A method and electronic device for predicting subway energy consumption

Abstract

This disclosure provides a method and electronic device for predicting subway energy consumption, used to improve the accuracy of energy consumption prediction. It includes: for any energy consumption prediction category in the same subway line, at first specified intervals, inputting the first historical energy consumption data of the energy consumption prediction category into the corresponding pre-trained subway energy consumption prediction model to obtain predicted energy consumption; wherein, the pre-trained subway energy consumption prediction model is obtained through the following method: for any energy consumption prediction category, at second specified intervals, performing feature expansion on the second historical energy consumption data corresponding to the energy consumption prediction category to obtain target historical energy consumption data, then clustering each energy consumption prediction category to obtain each target energy consumption prediction category; for any target energy consumption prediction category, fusing the target historical energy consumption data of each energy consumption prediction category belonging to the target energy consumption prediction category, and training the subway energy consumption prediction model based on the fused target historical energy consumption data.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular to a method and electronic device for predicting subway energy consumption. Background Technology

[0002] With rapid urban development, subways have become an indispensable mode of transportation. However, the macroclimate, tunnel background temperature, and passenger flow trends faced by the same subway line or station will change over time. Simultaneously, to adapt to these environmental changes, the subway system's air conditioning, lighting, and train operation will need to be modified accordingly. Furthermore, as society progresses, the daily, monthly, and seasonal energy consumption levels of each subway station will also evolve.

[0003] Typically, the location and environment of each station on a subway line are different, even if they are both underground or above ground. These stations also exhibit uncorrelated differences in depth, hydrology, station design, and passenger flow. Therefore, for each station, a regression model for energy consumption prediction based on historical energy consumption and surrounding environmental characteristics is ideally adjusted station-by-station. However, subway energy systems and stations usually begin operation simultaneously. This makes it difficult to obtain historical energy consumption data accumulated over a long period of operation specific to the target subway line, while historical time series from other lines or even other cities cannot fully cover the characteristics of the target subway line. Consequently, the accuracy of subway energy consumption prediction is relatively low. Summary of the Invention

[0004] The exemplary embodiments of this disclosure provide a method and electronic device for predicting subway energy consumption, which are used to improve the accuracy of subway energy consumption prediction.

[0005] The first aspect of this disclosure provides a method for predicting subway energy consumption, the method comprising:

[0006] For any energy consumption prediction category in the same subway line, every first specified time interval, the first historical energy consumption data corresponding to the energy consumption prediction category is input into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption.

[0007] The pre-trained subway energy consumption prediction model is obtained in the following way:

[0008] For any given energy consumption prediction category, at second specified intervals, second historical energy consumption data corresponding to the energy consumption prediction category is acquired, wherein the second historical energy consumption data includes historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and,

[0009] The target historical energy consumption data is obtained by using the aforementioned historical feature data to augment the features of the second historical energy consumption data.

[0010] Based on the target historical energy consumption data of each energy consumption prediction category, cluster the energy consumption prediction categories to obtain each target energy consumption prediction category;

[0011] For any target energy consumption prediction category, the target historical energy consumption data corresponding to each energy consumption prediction category in the target energy consumption prediction category are fused together, and the subway energy consumption prediction model is trained based on the fused target historical energy consumption data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category.

[0012] Each energy consumption prediction category is updated using each target energy consumption prediction category, and the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category is determined as a pre-trained subway energy consumption prediction model corresponding to each updated energy consumption prediction category.

[0013] In this embodiment, for any energy consumption prediction category within any given subway line, second historical energy consumption data corresponding to the predicted category is acquired at second specified intervals. Features are then augmented on each second historical energy consumption data set to obtain target historical energy consumption data. Based on this target historical energy consumption data, each energy consumption prediction category is clustered to obtain target energy consumption prediction categories. The target historical energy consumption data corresponding to energy consumption prediction categories belonging to the same category are then fused. Finally, the subway energy consumption prediction model is trained based on the fused target historical energy consumption data to obtain a trained subway energy consumption prediction model corresponding to the target energy consumption prediction category. Therefore, this embodiment improves the basic dataset for training new models by augmenting existing historical data, clustering different energy consumption prediction categories, and integrating data from similar categories. This further improves the training accuracy of the model, thus increasing the accuracy of subway energy consumption prediction. Furthermore, the reduced number of energy consumption prediction categories also reduces the number of corresponding models, further improving the efficiency of subway energy consumption prediction.

[0014] In one embodiment, the step of using the historical feature data to augment the second historical energy consumption data to obtain the target historical energy consumption data includes:

[0015] A preset algorithm is used to fuse the specified historical feature data from each of the historical feature data to obtain new feature data;

[0016] The new feature data is added to the second historical energy consumption data to obtain the target historical energy consumption data.

[0017] In this embodiment, a preset algorithm is used to fuse specified historical feature data to obtain new feature data, and the new feature data is added to the second historical energy consumption data, thereby increasing the basic data for model training and further improving the accuracy of the model.

[0018] In one embodiment, before clustering the energy consumption prediction categories based on the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category, the method further includes:

[0019] For any energy consumption prediction category, the target historical energy consumption data of the energy consumption prediction category is normalized to obtain normalized target historical energy consumption data, and the normalized target historical energy consumption data is determined as the target historical energy consumption data.

[0020] In this embodiment, before clustering each energy consumption prediction category, the target historical energy consumption data needs to be normalized. This eliminates the absolute quantitative differences among the data points in each target historical energy consumption dataset, further improving the model's accuracy and thus enhancing the accuracy of subway energy consumption prediction.

[0021] In one embodiment, before inputting the first historical energy consumption data corresponding to the energy consumption prediction category into a pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, the method further includes:

[0022] For any energy consumption prediction category, the first historical energy consumption data corresponding to the energy consumption prediction category is normalized to obtain the normalized first historical energy consumption data, and the normalized first historical energy consumption data is determined as the first historical energy consumption data.

[0023] After inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, the method further includes:

[0024] The predicted energy consumption is then restored to obtain the restored predicted energy consumption.

[0025] In this embodiment, before inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category, the first historical energy consumption data needs to be normalized. Furthermore, after obtaining the predicted energy consumption, the obtained predicted energy consumption needs to be restored. Since the model uses normalized data during training, in order to ensure accuracy, the data also needs to be normalized in practical applications. Moreover, to ensure the accuracy of the prediction results, the obtained predicted energy consumption needs to be restored, thereby ensuring the accuracy of the obtained predicted energy consumption.

[0026] In one embodiment, the clustering of each energy consumption prediction category based on the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category includes:

[0027] Using a preset clustering algorithm, the energy consumption prediction categories are clustered based on the target historical energy consumption data for each category to obtain the target energy consumption prediction categories; or,

[0028] Based on the target historical energy consumption data of each energy consumption prediction category, the correlation between each energy consumption prediction category is obtained, and two energy consumption prediction categories with a correlation greater than a specified threshold are added to the same target energy consumption prediction category.

[0029] In this embodiment, clustering algorithms or the correlation between different energy consumption prediction categories can be used to cluster the energy consumption prediction categories to obtain the target energy consumption prediction category. This makes the obtained target energy consumption prediction category more accurate.

[0030] A second aspect of this disclosure provides an electronic device, including a processor and a memory, wherein the processor and the memory are connected via a bus;

[0031] The memory stores a computer program, and the processor is configured to perform the following operations based on the computer program:

[0032] For any energy consumption prediction category in the same subway line, every first specified time interval, the first historical energy consumption data corresponding to the energy consumption prediction category is input into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption.

[0033] The pre-trained subway energy consumption prediction model is obtained in the following way:

[0034] For any given energy consumption prediction category, at second specified intervals, second historical energy consumption data corresponding to the energy consumption prediction category is acquired, wherein the second historical energy consumption data includes historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and,

[0035] The target historical energy consumption data is obtained by using the aforementioned historical feature data to augment the features of the second historical energy consumption data.

[0036] Based on the target historical energy consumption data of each energy consumption prediction category, cluster the energy consumption prediction categories to obtain each target energy consumption prediction category;

[0037] For any target energy consumption prediction category, the target historical energy consumption data corresponding to each energy consumption prediction category in the target energy consumption prediction category are fused together, and the subway energy consumption prediction model is trained based on the fused target historical energy consumption data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category.

[0038] Each energy consumption prediction category is updated using each target energy consumption prediction category, and the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category is determined as a pre-trained subway energy consumption prediction model corresponding to each updated energy consumption prediction category.

[0039] In one embodiment, the processor is specifically configured to perform feature augmentation on the second historical energy consumption data using the historical feature data to obtain target historical energy consumption data, specifically as follows:

[0040] A preset algorithm is used to fuse the specified historical feature data from each of the historical feature data to obtain new feature data;

[0041] The new feature data is added to the second historical energy consumption data to obtain the target historical energy consumption data.

[0042] In one embodiment, the processor is further configured to:

[0043] Before clustering the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category, for any energy consumption prediction category, the target historical energy consumption data of the energy consumption prediction category is normalized to obtain normalized target historical energy consumption data, and the normalized target historical energy consumption data is determined as the target historical energy consumption data.

[0044] In one embodiment, the processor is further configured to:

[0045] Before inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, for any energy consumption prediction category, the first historical energy consumption data corresponding to the energy consumption prediction category is normalized to obtain the normalized first historical energy consumption data, and the normalized first historical energy consumption data is determined as the first historical energy consumption data.

[0046] The process involves inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, and then performing a restoration process on the predicted energy consumption to obtain the restored predicted energy consumption.

[0047] In one embodiment, the processor performs clustering of the target historical energy consumption data based on each energy consumption prediction category to obtain each target energy consumption prediction category, specifically configured as follows:

[0048] Using a preset clustering algorithm, the energy consumption prediction categories are clustered based on the target historical energy consumption data for each category to obtain the target energy consumption prediction categories; or,

[0049] Based on the target historical energy consumption data of each energy consumption prediction category, the correlation between each energy consumption prediction category is obtained, and two energy consumption prediction categories with a correlation greater than a specified threshold are added to the same target energy consumption prediction category.

[0050] According to a third aspect provided in the embodiments of this disclosure, a computer storage medium is provided, the computer storage medium storing a computer program for performing the method as described in the first aspect. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of an applicable scenario according to one embodiment of the present disclosure;

[0053] Figure 2 This is a second schematic diagram illustrating an applicable scenario according to one embodiment of the present disclosure;

[0054] Figure 3 This is a third schematic diagram illustrating an applicable scenario according to one embodiment of the present disclosure;

[0055] Figure 4 This is a flowchart illustrating a method for training subway energy consumption prediction models corresponding to various energy consumption prediction categories according to an embodiment of the present disclosure.

[0056] Figure 5 One of the schematic diagrams illustrates the fusion of target historical energy consumption data according to an embodiment of this disclosure;

[0057] Figure 6 A second schematic diagram illustrating the fusion of target historical energy consumption data according to an embodiment of this disclosure;

[0058] Figure 7 This is a flowchart illustrating a method for predicting subway energy consumption according to an embodiment of the present disclosure.

[0059] Figure 8 A flowchart illustrating a method for predicting subway energy consumption according to an embodiment of this disclosure;

[0060] Figure 9 This is a subway energy consumption prediction device according to an embodiment of the present disclosure;

[0061] Figure 10 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0063] In this disclosure, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0064] The application scenarios described in this disclosure are for the purpose of more clearly illustrating the technical solutions of this disclosure and do not constitute a limitation on the technical solutions provided in this disclosure. Those skilled in the art will understand that with the emergence of new application scenarios, the technical solutions provided in this disclosure are also applicable to similar technical problems. In the description of this disclosure, unless otherwise stated, "multiple" means two or more.

[0065] In existing technologies, because the energy systems and stations of subway lines are put into operation simultaneously, it is difficult to accumulate historical energy consumption data over a long period of time specific to the target subway line. Furthermore, historical time series data from other lines or even other cities cannot fully cover the characteristics of the target subway line. Therefore, the accuracy of subway energy consumption prediction is relatively low.

[0066] Therefore, this disclosure provides a method for predicting subway energy consumption. For any given energy consumption prediction category, at second specified intervals, second historical energy consumption data corresponding to that category is acquired. Then, feature expansion is performed on each second historical energy consumption data to obtain target historical energy consumption data. Based on this target historical energy consumption data, each energy consumption prediction category is clustered to obtain target energy consumption prediction categories. The target historical energy consumption data corresponding to energy consumption prediction categories belonging to the same target energy consumption prediction category are then fused. Finally, the subway energy consumption prediction model is trained based on the fused target historical energy consumption data to obtain a trained subway energy consumption prediction model corresponding to the target energy consumption prediction category. Thus, this embodiment improves the basic dataset for training new models by expanding the features of existing historical data, clustering different energy consumption prediction categories, and integrating data from similar energy consumption prediction categories. This further improves the training accuracy of the model, thereby increasing the accuracy of subway energy consumption prediction. Furthermore, due to the reduction in the number of energy consumption prediction categories, the number of corresponding models is also reduced, further improving the efficiency of subway energy consumption prediction. The present invention will now be described in detail with reference to the accompanying drawings.

[0067] like Figure 1 The diagram illustrates an application scenario for a subway energy consumption prediction method, using electronic devices as servers as an example. This application scenario includes a terminal device 110 and a server 120. Server 120 can be implemented using a single server or multiple servers. Server 120 can be implemented using a physical server or a virtual server.

[0068] In one possible application scenario, for any energy consumption prediction category within the same subway line, every first specified time interval, server 120 inputs the first historical energy consumption data corresponding to the energy consumption prediction category into a pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, and sends the predicted energy consumption to terminal device 110 for display; wherein, the pre-trained subway energy consumption prediction model is obtained in the following manner: for any energy consumption prediction category, every second specified time interval, server 120 obtains the second historical energy consumption data corresponding to the energy consumption prediction category, wherein the second historical energy consumption data includes various historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and the second historical energy consumption data is feature-enlarged using the various historical feature data. The server 120 obtains target historical energy consumption data; then, based on the target historical energy consumption data of each energy consumption prediction category, the server 120 clusters each energy consumption prediction category to obtain each target energy consumption prediction category; for any target energy consumption prediction category, the server 120 merges the target historical energy consumption data corresponding to the energy consumption prediction category within the target energy consumption prediction category, and trains the subway energy consumption prediction model based on the merged target energy consumption historical data to obtain a trained subway energy consumption prediction model corresponding to the target energy consumption prediction category; then, the server 120 updates each energy consumption prediction category using each target energy consumption prediction category, and determines the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category as a pre-trained subway energy consumption prediction model corresponding to the updated energy consumption prediction category.

[0069] like Figure 2The diagram illustrates another application scenario of this application, which includes a terminal device 110, a server 120, and a memory 130. In one possible application scenario, for any energy consumption prediction category on the same subway line, every first specified time interval, the server 120 retrieves first historical energy consumption data from the memory 130, then inputs the first historical energy consumption data corresponding to the energy consumption prediction category into a pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, and sends the predicted energy consumption to the terminal device 110 for display; wherein, the pre-trained subway energy consumption prediction model is obtained in the following way: for any energy consumption prediction category, every second specified time interval, the server 120 retrieves second historical energy consumption data corresponding to the energy consumption prediction category from the memory 130, wherein the second historical energy consumption data includes various historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and uses the various historical feature data to perform feature expansion on the second historical energy consumption data to obtain the target historical energy consumption. The server 120 then clusters the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category. For any target energy consumption prediction category, the server 120 merges the target historical energy consumption data corresponding to the energy consumption prediction category within that target energy consumption prediction category, and trains the subway energy consumption prediction model based on the merged target historical energy consumption data to obtain a trained subway energy consumption prediction model corresponding to the target energy consumption prediction category. Then, the server 120 updates each energy consumption prediction category using each target energy consumption prediction category, and determines the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category as a pre-trained subway energy consumption prediction model corresponding to the updated energy consumption prediction category, and stores the pre-trained subway energy consumption prediction model corresponding to the updated energy consumption prediction category in the memory 130.

[0070] like Figure 3The diagram illustrates another application scenario of this application, which includes a terminal device 110 and a memory 130. In one possible application scenario, for any energy consumption prediction category on the same subway line, every first specified time interval, the terminal device 110 retrieves first historical energy consumption data from the memory 130, and then inputs the first historical energy consumption data corresponding to the energy consumption prediction category into a pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain and display the predicted energy consumption; wherein, the pre-trained subway energy consumption prediction model is obtained in the following way: for any energy consumption prediction category, every second specified time interval, the terminal device 110 retrieves second historical energy consumption data corresponding to the energy consumption prediction category from the memory 130, wherein the second historical energy consumption data includes historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and uses the historical feature data to perform feature expansion on the second historical energy consumption data to obtain target historical energy consumption data; then the terminal device 110... 10. Cluster the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category; for any target energy consumption prediction category, the terminal device 110 fuses the target historical energy consumption data corresponding to the energy consumption prediction category in the target energy consumption prediction category, and trains the subway energy consumption prediction model based on the fused target historical energy consumption data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category; then the terminal device 110 updates the target energy consumption prediction categories using each target energy consumption prediction category, and determines the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category as a pre-trained subway energy consumption prediction model corresponding to the updated energy consumption prediction category, and stores the pre-trained subway energy consumption prediction model corresponding to the updated energy consumption prediction category in the memory 130.

[0071] The description in this application focuses on a single terminal device 110, a single server 120, and a single memory 130. However, those skilled in the art should understand that the illustrated terminal device 110, server 120, and memory 130 are intended to illustrate the operation of the terminal device 110, server 120, and memory 130 involved in the technical solutions of this application, and do not imply any limitation on the number, type, or location of the terminal device 110, server 120, and memory 130. It should be noted that adding additional modules to or removing individual modules from the illustrated environment will not change the underlying concept of the exemplary embodiments of this application.

[0072] For example, terminal device 110 includes, but is not limited to: large visual screens, tablet computers, laptops, handheld computers, mobile internet devices (MID), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminal devices in industrial control, wireless terminal devices in autonomous driving, wireless terminal devices in smart grids, wireless terminal devices in transportation safety, wireless terminal devices in smart cities, or wireless terminal devices in smart homes, etc.; the terminal device may have a related client installed, which may be software (e.g., browsers, short video software, etc.), or web pages, mini-programs, etc.

[0073] It should be noted that the subway energy consumption prediction method proposed in this application is not only applicable to… Figure 1 , Figure 2 as well as Figure 3 The application scenarios shown can also be applied to any subway energy consumption prediction device.

[0074] The following describes an exemplary embodiment of the subway energy consumption prediction method of this application, in conjunction with the application scenarios described above and with reference to the accompanying drawings. It should be noted that the above application scenarios are only shown to facilitate understanding of the methods and principles of this application, and the implementation of this application is not limited in any way in this respect.

[0075] Before introducing the subway energy consumption prediction method in this application, we will first give a detailed introduction to the training method of the subway energy consumption prediction model corresponding to each energy consumption prediction category in this application.

[0076] like Figure 4 The diagram shown illustrates the training method for subway energy consumption prediction models corresponding to each energy consumption prediction category, including the following steps:

[0077] Step 401: For any energy consumption prediction category in the same subway line, every second specified time interval, obtain the second historical energy consumption data corresponding to the energy consumption prediction category, wherein the second historical energy consumption data includes each historical feature data and the historical energy consumption value corresponding to each historical feature data;

[0078] In this embodiment, the energy consumption prediction categories include, but are not limited to, power lighting energy consumption prediction categories, air conditioning energy consumption prediction categories, vehicle departure energy consumption prediction categories, and water energy consumption prediction categories. The energy consumption prediction categories described above are for illustrative purposes only. The energy consumption prediction categories in this embodiment can be set according to actual conditions, and this embodiment does not limit each energy consumption prediction category. For any given energy consumption prediction category, the format of its second historical feature data is shown in Table 1:

[0079]

[0080] Table 1

[0081] Each historical energy consumption data point represents energy consumption data from a different time period. The data value corresponding to each feature is the historical feature data described above. Furthermore, the features corresponding to each energy consumption prediction category may be the same or different.

[0082] It should be noted that the second specified duration in this embodiment can be set according to actual conditions, and this embodiment does not limit the specific value of the second specified duration. Furthermore, the features corresponding to each historical feature data in the second historical energy consumption data can be set according to actual conditions, and this embodiment does not limit them.

[0083] Step 402: Use the historical feature data to augment the second historical energy consumption data to obtain the target historical energy consumption data;

[0084] In one embodiment, the second historical energy consumption data is feature-enhanced in the following manner:

[0085] A preset algorithm is used to fuse the specified historical feature data in each of the historical feature data to obtain new feature data; the new feature data is added to the second historical energy consumption data to obtain the target historical energy consumption data.

[0086] The preset algorithms in this embodiment include, but are not limited to, multi-level difference and multi-level differentiation algorithms. Specific preset algorithms can be set according to actual conditions, and this embodiment does not limit the preset algorithms.

[0087] Taking Table 1 as an example, if the specified historical feature data in Table 1 is the historical feature data corresponding to Feature 1 and Feature 2, and the preset algorithm is to add the specified historical feature data, then the historical feature data corresponding to Feature 1 and Feature 2 will be added together to obtain new feature data. The new feature data is shown in Table 2.

[0088]

[0089] Table 2

[0090] As shown in Table 2, the new feature data obtained are the data in the data column corresponding to feature 5.

[0091] Step 403: Cluster the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category;

[0092] In one embodiment, the energy consumption prediction categories can be clustered in the following two ways to obtain each target energy consumption prediction category:

[0093] Method 1: Using a preset clustering algorithm, cluster the target energy consumption prediction categories based on the target historical energy consumption data of each energy consumption prediction category to obtain the target energy consumption prediction categories.

[0094] The clustering algorithms used in this embodiment include, but are not limited to, k-means clustering, DBSCAN (Density-Based Spatial Clustering of Applications with Noise), bi-kmeans, and OPTICS (Ordering points to identify the clustering structure). The specific clustering algorithm can be set according to the actual situation; this embodiment does not limit the clustering algorithm.

[0095] Method 2: Based on the target historical energy consumption data of each energy consumption prediction category, obtain the correlation between each energy consumption prediction category, and add two energy consumption prediction categories with a correlation greater than a specified threshold to the same target energy consumption prediction category.

[0096] The process of determining the correlation between different energy consumption prediction categories involves obtaining the covariance matrix between each category based on the target historical energy consumption data, and then using the covariance matrix to determine the correlation between the different energy consumption prediction categories.

[0097] For example, consider energy consumption prediction categories A, B, C, and D. If the correlation between energy consumption prediction categories A and C is determined to be greater than a specified threshold, then energy consumption prediction categories A and C are identified as belonging to the same target energy consumption prediction category and added to the same target energy consumption prediction category 1. If energy consumption prediction categories B and D are determined to belong to the same target energy consumption prediction category, then energy consumption prediction categories B and D are added to the same target energy consumption prediction category 2.

[0098] It should be noted that energy consumption prediction categories that do not belong to the same target energy consumption prediction category cannot be added to the same target energy consumption prediction category.

[0099] To further improve the accuracy of subway energy consumption prediction, in one embodiment, before performing step 403, the target historical energy consumption data is normalized to obtain normalized target historical energy consumption data, and the normalized target historical energy consumption data is determined as the target historical energy consumption data.

[0100] Specifically, in this embodiment, the normalization process is as follows: for any feature in any target historical energy consumption data, divide each historical feature data corresponding to the feature by a first target value to obtain the normalized historical feature data, and divide each historical energy consumption value in the target historical energy consumption data by a second target value to obtain the normalized historical energy consumption value. Based on the normalized historical feature data and the normalized historical energy consumption value, the normalized target historical energy consumption data is obtained.

[0101] Wherein, the first target value is the historical feature data with the largest value among the historical feature data corresponding to the feature in the target historical energy consumption data. The second target value is the historical energy consumption value with the largest value among all historical energy consumption values ​​in the target historical energy consumption data.

[0102] Step 404: For any target energy consumption prediction category, merge the target historical energy consumption data corresponding to the energy consumption prediction category in the target energy consumption prediction category, and train the subway energy consumption prediction model based on the merged target energy consumption historical data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category.

[0103] In one embodiment, the target historical energy consumption data corresponding to the energy consumption prediction category belonging to the target energy consumption prediction category are fused in the following two ways:

[0104] Method 1: If the historical feature data of each energy consumption prediction category belonging to the same target energy consumption prediction category have the same characteristics, then the target historical energy consumption data of each energy consumption prediction category are fused based on each feature.

[0105] For example, such as Figure 5 As shown in the figure, energy consumption prediction category 1 and energy consumption prediction category 2 belong to the same target energy consumption prediction category. It can be seen from the figure that the target historical energy consumption data of energy consumption prediction category 1 and energy consumption prediction category 2 contain the same features. Therefore, the historical feature data belonging to the same feature in the target historical energy consumption data of each target energy consumption prediction category are merged to obtain the merged target historical energy consumption data.

[0106] Method 2: If the historical feature data of different energy consumption prediction categories within the same target energy consumption prediction category are different, then the features are fused together, and the target historical energy consumption data of each energy consumption prediction category are fused together based on the fused features.

[0107] For example, such as Figure 6 As shown, taking energy consumption prediction category 3 and energy consumption prediction category 4 as examples where they belong to the same energy consumption prediction category. From Figure 5 As can be seen, the target historical energy consumption data of energy consumption prediction category 3 and the target historical energy consumption data of energy consumption prediction category 4 contain different features. Therefore, the features in the target historical energy consumption data of energy consumption prediction category 3 and the target historical energy consumption data of energy consumption prediction category 4 are first fused together. Then, based on the fused features, the target historical energy consumption data of energy consumption prediction category 3 and the target historical energy consumption data of energy consumption prediction category 4 are fused together to obtain the fused target historical energy consumption data.

[0108] In this embodiment, the subway energy consumption prediction model includes, but is not limited to, neural network models such as DNN (Deep Neural Networks), RNN (Recurrent Neural Networks), and wavelet CNN (Convolutional Neural Networks). The specific model selection can be set according to the actual situation. This embodiment does not limit the selection of the subway energy consumption prediction model.

[0109] The following section details the training method for training the subway energy consumption prediction model based on the fused historical target energy consumption data:

[0110] For any target energy consumption prediction category, the fused historical target energy consumption data corresponding to the target energy consumption prediction category is input into the subway energy consumption prediction model for feature extraction to obtain the predicted historical energy consumption value corresponding to each historical feature data in the fused historical target energy consumption data. Based on the predicted historical energy consumption value corresponding to each historical feature data and the historical energy consumption value, an error value is obtained. If the error value is greater than a specified error value, the specified model parameters of the subway energy consumption prediction model are adjusted, and the process of inputting the fused historical target energy consumption data corresponding to the target energy consumption prediction category into the subway energy consumption prediction model for feature extraction is returned until the error value is not greater than the specified error value. Then, the training of the subway energy consumption prediction model ends, and a trained subway energy consumption prediction model corresponding to the target energy consumption prediction category is obtained.

[0111] The adjustment method for the specified model parameters can be to increase or decrease the specified parameters by a fixed value each time. The adjustment methods for different specified parameters can be the same or different, and can be set according to the actual situation. This embodiment does not limit the adjustment method for the specified model parameters. Furthermore, the specified model parameters can also be set according to the actual situation; this embodiment does not limit the specified model parameters.

[0112] It should be noted that the specified error value in this embodiment can be set according to the actual situation, and this embodiment does not limit the specified error value.

[0113] Step 405: Update each energy consumption prediction category using each target energy consumption prediction category, and determine the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category as a pre-trained subway energy consumption prediction model corresponding to each updated energy consumption prediction category.

[0114] In one embodiment, updating the energy consumption prediction categories using each target energy consumption prediction category is done by determining each target energy consumption prediction category as an energy consumption prediction category.

[0115] After introducing the training methods for the subway energy consumption prediction models corresponding to each energy consumption prediction category, the following section provides a detailed introduction to the subway energy consumption prediction methods:

[0116] In one embodiment, for any energy consumption prediction category in the same subway line, at first specified intervals, the first historical energy consumption data corresponding to the energy consumption prediction category is input into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption.

[0117] In this embodiment, the first specified duration is less than the second specified duration. However, the specific values ​​of the first specified duration and the second specified duration can be set according to the actual situation. This embodiment does not limit the specific values ​​of the first specified duration and the second specified duration.

[0118] Furthermore, in this embodiment, both the first historical energy consumption data and the second historical energy consumption data are historical energy consumption data, only the corresponding time periods are different. For example, if the historical energy consumption data includes data from January 1, 2010 to December 30, 2012, then the data from January 1, 2010 to December 31, 2011 can be determined as the second historical energy consumption data, and the data from January 1, 2012 to December 30, 2012 can be determined as the first historical energy consumption data. The specific division method is not limited in this embodiment and can be set according to actual conditions.

[0119] To ensure the accuracy of subway energy consumption prediction, in one embodiment, before inputting the first historical energy consumption data into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category, the first historical energy consumption data is normalized to obtain normalized first historical energy consumption data, and the normalized first historical energy consumption data is determined as the first historical energy consumption data.

[0120] The normalization process is the same as that described above, and will not be repeated here.

[0121] To ensure that the predicted energy consumption is the true value, the predicted energy consumption is restored after it is obtained to obtain the restored predicted energy consumption.

[0122] The restoration process corresponds to the normalization process. In this embodiment, the predicted energy consumption is multiplied by the second target value to obtain the restored predicted energy consumption.

[0123] It should be noted that the normalization and restoration methods described in this embodiment are for illustrative purposes only and do not limit the scope of normalization and restoration methods. The specific normalization and restoration methods can be set according to the actual situation.

[0124] To further understand the technical solution of this disclosure, the following is in conjunction with... Figure 7 A detailed explanation may include the following steps:

[0125] Step 701: For any energy consumption prediction category in the same subway line, every second specified time interval, obtain the second historical energy consumption data corresponding to the energy consumption prediction category, wherein the second historical energy consumption data includes each historical feature data and the historical energy consumption value corresponding to each historical feature data;

[0126] Step 702: Use the historical feature data to augment the second historical energy consumption data to obtain the target historical energy consumption data;

[0127] Step 703: Normalize the target historical energy consumption data to obtain normalized target historical energy consumption data, and determine the normalized target historical energy consumption data as the target historical energy consumption data;

[0128] Step 704: Cluster the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category;

[0129] Step 705: For any target energy consumption prediction category, merge the target historical energy consumption data corresponding to the energy consumption prediction category in the target energy consumption prediction category, and train the subway energy consumption prediction model based on the merged target energy consumption historical data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category.

[0130] Step 706: Update each energy consumption prediction category using each target energy consumption prediction category, and determine the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category as a pre-trained subway energy consumption prediction model corresponding to each updated energy consumption prediction category.

[0131] Step 707: For any energy consumption prediction category in the same subway line, normalize the first historical energy consumption data corresponding to the energy consumption prediction category to obtain normalized first historical energy consumption data, and determine the normalized first historical energy consumption data as the first historical energy consumption data.

[0132] Step 708: Every first specified time interval, input the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption;

[0133] Step 709: Perform a restoration process on the predicted energy consumption to obtain the restored predicted energy consumption.

[0134] To better understand the technical solution of the present invention, such as Figure 8 The flowchart shown is for this embodiment. Taking any energy consumption prediction category within the same subway line as an example, firstly, historical energy consumption data is acquired, including first historical energy consumption data and second historical energy consumption data. Second historical energy consumption data is acquired every second specified time interval. Then, the subway energy consumption prediction model for each energy consumption prediction category is trained based on the second historical energy consumption data, resulting in a trained subway energy consumption prediction model corresponding to each energy consumption prediction category. Finally, the normalization coefficient and restoration coefficient are recorded.

[0135] Then, at each specified time interval, the first historical energy consumption data is acquired, and the first historical energy consumption data is normalized using a normalization coefficient. Then, the corresponding subway energy consumption prediction model is loaded, and the normalized first historical energy consumption data is input into the pre-trained subway energy consumption prediction model to obtain the predicted energy consumption. The predicted energy consumption is then restored based on the recorded restoration coefficient to obtain the restored predicted energy consumption.

[0136] The training of the subway energy consumption prediction model includes feature expansion, data normalization, energy consumption prediction category clustering, target historical energy consumption data fusion, and model training.

[0137] Based on the same disclosed concept, the subway energy consumption prediction method described above can also be implemented by a subway energy consumption prediction device. The effect of this subway energy consumption prediction device is similar to that of the aforementioned method, and will not be described again here.

[0138] Figure 9 This is a schematic diagram of a subway energy consumption prediction device according to an embodiment of the present disclosure.

[0139] like Figure 9 As shown, the subway energy consumption prediction device 900 disclosed herein may include an energy consumption prediction module 910 and a model training module 920.

[0140] The energy consumption prediction module 910 is used to input the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained metro energy consumption prediction model corresponding to the energy consumption prediction category at first specified intervals for any energy consumption prediction category in the same metro line, so as to obtain the predicted energy consumption.

[0141] Model training module 920 is used to obtain the pre-trained subway energy consumption prediction model in the following manner:

[0142] For any given energy consumption prediction category, at second specified intervals, second historical energy consumption data corresponding to the energy consumption prediction category is acquired, wherein the second historical energy consumption data includes historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and,

[0143] The target historical energy consumption data is obtained by using the aforementioned historical feature data to augment the features of the second historical energy consumption data.

[0144] Based on the target historical energy consumption data of each energy consumption prediction category, cluster the energy consumption prediction categories to obtain each target energy consumption prediction category;

[0145] For any target energy consumption prediction category, the target historical energy consumption data corresponding to each energy consumption prediction category in the target energy consumption prediction category are fused together, and the subway energy consumption prediction model is trained based on the fused target historical energy consumption data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category.

[0146] Each energy consumption prediction category is updated using each target energy consumption prediction category, and the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category is determined as a pre-trained subway energy consumption prediction model corresponding to each updated energy consumption prediction category.

[0147] In one embodiment, the model training module 920 performs feature augmentation on the second historical energy consumption data using the historical feature data to obtain target historical energy consumption data, specifically for:

[0148] A preset algorithm is used to fuse the specified historical feature data from each of the historical feature data to obtain new feature data;

[0149] The new feature data is added to the second historical energy consumption data to obtain the target historical energy consumption data.

[0150] In one embodiment, the apparatus further includes:

[0151] The first normalization module 930 is used to perform normalization processing on the target historical energy consumption data of each energy consumption prediction category before clustering the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category, and to determine the normalized target historical energy consumption data as the target historical energy consumption data.

[0152] In one embodiment, the apparatus further includes:

[0153] The second normalization module 940 is used to input the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption. Before that, for any energy consumption prediction category, the first historical energy consumption data corresponding to the energy consumption prediction category is normalized to obtain the normalized first historical energy consumption data, and the normalized first historical energy consumption data is determined as the first historical energy consumption data.

[0154] The restoration module 950 is used to input the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category, obtain the predicted energy consumption, and then perform restoration processing on the predicted energy consumption to obtain the restored predicted energy consumption.

[0155] In one embodiment, the model training module 920 performs clustering of the target historical energy consumption data based on each energy consumption prediction category to obtain each target energy consumption prediction category, specifically for:

[0156] Using a preset clustering algorithm, the energy consumption prediction categories are clustered based on the target historical energy consumption data for each category to obtain the target energy consumption prediction categories; or,

[0157] Based on the target historical energy consumption data of each energy consumption prediction category, the correlation between each energy consumption prediction category is obtained, and two energy consumption prediction categories with a correlation greater than a specified threshold are added to the same target energy consumption prediction category.

[0158] After introducing a subway energy consumption prediction method and apparatus according to an exemplary embodiment of the present disclosure, an electronic device according to another exemplary embodiment of the present disclosure will be introduced next.

[0159] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0160] In some possible implementations, the electronic device according to this disclosure may include at least one processor and at least one computer storage medium. The computer storage medium stores program code that, when executed by the processor, causes the processor to perform the steps in the subway energy consumption prediction method according to various exemplary embodiments of this disclosure described above. For example, the processor may perform actions such as... Figure 7 Steps 701-709 are shown.

[0161] The following reference Figure 10 To describe an electronic device 1000 according to such an embodiment of the present disclosure. Figure 10 The electronic device 1000 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0162] like Figure 10 As shown, the electronic device 1000 is manifested in the form of a general electronic device. The components of the electronic device 1000 may include, but are not limited to: at least one processor 1001, at least one computer storage medium 1002, and a bus 1003 connecting different system components (including the computer storage medium 1002 and the processor 1001).

[0163] Bus 1003 represents one or more of several bus structures, including computer storage media bus or computer storage media controller, peripheral bus, processor, or local bus using any of the various bus structures.

[0164] Computer storage medium 1002 may include readable media in the form of volatile computer storage media, such as random access computer storage medium (RAM) 1021 and / or cache storage medium 1022, and may further include read-only computer storage medium (ROM) 1023.

[0165] The computer storage medium 1002 may also include a program / utility 1025 having a set (at least one) of program modules 1024, such program modules 1024 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0166] Electronic device 1000 can also communicate with one or more external devices 1004 (e.g., keyboard, pointing device, etc.), one or more devices that enable a user to interact with electronic device 1000, and / or any device that enables electronic device 1000 to communicate with one or more other electronic devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 1005. Furthermore, electronic device 1000 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 1006. As shown, network adapter 1006 communicates with other modules used in electronic device 1000 via bus 1003. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 1000, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0167] In some possible implementations, various aspects of the subway energy consumption prediction method provided in this disclosure can also be implemented in the form of a program product, which includes program code that, when the program product is run on a computer device, causes the computer device to perform the steps in the subway energy consumption prediction method according to various exemplary embodiments of this disclosure as described above.

[0168] The program product may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access computer storage media (RAM), read-only computer storage media (ROM), erasable programmable read-only computer storage media (EPROM or flash memory), optical fibers, portable compact disk read-only computer storage media (CD-ROM), optical computer storage media, magnetic computer storage media, or any suitable combination thereof.

[0169] The program product for predicting subway energy consumption according to embodiments of this disclosure can be a portable compact disc read-only computer storage medium (CD-ROM) and include program code, and can run on an electronic device. However, the program product of this disclosure is not limited thereto. In this document, the readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0170] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying readable program code. This propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0171] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0172] Program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's electronic device, partially on the user's device, as a standalone software package, partially on the user's electronic device and partially on a remote electronic device, or entirely on a remote electronic device or server. In cases involving remote electronic devices, the remote electronic device can be connected to the user's electronic device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external electronic device (e.g., via the Internet using an Internet service provider).

[0173] It should be noted that although several modules of the apparatus have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules described above can be embodied in one module. Conversely, the features and functions of one module described above can be further divided and embodied by multiple modules.

[0174] Furthermore, although the operations of the methods disclosed herein are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all of the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0175] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk computer storage media, CD-ROMs, optical computer storage media, etc.) containing computer-usable program code.

[0176] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0177] These computer program instructions may also be stored in a computer-readable computer storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable computer storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0178] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0179] Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from its spirit and scope. Therefore, if such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, this disclosure is also intended to include such modifications and variations.

Claims

1. A method for predicting subway energy consumption, characterized in that, The method includes: For any energy consumption prediction category in the same subway line, every first specified time interval, the first historical energy consumption data corresponding to the energy consumption prediction category is input into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption. The pre-trained subway energy consumption prediction model is obtained in the following way: For any given energy consumption prediction category, at second specified intervals, second historical energy consumption data corresponding to the energy consumption prediction category is acquired, wherein the second historical energy consumption data includes historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and, The target historical energy consumption data is obtained by using the aforementioned historical feature data to augment the features of the second historical energy consumption data. Based on the target historical energy consumption data of each energy consumption prediction category, the energy consumption prediction categories are clustered to obtain each target energy consumption prediction category; For any target energy consumption prediction category, the target historical energy consumption data corresponding to each energy consumption prediction category in the target energy consumption prediction category are fused together, and the subway energy consumption prediction model is trained based on the fused target historical energy consumption data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category. Each energy consumption prediction category is updated using each target energy consumption prediction category, and the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category is determined as a pre-trained subway energy consumption prediction model corresponding to each updated energy consumption prediction category.

2. The method according to claim 1, characterized in that, The step of using the historical feature data to augment the second historical energy consumption data to obtain the target historical energy consumption data includes: A preset algorithm is used to fuse the specified historical feature data from each historical feature data to obtain new feature data; The new feature data is added to the second historical energy consumption data to obtain the target historical energy consumption data.

3. The method according to claim 1, characterized in that, Before clustering the target energy consumption prediction categories based on the historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category, the method further includes: For any energy consumption prediction category, the target historical energy consumption data of the energy consumption prediction category is normalized to obtain normalized target historical energy consumption data, and the normalized target historical energy consumption data is determined as the target historical energy consumption data.

4. The method according to claim 1, characterized in that, Before inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, the method further includes: For any energy consumption prediction category, the first historical energy consumption data corresponding to the energy consumption prediction category is normalized to obtain the normalized first historical energy consumption data, and the normalized first historical energy consumption data is determined as the first historical energy consumption data. After inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, the method further includes: The predicted energy consumption is then restored to obtain the restored predicted energy consumption.

5. The method according to claim 1, characterized in that, The target energy consumption prediction categories are clustered based on the historical energy consumption data of each category to obtain each target energy consumption prediction category, including: Using a preset clustering algorithm, the energy consumption prediction categories are clustered based on the target historical energy consumption data for each category to obtain the target energy consumption prediction categories; or, Based on the target historical energy consumption data of each energy consumption prediction category, the correlation between each energy consumption prediction category is obtained, and two energy consumption prediction categories with a correlation greater than a specified threshold are added to the same target energy consumption prediction category.

6. An electronic device, characterized in that, It includes a processor and a memory, which are connected via a bus; The memory stores a computer program, and the processor is configured to perform the following operations based on the computer program: For any energy consumption prediction category in the same subway line, every first specified time interval, the first historical energy consumption data corresponding to the energy consumption prediction category is input into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption. The pre-trained subway energy consumption prediction model is obtained in the following way: For any given energy consumption prediction category, at second specified intervals, second historical energy consumption data corresponding to the energy consumption prediction category is acquired, wherein the second historical energy consumption data includes historical feature data and historical energy consumption values ​​corresponding to each historical feature data; and, The target historical energy consumption data is obtained by using the aforementioned historical feature data to augment the features of the second historical energy consumption data. Based on the target historical energy consumption data of each energy consumption prediction category, the energy consumption prediction categories are clustered to obtain each target energy consumption prediction category; For any target energy consumption prediction category, the target historical energy consumption data corresponding to each energy consumption prediction category in the target energy consumption prediction category are fused together, and the subway energy consumption prediction model is trained based on the fused target historical energy consumption data to obtain the trained subway energy consumption prediction model corresponding to the target energy consumption prediction category. Each energy consumption prediction category is updated using each target energy consumption prediction category, and the trained subway energy consumption prediction model corresponding to each target energy consumption prediction category is determined as a pre-trained subway energy consumption prediction model corresponding to each updated energy consumption prediction category.

7. The electronic device according to claim 6, characterized in that, The processor is specifically configured to perform feature augmentation on the second historical energy consumption data using the historical feature data to obtain the target historical energy consumption data. A preset algorithm is used to fuse the specified historical feature data from each historical feature data to obtain new feature data; The new feature data is added to the second historical energy consumption data to obtain the target historical energy consumption data.

8. The electronic device according to claim 6, characterized in that, The processor is also configured to: Before clustering the target historical energy consumption data of each energy consumption prediction category to obtain each target energy consumption prediction category, for any energy consumption prediction category, the target historical energy consumption data of the energy consumption prediction category is normalized to obtain normalized target historical energy consumption data, and the normalized target historical energy consumption data is determined as the target historical energy consumption data.

9. The electronic device according to claim 6, characterized in that, The processor is also configured to: Before inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, for any energy consumption prediction category, the first historical energy consumption data corresponding to the energy consumption prediction category is normalized to obtain the normalized first historical energy consumption data, and the normalized first historical energy consumption data is determined as the first historical energy consumption data. The process involves inputting the first historical energy consumption data corresponding to the energy consumption prediction category into the pre-trained subway energy consumption prediction model corresponding to the energy consumption prediction category to obtain the predicted energy consumption, and then performing a restoration process on the predicted energy consumption to obtain the restored predicted energy consumption.

10. The electronic device according to claim 6, characterized in that, The processor executes the clustering of the target historical energy consumption data based on each energy consumption prediction category to obtain each target energy consumption prediction category, specifically configured as follows: Using a preset clustering algorithm, the energy consumption prediction categories are clustered based on the target historical energy consumption data for each category to obtain the target energy consumption prediction categories; or, Based on the target historical energy consumption data of each energy consumption prediction category, the correlation between each energy consumption prediction category is obtained, and two energy consumption prediction categories with a correlation greater than a specified threshold are added to the same target energy consumption prediction category.

Images