characteristic data can be bidding space, inter-provincial interconnection lines, spot electricity prices, provincial load, coal price data, new energy power generation, hydropower power generation, minimum power generation of thermal power units and power generation surplus of thermal power units. Market image data refers to market characteristic data after image processing and presented in the form of images. Historical meteorological characteristic data refers to meteorological image data (such as temperature maps, wind maps, cloud maps, etc.) disclosed and recorded by the Meteorological Bureau in the past period of time (such as one day, etc.). Future meteorological characteristic data refers to meteorological image data disclosed and recorded by the Meteorological Bureau in the future period of time (such as one day). Meteorological image data refers to meteorological characteristic data after reconstruction processing and presented in the form of images.

[0042] In the specific implementation, the historical market characteristic data and future market characteristic data at the same time point can be spliced ​​to obtain the integrated market data at each time point. Then the integrated market data at each time point is normalized to obtain the normalized market data at each time point. The market characteristic data at each time point is then constructed into a matrix (such as behavioral characteristics, listed as time points), and each element in the matrix is ​​mapped to a color value. Finally, the image processing library (such as OpenCV or PIL library in Python) is used to convert the above color-mapped matrix into an image to obtain the market image data at each time point.

[0043] At the same time, the historical and future meteorological characteristic data at each time point are split into three primary color (red, green and blue) channels to obtain meteorological image data at each time point. The meteorological image data at each time point includes image data of three color channels.

[0044] In this embodiment, by processing the historical market characteristic data and future market characteristic data of the power market at each time point in a graphical manner, the market image data at each time point is obtained, which not only helps to capture the spatial and temporal relationship of the power market and provide rich characteristic information, thereby improving the quality of market data, but also enhances the learning ability of the model, so that it can more accurately understand and predict market dynamics. And the historical meteorological characteristic data and future meteorological characteristic data at each time point are reconstructed to obtain meteorological image data at each time point, which can avoid mutual interference between different meteorological characteristics, while retaining their original information in different channels, improving the expression ability and quality of meteorological data, and thus improving the reliability of the model.

[0045] Step 120: Perform dimensionality reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and perform dimensionality reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point.

[0046] Specifically, the one-dimensional market vector data refers to the one-dimensional vector data obtained by reducing the dimension of the market image data. The one-dimensional meteorological vector data refers to the one-dimensional vector data obtained by reducing the dimension of the meteorological image data.

[0047] In the specific implementation, first, according to the actual situation or needs, a suitable dimensionality reduction method (such as principal component analysis, linear discriminant analysis, t-distributed random neighbor embedding, uniform manifold approximation and projection, etc.) is selected, and then the selected dimensionality reduction method is applied to perform dimensionality reduction processing on the market image data and meteorological image data at each time point to obtain one-dimensional market vector data and one-dimensional meteorological vector data at each time point.

[0048] In this embodiment, by performing dimensionality reduction processing on the market image data and meteorological image data at each time point, one-dimensional market vector data and one-dimensional meteorological vector data at each time point are obtained, which can significantly reduce the dimension of the data, reduce the complexity of subsequent data analysis and calculation, and improve processing efficiency and the speed of information generation.

[0049] Step 130: fuse the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fused data at each time point.

[0050] Specifically, the target fusion data refers to data obtained by fusing one-dimensional market vector data and one-dimensional meteorological vector data.

[0051] In a specific implementation, the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point can be directly spliced ​​together using a splicing function to obtain the target fusion data at each time point.

[0052] In this embodiment, by fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point, the target fusion data at each time point is obtained, which improves the comprehensiveness of the data information, provides richer feature information for the prediction model, and helps the model to better capture the complexity of market dynamics and meteorological changes, thereby improving the accuracy of the prediction and further improving the reliability of decision-making information.

[0053] Step 140: Input the target fusion data at each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day.

[0054] Specifically, the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum of at least two sub-prediction models. For example, the prediction model includes sub-model A and sub-model B, the weight of sub-model A is a1, and the weight of sub-model B is a2. When the result of sub-model A is y1 and the result of sub-model B is y2, the result of the prediction model is a1×y1+a2×y2. The prediction model refers to a model for predicting future electricity prices in the power market based on the target fusion data of each historical time point and the actual electricity price corresponding to each historical time point. The sub-prediction model refers to a component in the prediction model, and each sub-model is trained based on a different algorithm.

[0055] In the specific implementation, after obtaining the target fusion data at each time point, the target fusion data at each time point can be input into a pre-trained prediction model, so that the prediction model calculates the electricity price at each time point on the prediction day based on the input target fusion data, and outputs the predicted electricity price at each time point on the prediction day in the form of an array or time series.

[0056] In this embodiment, the target fusion data at each time point is input into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, thereby improving the accuracy and efficiency of the determined predicted electricity price.

[0057] Step 150: Generate participation decision information for participants in the power market based on the predicted electricity prices at each time point on the prediction day.

[0058] Specifically, decision-making information refers to information generated based on forecast results and used to guide power market participants (such as power generation companies, users, power sales companies, etc.) to make decisions.

[0059] For example, if the electricity market participant is a power generation enterprise, power generation output recommendations and market quotation recommendations at each time point can be generated for the power generation enterprise based on the predicted electricity prices at each time point on the forecast day. For example, during periods when the predicted electricity prices are higher, the output can be increased to maximize profits; during periods when the predicted electricity prices are lower, the output can be reduced or equipment maintenance can be performed to reduce operating costs.

[0060] For example, if the electricity market participant is a user, electricity load adjustment suggestions for each time point can be generated for the user based on the predicted electricity price at each time point on the forecast day. For example: reduce unnecessary electricity consumption during periods when the predicted electricity price is higher, such as postponing the use of high-energy consumption equipment; increase electricity consumption during periods when the predicted electricity price is lower, such as for washing clothes or heating hot water.

[0061] For example, if the electricity market participant is a power sales company, a power purchase amount recommendation and a power sales price recommendation for each time point can be generated for the power sales company based on the predicted electricity price at each time point on the forecast day.

[0062] In this embodiment, decision-making information is generated for the participants in the power market based on the predicted electricity prices at each time point on the forecast day, which not only improves the efficiency of information generation, but also improves the scientificity and accuracy of decision-making, thereby improving the satisfaction of the participants in the power market and promoting the healthy development of the power market.

[0063] In the embodiment of the present invention, the historical market characteristic data and future market characteristic data of the power market at each time point are processed by image processing to obtain the market image data at each time point, which not only helps to capture the spatial and temporal relationship of the power market and provide rich characteristic information, thereby improving the quality of market data, but also enhances the learning ability of the model, so that it can more accurately understand and predict market dynamics. The historical meteorological characteristic data and future meteorological characteristic data at each time point are reconstructed to obtain the meteorological image data at each time point, which can avoid mutual interference between different meteorological characteristics, while retaining their original information in different channels, improving the expression ability and quality of meteorological data, and thus improving the reliability of the model. After that, the market image data and meteorological image data at each time point are respectively processed by dimensionality reduction to obtain one-dimensional market vector data and one-dimensional meteorological vector data at each time point, which can significantly reduce the dimension of the data, reduce the complexity of subsequent data analysis and calculation, and improve processing efficiency and the speed of information generation. Then, the one-dimensional market vector data and one-dimensional meteorological vector data at each time point are fused to obtain the target fusion data at each time point, which improves the comprehensiveness of the data information, provides richer feature information for the prediction model, and helps the model better capture the complexity of market dynamics and meteorological changes, thereby improving the accuracy of the prediction and the reliability of decision-making information. The target fusion data at each time point is input into the pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, which improves the accuracy and efficiency of the determined predicted electricity price. Finally, based on the predicted electricity price at each time point on the prediction day, decision-making information is generated for the participants of the power market, which not only improves the efficiency of information generation, but also improves the scientificity and accuracy of decision-making, thereby improving the satisfaction of the participants in the power market and promoting the healthy development of the power market.

[0064] Figure 2 A flowchart of another information generation method provided by an embodiment of the present invention, which is specific based on the above embodiment. In this embodiment, the method may also include:

[0065] Step 210: splice the historical market characteristic data and the future market characteristic data at the same time point to obtain the integrated market data at each time point.

[0066] Specifically, integrated market data refers to a data set obtained by splicing historical market characteristic data and future market characteristic data at the same time point, represented in the form of a matrix, for example: each row of the matrix corresponds to a market characteristic, and each column represents a different time point.

[0067] For example, historical market characteristic data: t1: the electricity price is 50 yuan / MWh, and the electricity trading volume is 1000 MWh. t2: the electricity price is 60 yuan / MWh, and the electricity trading volume is 1200 MWh. t3: ​​the electricity price is 55 yuan / MWh, and the electricity trading volume is 1100 MWh. Future market characteristic data: t1: the predicted electricity price is 52 yuan / MWh, and the predicted electricity trading volume is 1050 MWh. t2: the predicted electricity price is 62 yuan / MWh, and the predicted electricity trading volume is 1250 MWh. t3: ​​the predicted electricity price is 58 yuan / MWh, and the predicted electricity trading volume is 1150 MWh. The historical market characteristic data and future market characteristic data at the same time point are spliced ​​to obtain the integrated market data at each time point: the integrated market data at t1 is: [[50, 52], [1000, 1050]], the integrated market data at t2 is: [[60, 62], [1200, 1250]], and the integrated market data at t3 is: [[55, 58], [1100, 1150]].

[0068] In this embodiment, by splicing the historical market characteristic data and the future market characteristic data at the same time point, the integrated market data at each time point is obtained, thereby improving the integrity of the market data.

[0069] Step 211: normalize the integrated market data at each time point to obtain standard market data at each time point.

[0070] Specifically, standard market data refers to data obtained after normalizing the integrated market data at each time point, which is expressed in the form of a matrix.

[0071] For example, assuming that the value range of electricity price is [0,100], and the value range of electricity trading volume is [0,2000]. The integrated market data obtained above is normalized to obtain the standard market data at each time point: the standard market data at t1 is: [[0.5, 0.52], [0.5, 0.525]], the standard market data at t2 is: [[0.6, 0.62], [0.6, 0.625]], and the standard market data at t3 is: [[0.55, 0.58], [0.55, 0.575]].

[0072] In this embodiment, by normalizing the integrated market data at each time point to obtain standard market data at each time point, dimensional differences in the data can be eliminated, data comparability can be improved, the model can converge faster, and model performance can be improved.

[0073] Step 212: grayscale the standard market data at each time point to obtain the market image data corresponding to each time point.

[0074] Specifically, each element in the matrix corresponding to the standard market data at each time point can be mapped to a color value, and then the image processing library (such as OpenCV or PIL library in Python) can be used to convert the above color mapped matrix into an image to obtain the market image data at each time point.

[0075] In this embodiment, by graying the standard market data at each time point, the market image data corresponding to each time point is obtained, which not only helps to capture the spatial and temporal relationship of the electricity market and provide rich feature information, thereby improving the quality of market data, but also enhances the learning ability of the model, enabling it to more accurately understand and predict market dynamics.

[0076] Step 213: perform three-primary color separation on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain meteorological image data at each time point.

[0077] In the specific implementation, first, determine the data format of the historical meteorological characteristic data and the future meteorological characteristic data (such as RGB, grayscale, CMYK, HSV, HSL, etc.). If the data format of the historical meteorological characteristic data and the future meteorological characteristic data is not in RGB format, use an image processing library (such as OpenCV or PIL library in Python) to convert it into RGB format, and then use an image processing library (such as Pillow library in Python) to split the historical meteorological characteristic data and the future meteorological characteristic data at each time point into three color channel maps of red, green, and blue at each time point to obtain the meteorological image data at each time point.

[0078] In this embodiment, by performing three-primary color splitting on the historical and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point, mutual interference between different meteorological characteristics can be avoided while retaining their original information in different channels, thereby improving the expression ability and quality of meteorological data and thus improving the reliability of the model.

[0079] Step 214: perform dimensionality reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and perform dimensionality reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point.

[0080] Furthermore, the market image data at each time point is subjected to dimensionality reduction processing to obtain one-dimensional market vector data at each time point, including: performing convolution and pooling processing on the market image data at each time point to obtain characteristic market image data at each time point; flattening processing on the characteristic market image data at each time point to obtain one-dimensional market vector data at each time point; correspondingly, the meteorological image data at each time point is subjected to dimensionality reduction processing to obtain one-dimensional meteorological vector data at each time point, including: performing convolution and pooling processing on the meteorological image data at each time point to obtain characteristic meteorological data at each time point; flattening processing on the characteristic meteorological data at each time point to obtain one-dimensional meteorological vector data at each time point.

[0081] In the specific implementation, the convolutional pooling layer can be used to perform convolution and pooling processing on the market image data at each time point to obtain the characteristic market image data at each time point. The specific formula is: 1 =conv n (G 1 ), where f 1 is the feature market image data, conv n is a convolutional pooling layer with n layers, G 1 is the market image data. Next, the characteristic market image data at each time point is flattened to obtain the one-dimensional market vector data at each time point. The specific formula is: 2 = flatten(f 1 ), where flatten is the operation of flattening the matrix to a 1-dimensional vector (dimension is 1×n), and f 2 is the one-dimensional market vector data. Then the same method is used to process the meteorological image data at each time point to obtain the one-dimensional meteorological vector data at each time point.

[0082] In this embodiment, through the above steps, the robustness of the features can be enhanced while reducing the amount of data, making the features more prominent, thereby accelerating the training speed of the model and improving its generalization ability, and also increasing the speed of subsequent decision generation.

[0083] Step 215: splice the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain fused data at each time point.

[0084] Specifically, fused data refers to data obtained by splicing one-dimensional market vector data and one-dimensional meteorological vector data.

[0085] In the specific implementation, the concat function can be used to splice the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain the fused data at each time point to enhance the integrity of the data, thereby improving the accuracy of the model and providing a more comprehensive decision-making basis.

[0086] Step 216: Perform dimensionality reduction processing on the fused data at each time point to obtain target fused data at each time point.

[0087] Specifically, in this embodiment, the target fusion data refers to one-dimensional vector data obtained after dimensionality reduction of the fusion data.

[0088] In the specific implementation, the fully connected neural network can be used to reduce the dimension of the fused data at each time point to obtain the target fused data at each time point. The specific formula is: x = line y (c), where x represents the target fusion data, line y represents a y-layer fully connected neural network, and c represents fused data.

[0089] In this embodiment, by performing dimensionality reduction processing on the fused data at each time point to obtain the target fused data at each time point, data redundancy can be reduced, data quality can be improved, and the efficiency of decision information generation can be improved.

[0090] Step 217: Input the target fusion data at each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day.

[0091] Furthermore, the training process of the prediction model includes: obtaining the target fusion data of each first historical time point and the actual electricity price corresponding to each first historical time point; using the target fusion data of each first historical time point and the actual electricity price corresponding to each first historical time point as training data to perform model training on at least two sub-prediction models, and calculating the loss function of at least two sub-prediction models; optimizing at least two sub-prediction models based on the back propagation algorithm until the loss function of at least two sub-prediction models converges.

[0092] Specifically, the first historical time point refers to a time point in the historical data set used to train the prediction model. The actual electricity price refers to the electricity price actually occurring at the historical time point. At least two sub-prediction models are deep learning models using different algorithms, for example, the sub-prediction models can be XGBoost, SVR, LSTM, GRU, and Transformer.

[0093] In a specific implementation, the target fusion data of each first historical time point and the actual electricity price corresponding to each first historical time point are first obtained, and then the target fusion data of each first historical time point and the actual electricity price corresponding to each first historical time point are used as training data to train at least two sub-prediction models, and the loss function of at least two sub-prediction models is calculated during the training process until the loss functions of all sub-prediction models converge. Thereafter, the model can be optimized based on a back propagation algorithm (such as a gradient descent method) until a preset optimization condition is met (such as the number of iterations reaches a preset number) to obtain a prediction model.

[0094] In this embodiment, through the above steps, the advantages of different algorithms can be brought into play, the error of the prediction model can be reduced, and the prediction accuracy can be improved.

[0095] Furthermore, after optimizing at least two sub-prediction models based on the back-propagation algorithm until the loss functions of the at least two sub-prediction models converge, it also includes: obtaining target fusion data of each second historical time point and the actual electricity price corresponding to each second historical time point; inputting the target fusion data of each second historical time point into the prediction model to obtain the predicted electricity price corresponding to each second historical time point; calculating the accuracy of the prediction model based on the actual electricity price and the predicted electricity price corresponding to each second historical time point; when the accuracy is lower than the preset accuracy, adjusting the weights of at least two sub-prediction models in the prediction model according to the accuracy of the prediction model.

[0096] Specifically, the second historical time point refers to a time point in the historical data set used to verify the performance of the prediction model. The predicted electricity price refers to the electricity price predicted by the prediction model based on the input target fusion data. The accuracy rate refers to an indicator that measures the prediction accuracy of the prediction model. The preset accuracy rate refers to an accuracy rate threshold pre-set according to actual conditions or needs, which is used to evaluate whether the prediction model needs to adjust the weight of the sub-model.

[0097] In the specific implementation, the target fusion data of each second historical time point and the actual electricity price corresponding to each second historical time point are first obtained, and then the target fusion data of each second historical time point is input into the prediction model to obtain the predicted electricity price corresponding to each second historical time point. Then, the accuracy of the prediction model is calculated based on the actual electricity price and predicted electricity price corresponding to each second historical time point. The specific calculation formula is as follows:

[0098]

[0099] Among them, ACC is the accuracy, y i ′ is the predicted electricity price at the i-th time point, y i is the actual electricity price at the i-th time point, and n is the number of time points.

[0100] The calculated accuracy is then compared with the preset accuracy. If the accuracy is lower than the preset accuracy, the weights of at least two sub-prediction models in the prediction model are adjusted according to the accuracy of the prediction model, and the accuracy is recalculated until the accuracy is not lower than the preset accuracy. For example, a particle swarm algorithm or an ensemble learning method can be used to adjust the weights.

[0101] In this embodiment, by adjusting the sub-model weights, the model can better adapt to new data features and patterns, thereby improving the generalization ability of the model under different data distributions, and further enhancing the accuracy and stability of the model.

[0102] Step 219: Generate participation decision information for participants in the power market based on the predicted electricity prices at each time point on the prediction day.

[0103] Therefore, the technical solution of the present invention first splices the historical market characteristic data and future market characteristic data at the same time point to obtain the integrated market data at each time point, thereby improving the integrity of the market data. Then, the integrated market data at each time point is normalized to obtain the standard market data at each time point, thereby eliminating the dimensional differences in the data, improving the data comparability, and enabling the model to converge faster and improve the model performance. Afterwards, the standard market data at each time point is grayed to obtain the market image data corresponding to each time point, which not only helps to capture the spatial and temporal relationship of the power market, provides rich feature information, thereby improving the quality of market data, but also enhances the learning ability of the model, enabling it to more accurately understand and predict market dynamics. The historical meteorological characteristic data and future meteorological characteristic data at each time point are split into three primary colors to obtain the meteorological image data at each time point, thereby avoiding mutual interference between different meteorological features, while retaining their original information in different channels, improving the expression ability and quality of meteorological data, and thereby improving the reliability of the model. Then, the market image data and meteorological image data at each time point are processed by dimensionality reduction to obtain one-dimensional market vector data and one-dimensional meteorological vector data at each time point, which can significantly reduce the dimension of the data, reduce the complexity of subsequent data analysis and calculation, and improve the processing efficiency and the speed of information generation. The one-dimensional market vector data and one-dimensional meteorological vector data at each time point are spliced ​​to obtain the fusion data at each time point to enhance the integrity of the data, thereby improving the accuracy of the model and providing a more comprehensive decision-making basis. The fusion data at each time point is processed by dimensionality reduction to obtain the target fusion data at each time point, which can reduce data redundancy, improve data quality, and thus improve the efficiency of decision-making information generation. The target fusion data at each time point is input into the pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, which improves the accuracy and efficiency of the determined predicted electricity price. Finally, based on the predicted electricity price at each time point on the prediction day, the decision-making information for the participants in the power market is generated, which not only improves the efficiency of information generation, but also improves the scientificity and accuracy of decision-making, thereby improving the satisfaction of the participants in the power market and promoting the healthy development of the power market.

[0104] Figure 3 A structural schematic diagram of an information generating device provided for an embodiment of the present invention. The device and the information generating methods of the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiments of the information generating device, reference can be made to the embodiments of the above-mentioned information generating methods.

[0105] like Figure 3 As shown, the device comprises:

[0106] The imaging module 310 is used to perform imaging processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, and to perform reconstruction processing on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point;

[0107] A dimension reduction module 320 is used to perform dimension reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and to perform dimension reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point;

[0108] A fusion module 330, for fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fusion data at each time point;

[0109] A prediction module 340 is used to input the target fusion data at each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, wherein the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum result of the at least two sub-prediction models;

[0110] The generation module 350 is used to generate participation decision information for the participants in the power market based on the predicted electricity prices at each time point on the predicted day.

[0111] On the basis of the above-mentioned embodiment, the imaging module 310 performs imaging processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, including:

[0112] splicing the historical market characteristic data and the future market characteristic data at the same time point to obtain the integrated market data at each time point;

[0113] Normalizing the integrated market data at each time point to obtain standard market data at each time point;

[0114] Grayscale processing is performed on the standard market data at each time point to obtain market image data corresponding to each time point.

[0115] On the basis of the above embodiment, the imaging module 310 reconstructs the historical meteorological characteristic data and the future meteorological characteristic data at each time point to obtain the meteorological image data at each time point, including:

[0116] The historical meteorological characteristic data and the future meteorological characteristic data at each time point are split into three primary colors to obtain the meteorological image data at each time point.

[0117] On the basis of the above embodiment, the dimension reduction module 320 performs dimension reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, including:

[0118] Performing convolution and pooling processing on the market image data at each time point to obtain characteristic market image data at each time point;

[0119] Flattening the characteristic market image data at each time point to obtain one-dimensional market vector data at each time point;

[0120] Accordingly, the dimension reduction module 320 performs dimension reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point, including:

[0121] Performing convolution and pooling processing on the meteorological image data at each time point to obtain characteristic meteorological data at each time point;

[0122] The characteristic meteorological data at each time point is flattened to obtain one-dimensional meteorological vector data at each time point.

[0123] On the basis of the above embodiment, the fusion module 330 is specifically used to: splice the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain the fusion data at each time point;

[0124] The fusion data at each time point is subjected to dimensionality reduction processing to obtain the target fusion data at each time point.

[0125] Based on the above embodiment, the training process of the prediction model includes:

[0126] Obtain target fusion data for each first historical time point and actual electricity prices corresponding to each first historical time point;

[0127] Using the target fusion data of each first historical time point and the actual electricity price corresponding to each first historical time point as training data to perform model training on the at least two sub-prediction models, and calculating the loss function of the at least two sub-prediction models; the at least two sub-prediction models are deep learning models using different algorithms;

[0128] The at least two sub-prediction models are optimized based on a back-propagation algorithm until the loss functions of the at least two sub-prediction models converge.

[0129] Based on the above embodiment, the device further includes:

[0130] A weight adjustment module is used to optimize the at least two sub-prediction models based on the back propagation algorithm until the loss function of the at least two sub-prediction models converges, obtain the target fusion data of each second historical time point and the actual electricity price corresponding to each second historical time point; input the target fusion data of each second historical time point into the prediction model to obtain the predicted electricity price corresponding to each second historical time point; calculate the accuracy of the prediction model based on the actual electricity price and the predicted electricity price corresponding to each second historical time point; when the accuracy is lower than the preset accuracy, adjust the weights of at least two sub-prediction models in the prediction model according to the accuracy of the prediction model.

[0131] The information generating device provided in the embodiment of the present invention can execute the information generating method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0132] It is worth noting that in the embodiment of the above-mentioned information generating device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0133] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 4 A block diagram of an exemplary computer device 4 suitable for use in implementing embodiments of the present invention is shown. Figure 4 The computer device 4 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0134] like Figure 4 As shown, the computer device 4 is in the form of a general-purpose computing electronic device. The components of the computer device 4 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0135] Bus 18 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 architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0136] The computer device 4 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 4, including volatile and non-volatile media, removable and non-removable media.

[0137] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 4 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 4 not shown, usually called a "hard drive"). Although Figure 4 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 18 via one or more data medium interfaces. The system memory 28 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 invention.

[0138] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 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. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0139] The computer device 4 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the computer device 4, and / or any device that enables the computer device 4 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 22. Furthermore, the computer device 4 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 20. Figure 4 As shown, the network adapter 20 communicates with other modules of the computer device 4 via the bus 18. It should be understood that although Figure 4Not shown, other hardware and / or software modules may be used in conjunction with computer device 4, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0140] The processing unit 16 executes various functional applications and page displays by running the program stored in the system memory 28, for example, implementing the information generation method provided by the embodiment of the present invention, which includes:

[0141] Performing image processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, and performing reconstruction processing on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point;

[0142] Performing dimensionality reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and performing dimensionality reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point;

[0143] Fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fused data at each time point;

[0144] Inputting the target fusion data at each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, wherein the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum result of the at least two sub-prediction models;

[0145] Based on the predicted electricity prices at each time point on the predicted day, participation decision information is generated for participants in the electricity market.

[0146] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the information generating method provided by any embodiment of the present invention.

[0147] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, for example, an information generation method provided by an embodiment of the present invention is implemented. The method includes:

[0148] Performing image processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, and performing reconstruction processing on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point;

[0149] Performing dimensionality reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and performing dimensionality reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point;

[0150] Fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fused data at each time point;

[0151] Inputting the target fusion data at each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, wherein the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum result of the at least two sub-prediction models;

[0152] Based on the predicted electricity prices at each time point on the predicted day, participation decision information is generated for participants in the electricity market.

[0153] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0154] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0155] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0156] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0157] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0158] In addition, the acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0159] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for generating information, characterized in that: include: Performing image processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, and performing reconstruction processing on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point; Performing dimensionality reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and performing dimensionality reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point; Fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fused data at each time point; Inputting the target fusion data at each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, wherein the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum result of the at least two sub-prediction models; Based on the predicted electricity prices at each time point on the predicted day, participation decision information is generated for participants in the electricity market.

2. The information generation method according to claim 1, characterized in that: The historical market characteristic data and future market characteristic data of the power market at each time point are processed into images to obtain the market image data at each time point, including: splicing the historical market characteristic data and the future market characteristic data at the same time point to obtain the integrated market data at each time point; Normalizing the integrated market data at each time point to obtain standard market data at each time point; Grayscale processing is performed on the standard market data at each time point to obtain market image data corresponding to each time point.

3. The information generation method according to claim 1, characterized in that: Reconstructing the historical meteorological characteristic data and the future meteorological characteristic data at each time point to obtain the meteorological image data at each time point includes: The historical meteorological characteristic data and the future meteorological characteristic data at each time point are split into three primary colors to obtain the meteorological image data at each time point.

4. The information generation method according to claim 1, characterized in that: Performing dimensionality reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, including: Performing convolution and pooling processing on the market image data at each time point to obtain characteristic market image data at each time point; Flattening the characteristic market image data at each time point to obtain one-dimensional market vector data at each time point; Accordingly, the meteorological image data at each time point is subjected to dimensionality reduction processing to obtain one-dimensional meteorological vector data at each time point, including: Performing convolution and pooling processing on the meteorological image data at each time point to obtain characteristic meteorological data at each time point; The characteristic meteorological data at each time point is flattened to obtain one-dimensional meteorological vector data at each time point.

5. The information generation method according to claim 1, characterized in that: The one-dimensional market vector data and the one-dimensional meteorological vector data at each time point are fused to obtain target fused data at each time point, including: splicing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain fused data at each time point; The fusion data at each time point is subjected to dimensionality reduction processing to obtain the target fusion data at each time point.

6. The information generating method according to claim 1, characterized in that: The training process of the prediction model includes: Obtain target fusion data for each first historical time point and actual electricity prices corresponding to each first historical time point; Using the target fusion data of each first historical time point and the actual electricity price corresponding to each first historical time point as training data to perform model training on the at least two sub-prediction models, and calculating the loss function of the at least two sub-prediction models; the at least two sub-prediction models are deep learning models using different algorithms; The at least two sub-prediction models are optimized based on a back-propagation algorithm until the loss functions of the at least two sub-prediction models converge.

7. The information generating method according to claim 6, characterized in that: After optimizing the at least two sub-prediction models based on the back propagation algorithm until the loss functions of the at least two sub-prediction models converge, the method further includes: Obtaining target fusion data of each second historical time point and actual electricity prices corresponding to each second historical time point; Inputting the target fusion data of each second historical time point into the prediction model to obtain the predicted electricity price corresponding to each second historical time point; Calculating the accuracy of the prediction model based on the actual electricity price and the predicted electricity price corresponding to each second historical time point; When the accuracy is lower than a preset accuracy, the weights of at least two sub-prediction models in the prediction model are adjusted according to the accuracy of the prediction model.

8. An information generating device, characterized in that: include: An imaging module is used to perform imaging processing on the historical market characteristic data and future market characteristic data of the power market at each time point to obtain the market image data at each time point, and to perform reconstruction processing on the historical meteorological characteristic data and future meteorological characteristic data at each time point to obtain the meteorological image data at each time point; A dimension reduction module, used to perform dimension reduction processing on the market image data at each time point to obtain one-dimensional market vector data at each time point, and to perform dimension reduction processing on the meteorological image data at each time point to obtain one-dimensional meteorological vector data at each time point; A fusion module, used for fusing the one-dimensional market vector data and the one-dimensional meteorological vector data at each time point to obtain target fusion data at each time point; A prediction module, used for inputting the target fusion data of each time point into a pre-trained prediction model to obtain the predicted electricity price at each time point on the prediction day, wherein the prediction model includes at least two sub-prediction models, and the output result of the prediction model is the weighted sum result of the at least two sub-prediction models; A generation module is used to generate participation decision information for participants in the power market based on the predicted electricity prices at each time point on the predicted day.

9. A computer device, characterized in that: The computer device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the information generating method described in any one of claims 1-7.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to perform the information generation method described in any one of claims 1 to 7 when executed by a computer processor.

Images