Multi-mode intelligent fusion electricity price prediction method and device

Abstract

The embodiment of the invention provides a multi-mode intelligent fusion electricity price prediction method and device. The method comprises the steps of obtaining a data set of a target energy storage power station in a future time period; inputting the data set of the future time period into a pre-trained electricity price classification model to obtain an electricity price category label of the future time period; inputting the data set of the future time period into a pre-trained electricity price prediction model to obtain an electricity price prediction value of the future time period; and performing fusion processing on the electricity price category label and the electricity price prediction value to obtain a final electricity value of the target energy storage power station in a future time period.

Description

Technical Field

[0001] The present application relates to the technical field of electricity price prediction, and in particular to a multi-mode intelligent fusion electricity price prediction method and device. Background Art

[0002] With the rapid development of renewable energy and the gradual opening of the power market, electricity price volatility is increasing. Traditional electricity price forecasting methods mainly rely on statistical methods such as time series analysis and linear regression. Although they can predict future electricity prices to a certain extent, they often have the following defects:

[0003] 1) Insufficient accuracy: When faced with complex electricity price fluctuations, the accuracy of traditional models is difficult to meet market demand, especially after emergencies or policy adjustments, the prediction deviation will increase significantly.

[0004] 2) Failure to consider the characteristics of electricity price range: Existing electricity price forecasting models usually only give a specific forecast value without considering the range volatility of electricity prices, which leads to insufficient flexibility of the forecast results in practical applications.

[0005] 3) Model uniformity: Existing technologies often focus on a single prediction or classification and fail to effectively combine the advantages and disadvantages of both, resulting in low prediction stability and reliability. Summary of the invention

[0006] In view of this, the embodiments of the present application provide a multi-mode intelligent fusion electricity price prediction method. One or more embodiments of the present application also involve a multi-mode intelligent fusion electricity price prediction device, a computing device, a computer-readable storage medium and a computer program to solve the technical defects existing in the prior art.

[0007] According to a first aspect of an embodiment of the present application, a multi-mode intelligent fusion electricity price prediction method is provided, comprising:

[0008] Obtain a data set of a future time period of a target energy storage power station;

[0009] The electricity price category label of the future time period is obtained by inputting the data set of the future time period into a pre-trained electricity price classification model;

[0010] The predicted value of the electricity price for the future time period is obtained by inputting the data set for the future time period into a pre-trained electricity price prediction model;

[0011] The final electricity value of the target energy storage power station in the future time period is obtained by fusing the electricity price category label with the electricity price forecast value.

[0012] Preferably, it also includes:

[0013] Collect historical data sets of each energy storage power station, wherein the historical data sets at least include time series corresponding to electricity price data, meteorological data and load data; wherein the meteorological data at least includes temperature, humidity and wind speed; and the load data at least includes power demand data and industrial power consumption data.

[0014] Preferably, the pre-trained electricity price classification model includes:

[0015] By analyzing the electricity price data in the historical data set of each energy storage power station, the electricity price data is divided into multiple electricity price intervals from small to large and non-overlapping, and an electricity price category label is configured for each electricity price interval;

[0016] Divide the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of first sub-historical data sets, and configure a corresponding electricity price category label for each first sub-historical data set according to the plurality of electricity price intervals and the electricity price data in each first sub-historical data set;

[0017] An electricity price classification model is constructed, and each of the first sub-historical data sets is used as a model input, and the electricity price classification model is trained using the corresponding electricity price category label configured for each of the first sub-historical data sets as a model output to obtain a trained electricity price classification model.

[0018] Preferably, the pre-trained electricity price prediction model includes:

[0019] Dividing the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of second sub-historical data sets;

[0020] An electricity price prediction model is constructed, and each of the second sub-historical data sets is used as a model input, and the electricity price prediction model is trained using the electricity price data corresponding to each of the second sub-historical data sets as a model output to obtain a trained electricity price prediction model.

[0021] Preferably, by fusing the electricity price category label with the electricity price forecast value, the final electricity value of the target energy storage power station in the future time period is obtained, which includes:

[0022] Determine the lower limit value and the upper limit value of the electricity price for the future time period by using the electricity price category label for the future time period;

[0023] According to the electricity price forecast value and the lower limit value and the upper limit value of the electricity price in the future time period, the final electricity value of the target energy storage power station in the future time period is obtained.

[0024] Preferably, using the electricity price category label of the future time period to determine the lower limit value and the upper limit value of the electricity price of the future time period includes:

[0025] Determining the electricity price range corresponding to the electricity price category label according to the electricity price category label of the future time period;

[0026] The lower limit value and the upper limit value of the electricity price in the future time period are calculated according to the electricity price range corresponding to the electricity price category label and the real electricity value corresponding to the electricity price range.

[0027] Preferably, according to the electricity price forecast value and the lower limit value and upper limit value of the electricity price in the future time period, obtaining the final electricity value of the target energy storage power station in the future time period includes:

[0028] Comparing the electricity price forecast value with the lower limit value and the upper limit value of the electricity price in the future time period respectively;

[0029] If the electricity price forecast value is greater than or equal to the electricity price lower limit value and less than or equal to the electricity price upper limit value, the electricity price forecast value is used as the final electricity value of the target energy storage power station in the future time period;

[0030] If the electricity price forecast value is less than the electricity price lower limit value, the electricity price lower limit value is used as the final electricity value of the target energy storage power station in the future time period;

[0031] If the electricity price forecast value is greater than the electricity price upper limit value, the electricity price upper limit value is used as the final electricity value of the target energy storage power station in the future time period.

[0032] According to a second aspect of an embodiment of the present application, a multi-mode intelligent fusion electricity price prediction device is provided, comprising:

[0033] An acquisition module is configured to acquire a data set of a future time period of a target energy storage power station;

[0034] A first prediction module is configured to obtain an electricity price category label for the future time period by inputting the data set for the future time period into a pre-trained electricity price classification model;

[0035] A second prediction module is configured to obtain a predicted electricity price value for the future time period by inputting the data set for the future time period into a pre-trained electricity price prediction model;

[0036] The electricity price fusion module is configured to obtain the final electricity value of the target energy storage power station in the future time period by fusing the electricity price category label with the electricity price forecast value.

[0037] According to a third aspect of an embodiment of the present application, a computing device is provided, including:

[0038] Memory and processor;

[0039] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement any step of the multi-mode intelligent fusion electricity price prediction method.

[0040] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of any one of the multi-mode intelligent fusion electricity price prediction methods are implemented.

[0041] According to a fifth aspect of an embodiment of the present application, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned multi-mode intelligent fusion electricity price prediction method.

[0042] The multi-mode intelligent fusion electricity price prediction scheme provided in the embodiment of the present application is achieved by obtaining a data set of a future time period of the target energy storage power station; by inputting the data set of the future time period into a pre-trained electricity price classification model to obtain an electricity price category label for the future time period; by inputting the data set of the future time period into a pre-trained electricity price prediction model to obtain an electricity price prediction value for the future time period; by fusing the electricity price category label with the electricity price prediction value to obtain the final electricity value of the future time period of the target energy storage power station. The multi-mode intelligent fusion electricity price prediction scheme provided in the embodiment of the present application is designed to improve the accuracy and stability of electricity price prediction by constructing an electricity price classification model and an electricity price prediction model, and by fusing the output results of the two models to determine the final electricity price data for the future time period, thereby achieving a more accurate electricity price output value. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of a multi-mode intelligent fusion electricity price prediction method provided by an embodiment of the present application;

[0044] Figure 2 This is a schematic diagram of the structure of a multi-mode intelligent fusion electricity price prediction device provided by an embodiment of the present application;

[0045] Figure 3 It is a process flow chart of a multi-mode intelligent fusion electricity price prediction method provided by an embodiment of the present application;

[0046] Figure 4 It is a structural block diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0047] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0048] The terms used in one or more embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present application. The singular forms of "a", "said" and "the" used in one or more embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present application refers to and includes any or all possible combinations of one or more associated listed items.

[0049] It should be understood that, although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present application, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0050] In the present application, a multi-mode intelligent fusion electricity price prediction method is provided. The present application also involves a multi-mode intelligent fusion electricity price prediction device, a computing device, a computer-readable storage medium, and a computer program, which are described in detail one by one in the following embodiments.

[0051] First, by combining the electricity price prediction model with the electricity price classification model, this application realizes the complementary advantages and disadvantages of the two models, and significantly improves the accuracy and stability of electricity price prediction. Because the classification model is generally less difficult to fit than the prediction model, the prediction accuracy is also higher. Secondly, electricity price interval classification: the use of electricity price interval classification makes the prediction results more applicable and flexible, and can adapt to electricity price fluctuations in different market environments. Thirdly, dynamic adjustment capability: by dividing the electricity price into multiple intervals, the final electricity price output value can be dynamically adjusted according to the actual electricity price situation, which improves the response speed and adaptability of the system. Finally, wide applicability: this method is not only suitable for electricity price forecasting in the power market, but can also be extended to other fields that require price forecasting (such as energy, commodities, etc.).

[0052] Figure 1A flowchart of a multi-mode intelligent fusion electricity price prediction method provided according to an embodiment of the present application is shown, which specifically includes the following steps.

[0053] Step S101: Acquire a data set of a future time period of a target energy storage power station;

[0054] Step S102: obtaining an electricity price category label for the future time period by inputting the data set for the future time period into a pre-trained electricity price classification model;

[0055] In an optional implementation, it also includes:

[0056] Collect historical data sets of each energy storage power station, wherein the historical data sets at least include time series corresponding to electricity price data, meteorological data and load data; wherein the meteorological data at least includes temperature, humidity and wind speed; and the load data at least includes power demand data and industrial power consumption data.

[0057] In an optional implementation, the pre-trained electricity price classification model includes:

[0058] By analyzing the electricity price data in the historical data set of each energy storage power station, the electricity price data is divided into multiple electricity price intervals from small to large and non-overlapping, and an electricity price category label is configured for each electricity price interval;

[0059] Divide the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of first sub-historical data sets, and configure a corresponding electricity price category label for each first sub-historical data set according to the plurality of electricity price intervals and the electricity price data in each first sub-historical data set;

[0060] An electricity price classification model is constructed, and each of the first sub-historical data sets is used as a model input, and the electricity price classification model is trained using the corresponding electricity price category label configured for each of the first sub-historical data sets as a model output to obtain a trained electricity price classification model.

[0061] Step S103: obtaining a predicted electricity price value for the future time period by inputting the data set for the future time period into a pre-trained electricity price prediction model;

[0062] In an optional implementation, it also includes:

[0063] Collect historical data sets of each energy storage power station, wherein the historical data sets at least include time series corresponding to electricity price data, meteorological data and load data; wherein the meteorological data at least includes temperature, humidity and wind speed; and the load data at least includes power demand data and industrial power consumption data.

[0064] In an optional implementation, the pre-trained electricity price prediction model includes:

[0065] Dividing the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of second sub-historical data sets;

[0066] An electricity price prediction model is constructed, and each of the second sub-historical data sets is used as a model input, and the electricity price prediction model is trained using the electricity price data corresponding to each of the second sub-historical data sets as a model output to obtain a trained electricity price prediction model.

[0067] Step S104: The final electricity value of the target energy storage power station in the future time period is obtained by fusing the electricity price category label with the electricity price forecast value.

[0068] In an optional implementation, the final electricity value of the target energy storage power station in the future time period is obtained by fusing the electricity price category label with the electricity price forecast value, including:

[0069] Determine the lower limit value and the upper limit value of the electricity price for the future time period by using the electricity price category label for the future time period;

[0070] According to the electricity price forecast value and the lower limit value and the upper limit value of the electricity price in the future time period, the final electricity value of the target energy storage power station in the future time period is obtained.

[0071] In an optional implementation, using the electricity price category label of the future time period to determine the lower limit value and the upper limit value of the electricity price of the future time period includes:

[0072] Determining the electricity price range corresponding to the electricity price category label according to the electricity price category label of the future time period;

[0073] The lower limit value and the upper limit value of the electricity price in the future time period are calculated according to the electricity price range corresponding to the electricity price category label and the real electricity value corresponding to the electricity price range.

[0074] In an optional implementation, obtaining the final electricity value of the target energy storage power station in the future time period according to the electricity price forecast value and the lower limit and upper limit of the electricity price in the future time period includes:

[0075] Comparing the electricity price forecast value with the lower limit value and the upper limit value of the electricity price in the future time period respectively;

[0076] If the electricity price forecast value is greater than or equal to the electricity price lower limit value and less than or equal to the electricity price upper limit value, the electricity price forecast value is used as the final electricity value of the target energy storage power station in the future time period;

[0077] If the electricity price forecast value is less than the electricity price lower limit value, the electricity price lower limit value is used as the final electricity value of the target energy storage power station in the future time period;

[0078] If the electricity price forecast value is greater than the electricity price upper limit value, the electricity price upper limit value is used as the final electricity value of the target energy storage power station in the future time period.

[0079] The multi-mode intelligent fusion electricity price prediction solution provided in the embodiment of the present application enhances flexibility and adaptability by classifying electricity price ranges through constructing an electricity price classification model; by constructing an electricity price classification model and an electricity price prediction model, the advantages of the classification model and the prediction model are integrated, the limitations of a single model are reduced, and the accuracy and stability of electricity price prediction are improved, especially when the electricity price fluctuates greatly, the electricity price prediction results are optimized, and ultimately a more accurate electricity price output value is achieved.

[0080] The following combination Figure 3 Taking the application of the multi-mode intelligent fusion electricity price prediction method provided by this application in the field of electricity price prediction as an example, the multi-mode intelligent fusion electricity price prediction method is further explained. Among them, Figure 3 A processing flow chart of a multi-mode intelligent fusion electricity price prediction method provided by an embodiment of the present application is shown, which specifically includes the following steps.

[0081] Step 301: data collection and processing;

[0082] Collect multi-dimensional data sources, including but not limited to historical electricity price data t , meteorological data (such as temperature T t 、Humidity t , wind speed W t ), load data (such as power demand D t 、Industrial electricity dataI t ) etc. The historical electricity price data can be expressed as a time series: Y = [y t-n ,y t-n+1 ,…,y t ]Where n represents the total amount of data; t represents time.

[0083] Step 302: constructing an electricity price classification model;

[0084] The purpose of the electricity price classification model is to classify the predicted electricity price into predefined electricity price ranges in order to better control the final electricity price output. The specific steps are as follows:

[0085] Define the electricity price range:

[0086] According to the analysis of historical electricity price data, the electricity price is divided into multiple intervals [L j ,U j ], where j represents different intervals, and the value is 1, 2, or 3. For example, the following electricity price intervals can be set:

[0087] Low electricity price range: [0,0.2]

[0088] China electricity price range: [0.2,0.5]

[0089] High electricity price range: [0.5,1.0]

[0090] The division of these intervals can be based on the distribution of historical electricity prices and dynamically adjusted in combination with market demand and policy factors.

[0091] Specifically, the electricity price is divided into intervals ranging from 0 to 1, and three electricity price intervals of low, medium and high are obtained. It should be pointed out that the lower limit value and the upper limit value in the interval do not represent the electricity price.

[0092] Annotation category:

[0093] Assign a corresponding category label C to each electricity price range j ,For example:

[0094] Low electricity price range: category label C1;

[0095] China electricity price range: category label C2;

[0096] High electricity price range: category label C3.

[0097] These category labels will be used to train the classification model to help the model understand the meaning of each electricity price range.

[0098] Feature and label construction:

[0099] Using feature set X t As input, the class label C j Model training is performed as output. Feature set X t This can include a number of factors, such as:

[0100] Time characteristics: such as hours t , Week W t Holidays t .

[0101] Environmental characteristics: such as temperature T t 、Humidity t , wind speed W t wait.

[0102] Load characteristics: Load data of the previous few days or hours D t , Industrial electricity data I t .

[0103] Feature set construction example: X t =[h t ,W t , Holidays t ,T t ,H t ,W t ,D t-1 ,D t-2 ,I t-1 ,I t-2 ].

[0104] Train the electricity price classification model:

[0105] Choose a suitable classification algorithm (such as support vector machine, decision tree, random forest or deep learning model), that is, choose according to the actual situation of the project, for example: according to the magnitude of existing data (how much data), the data type of existing data, numerical / categorical, the length of the final prediction time range, etc. These models are all traditional models, and the selection varies from person to person, time to time, and place to place. The core is to have a classification algorithm. Train the classification model. The input of the model is the feature set X t , the output is the category label C j :C=Classify(X t ).

[0106] The goal of the classification model is to t The predicted electricity price range category C j . You can use the cross-extraction loss function to optimize the model: Loss Among them, y im is the true category label (one-hot encoding), is the predicted probability, and M is the total number of categories.

[0107] Model Evaluation:

[0108] Use the test set to evaluate the classification model, and measure the performance of the model through indicators such as accuracy, recall, and F1 score: Accuracy = TP + TNN, where TP is the true positive example, TN is the true negative example, and N is the total number of samples.

[0109] Model Evaluation:

[0110] Use the test set to evaluate the classification model and measure the performance of the model through indicators such as accuracy, recall, and F1 score: Among them, TP is a true positive example, TN is a true negative example, and N is the total number of samples.

[0111] Step 303: constructing an electricity price prediction model;

[0112] 2. Build an electricity price prediction model

[0113] The main goal of the electricity price forecasting model is to accurately predict future electricity prices based on historical data and multiple influencing factors. The construction process includes the following steps:

[0114] Feature Engineering:

[0115] In order to improve the predictive ability of the model, it is necessary to construct an effective feature set X t Features may include:

[0116] Time characteristics: such as hours t , Week W t Holidays t .

[0117] Environmental characteristics: such as temperature T t 、Humidity t , wind speed W t wait.

[0118] Load characteristics: Load data of the previous few days or hours D t , Industrial electricity data I t .

[0119] Feature set construction example: X t =[h t ,W t , Holidays t ,T t ,H t ,W t ,D t-1 ,D t-2 ,I t-1 ,I t-2 ].

[0120] Model selection:

[0121] The present invention adopts long short-term memory network (LSTM) or deep neural network (DNN) as the main model to realize electricity price prediction. LSTM can effectively capture the temporal dependence of electricity price data due to its advantages in processing time series data, and is suitable for the dynamic fluctuation of electricity prices in the power market. On the other hand, DNN has a strong nonlinear modeling ability, can extract complex features through a multi-layer network structure, and has strong adaptability.

[0122] In specific implementation, you can choose a suitable model based on the characteristics of the data and prediction requirements:

[0123] LSTM model: suitable for time series data that requires long-term dependencies.

[0124] DNN model: suitable for situations where there are many input features and nonlinear relationships need to be captured.

[0125] The selection process includes preliminary design of the model architecture and performance comparison of multiple models to find the best predictive model.

[0126] Model building and training:

[0127] LSTM model construction:

[0128] The basic structure of LSTM includes an input layer, multiple LSTM layers, and an output layer. The input of the model is the feature set X t , the output is the predicted value P of future electricity price.

[0129] DNN model construction:

[0130] The DNN model consists of multiple fully connected layers, and the output of each layer is transformed nonlinearly through activation functions (such as ReLU, Sigmoid, etc.). The basic architecture of the model can be expressed as: (l) =σ(W (l) ·h (l-1) +b (l) ), where h (l) is the output of the lth layer, W (l) is the weight of the layer, b (l) is the bias and σ is the activation function.

[0131] Definition and optimization of loss function:

[0132] The goal of model training is to optimize model parameters by minimizing the loss function. The mean square error (MSE) is used as the loss function, and its calculation formula is: Among them, P i is the model prediction value, y i is the actual electricity price, and n is the number of samples.

[0133] During the training process, backpropagation and optimization algorithms (such as Adam and SGD) are used to update model parameters to reduce prediction errors.

[0134] Iteration of the training process:

[0135] In each training iteration, the predictions are calculated by forward propagation and the gradient is calculated according to the loss function. The weights are updated using an optimization algorithm until the convergence condition is met or the preset number of iterations is reached.

[0136] During the training process, the cross-validation method is used to evaluate the generalization ability of the model to ensure that the model performs well on unknown data. The loss value of the validation set is monitored to prevent the model from overfitting.

[0137] Model Evaluation:

[0138] After training is completed, the model is evaluated using the test set to compare the difference between the predicted results and the actual electricity price. The performance of the model is evaluated by calculating indicators such as MSE and root mean square error (RMSE):

[0139]

[0140] Model predictions:

[0141] After training, the model can output the predicted value P of future electricity prices, expressed as: P = f(Xt)

[0142] Among them, f represents the trained model function, and P is the future electricity price prediction based on the current feature set Xt.

[0143] Step 304: integrating the electricity price classification result with the electricity price prediction result;

[0144] Step 305: Output the electricity price fusion prediction result.

[0145] After obtaining the electricity price prediction value P and the classification result C, the process of fusion of the results to determine the final electricity price output E is as follows:

[0146] Interval determination:

[0147] According to the interval C output by the electricity price classification model j Determine the upper and lower limits: L j ,U j =GetLimits(C j )

[0148] For example, for the electricity price range C j , the lower and upper bounds can be expressed as follows: u j =L j +Δ j Among them, Δ j is the width of the jth interval, J is the total number of intervals, min(Y) is the minimum value in the historical electricity price data, and j is the interval.

[0149] Final electricity price output:

[0150] According to the comparison between the predicted value P and the upper and lower limits of the interval, the final electricity price output value E is calculated as follows:

[0151]

[0152] Where P i is the predicted value of electricity price, L j ,U j P i The corresponding electricity price range prediction results.

[0153] This method ensures that the final output electricity price E is always within a reasonable range.

[0154] The multi-mode intelligent fusion electricity price prediction solution provided in the embodiment of the present application aims to construct an electricity price classification model and an electricity price prediction model, and determine the final electricity price data for the future time period by integrating the output results of the two models, thereby improving the accuracy and stability of electricity price prediction and ultimately achieving a more accurate electricity price output value.

[0155] This application also provides an embodiment of a multi-mode intelligent fusion electricity price prediction device, Figure 2 The schematic diagram of the structure of a multi-mode intelligent fusion electricity price prediction device provided by an embodiment of the present application is shown. Figure 2 As shown, the device comprises:

[0156] An acquisition module 201 is configured to acquire a data set of a future time period of a target energy storage power station;

[0157] The first prediction module 202 is configured to obtain the electricity price category label of the future time period by inputting the data set of the future time period into a pre-trained electricity price classification model;

[0158] The second prediction module 203 is configured to obtain a predicted electricity price value for the future time period by inputting the data set for the future time period into a pre-trained electricity price prediction model;

[0159] The electricity price fusion module 204 is configured to obtain the final electricity value of the target energy storage power station in the future time period by fusing the electricity price category label with the electricity price forecast value.

[0160] The acquisition module 201 is further configured to:

[0161] Collect historical data sets of each energy storage power station, wherein the historical data sets at least include time series corresponding to electricity price data, meteorological data and load data; wherein the meteorological data at least includes temperature, humidity and wind speed; and the load data at least includes power demand data and industrial power consumption data.

[0162] The first prediction module 202 is further configured to:

[0163] By analyzing the electricity price data in the historical data set of each energy storage power station, the electricity price data is divided into multiple electricity price intervals from small to large and non-overlapping, and an electricity price category label is configured for each electricity price interval;

[0164] Divide the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of first sub-historical data sets, and configure a corresponding electricity price category label for each first sub-historical data set according to the plurality of electricity price intervals and the electricity price data in each first sub-historical data set;

[0165] An electricity price classification model is constructed, and each of the first sub-historical data sets is used as a model input, and the electricity price classification model is trained using the corresponding electricity price category label configured for each of the first sub-historical data sets as a model output to obtain a trained electricity price classification model.

[0166] The second prediction module 203 is further configured to:

[0167] Dividing the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of second sub-historical data sets;

[0168] An electricity price prediction model is constructed, and each of the second sub-historical data sets is used as a model input, and the electricity price prediction model is trained using the electricity price data corresponding to each of the second sub-historical data sets as a model output to obtain a trained electricity price prediction model.

[0169] The electricity price fusion module 204 is further configured to:

[0170] Determine the lower limit value and the upper limit value of the electricity price for the future time period by using the electricity price category label for the future time period;

[0171] According to the electricity price forecast value and the lower limit value and the upper limit value of the electricity price in the future time period, the final electricity value of the target energy storage power station in the future time period is obtained.

[0172] The electricity price fusion module 204 is further configured to:

[0173] Determining the electricity price range corresponding to the electricity price category label according to the electricity price category label of the future time period;

[0174] The lower limit value and the upper limit value of the electricity price in the future time period are calculated according to the electricity price range corresponding to the electricity price category label and the real electricity value corresponding to the electricity price range.

[0175] The electricity price fusion module 204 is further configured to:

[0176] Comparing the electricity price forecast value with the lower limit value and the upper limit value of the electricity price in the future time period respectively;

[0177] If the electricity price forecast value is greater than or equal to the electricity price lower limit value and less than or equal to the electricity price upper limit value, the electricity price forecast value is used as the final electricity value of the target energy storage power station in the future time period;

[0178] If the electricity price forecast value is less than the electricity price lower limit value, the electricity price lower limit value is used as the final electricity value of the target energy storage power station in the future time period;

[0179] If the electricity price forecast value is greater than the electricity price upper limit value, the electricity price upper limit value is used as the final electricity value of the target energy storage power station in the future time period.

[0180] The above is a schematic scheme of a multi-mode intelligent fusion electricity price prediction device of this embodiment. It should be noted that the technical scheme of the multi-mode intelligent fusion electricity price prediction device and the technical scheme of the multi-mode intelligent fusion electricity price prediction method mentioned above belong to the same concept. For the details not described in detail in the technical scheme of the multi-mode intelligent fusion electricity price prediction device, please refer to the description of the technical scheme of the multi-mode intelligent fusion electricity price prediction method mentioned above.

[0181] Figure 4 The block diagram of a computing device 400 according to an embodiment of the present application is shown. The components of the computing device 400 include but are not limited to a memory 410 and a processor 420. The processor 420 is connected to the memory 410 via a bus 430, and the database 450 is used to store data.

[0182] The computing device 400 also includes an access device 440 that enables the computing device 400 to communicate via one or more networks 460. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 440 may include one or more of any type of network interface (e.g., a network interface card (NIC)) wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Wide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0183] In one embodiment of the present application, the above components of the computing device 400 and Figure 4 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 4The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of the present application. Those skilled in the art may add or replace other components as needed.

[0184] Computing device 400 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. Computing device 400 may also be a mobile or stationary server.

[0185] The processor 420 is used to execute the following computer executable instructions, which, when executed by the processor, implement the steps of the above-mentioned multi-mode intelligent fusion electricity price prediction method.

[0186] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above-mentioned multi-mode intelligent fusion electricity price prediction method belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the above-mentioned multi-mode intelligent fusion electricity price prediction method.

[0187] An embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the above-mentioned multi-mode intelligent fusion electricity price prediction method.

[0188] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned multi-mode intelligent fusion electricity price prediction method belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the above-mentioned multi-mode intelligent fusion electricity price prediction method.

[0189] An embodiment of the present application further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned multi-mode intelligent fusion electricity price prediction method.

[0190] The above is a schematic scheme of a computer program of this embodiment. It should be noted that the technical scheme of the computer program and the technical scheme of the above-mentioned multi-mode intelligent fusion electricity price prediction method belong to the same concept, and the details not described in detail in the technical scheme of the computer program can be referred to the description of the technical scheme of the above-mentioned multi-mode intelligent fusion electricity price prediction method.

[0191] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0192] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0193] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described action sequence, because according to the embodiments of the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of the present application.

[0194] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0195] The preferred embodiments of the present application disclosed above are only used to help explain the present application. The optional embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of the present application. The present application selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of the present application, so that those skilled in the art can understand and use the present application well. The present application is only limited by the claims and their full scope and equivalents.

Claims

1. A multi-mode intelligent fusion electricity price prediction method, characterized in that: include: Obtain a data set of a future time period of a target energy storage power station; The electricity price category label of the future time period is obtained by inputting the data set of the future time period into a pre-trained electricity price classification model; The predicted value of the electricity price for the future time period is obtained by inputting the data set for the future time period into a pre-trained electricity price prediction model; The final electricity value of the target energy storage power station in the future time period is obtained by fusing the electricity price category label with the electricity price forecast value.

2. The method according to claim 1, characterized in that Also includes: Collect historical data sets of each energy storage power station, wherein the historical data sets at least include time series corresponding to electricity price data, meteorological data and load data; wherein the meteorological data at least includes temperature, humidity and wind speed; and the load data at least includes power demand data and industrial power consumption data.

3. The method according to claim 2, characterized in that The pre-trained electricity price classification model includes: By analyzing the electricity price data in the historical data set of each energy storage power station, the electricity price data is divided into multiple electricity price intervals from small to large and non-overlapping, and an electricity price category label is configured for each electricity price interval; Divide the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of first sub-historical data sets, and configure a corresponding electricity price category label for each first sub-historical data set according to the plurality of electricity price intervals and the electricity price data in each first sub-historical data set; An electricity price classification model is constructed, and each of the first sub-historical data sets is used as a model input, and the electricity price classification model is trained using the corresponding electricity price category label configured for each of the first sub-historical data sets as a model output to obtain a trained electricity price classification model.

4. The method according to claim 3, characterized in that The pre-trained electricity price prediction model includes: Dividing the time series corresponding to the meteorological data and the load data in the historical data set of each energy storage power station into a plurality of second sub-historical data sets; An electricity price prediction model is constructed, and each of the second sub-historical data sets is used as a model input, and the electricity price prediction model is trained using the electricity price data corresponding to each of the second sub-historical data sets as a model output to obtain a trained electricity price prediction model.

5. The method according to claim 4, characterized in that By fusing the electricity price category label with the electricity price forecast value, the final electricity value of the target energy storage power station in the future time period is obtained, including: Determine the lower limit value and the upper limit value of the electricity price for the future time period by using the electricity price category label for the future time period; According to the electricity price forecast value and the lower limit value and the upper limit value of the electricity price in the future time period, the final electricity value of the target energy storage power station in the future time period is obtained.

6. The method according to claim 5, characterized in that Determining the lower limit value and the upper limit value of the electricity price for the future time period by using the electricity price category label for the future time period includes: Determining the electricity price range corresponding to the electricity price category label according to the electricity price category label of the future time period; The lower limit value and the upper limit value of the electricity price in the future time period are calculated according to the electricity price range corresponding to the electricity price category label and the real electricity value corresponding to the electricity price range.

7. The method according to claim 5, characterized in that According to the electricity price forecast value and the lower limit and upper limit of the electricity price in the future time period, the final electricity value of the target energy storage power station in the future time period is obtained, which includes: Comparing the electricity price forecast value with the lower limit value and the upper limit value of the electricity price in the future time period respectively; If the electricity price forecast value is greater than or equal to the electricity price lower limit value and less than or equal to the electricity price upper limit value, the electricity price forecast value is used as the final electricity value of the target energy storage power station in the future time period; If the electricity price forecast value is less than the electricity price lower limit value, the electricity price lower limit value is used as the final electricity value of the target energy storage power station in the future time period; If the electricity price forecast value is greater than the electricity price upper limit value, the electricity price upper limit value is used as the final electricity value of the target energy storage power station in the future time period.

8. A multi-mode intelligent fusion electricity price prediction device, characterized in that: include: An acquisition module is configured to acquire a data set of a future time period of a target energy storage power station; A first prediction module is configured to obtain an electricity price category label for the future time period by inputting the data set for the future time period into a pre-trained electricity price classification model; A second prediction module is configured to obtain a predicted electricity price value for the future time period by inputting the data set for the future time period into a pre-trained electricity price prediction model; The electricity price fusion module is configured to obtain the final electricity value of the target energy storage power station in the future time period by fusing the electricity price category label with the electricity price forecast value.

9. A computing device comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the multi-mode intelligent fusion electricity price prediction method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the multi-mode intelligent fusion electricity price prediction method described in any one of claims 1 to 7.

Images