Enterprise energy consumption data prediction method, system and equipment

Abstract

The invention discloses an enterprise energy consumption data prediction method, system and equipment, belongs to the technical field of energy consumption prediction, and is used for solving the following technical problems that the existing energy consumption data prediction technology mainly has the defects of excessive dependence on a complete data set and poor non-linear data processing capability. The method comprises the steps of preprocessing pre-collected enterprise energy consumption data based on a K-means algorithm; performing feature extraction on the preprocessed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least comprise time features and statistical features; constructing an energy consumption prediction model, and training the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data features; in the training process, based on a grid search method, determining optimal parameters of the model to obtain an optimized energy consumption prediction model; and inputting the latest energy consumption data of the target enterprise into the optimized energy consumption prediction model to obtain an energy consumption prediction result in a future preset time period.

Description

Technical Field

[0001] The present invention relates to the technical field of energy consumption prediction, and in particular to a method, system and device for predicting enterprise energy consumption data. Background Art

[0002] With the rapid development of smart grid and Internet of Things technology, the collection and analysis of energy consumption data has become increasingly important. By predicting energy consumption data, enterprises can foresee future energy consumption trends, thereby providing data support for their energy use decisions, reducing unnecessary energy waste, achieving energy conservation and emission reduction, and reducing enterprise operating costs.

[0003] At present, traditional energy consumption data analysis and prediction methods mainly include statistical methods and machine learning methods. Common technical solutions include linear regression, time series analysis (such as ARIMA model) and simple neural network models. These methods usually rely on complete data sets, and current technologies are limited in dealing with missing values ​​and outliers in data sets. In addition, traditional methods such as ARIMA model and linear regression are mainly applicable to linear data, and have poor processing capabilities for nonlinear data, resulting in certain limitations in the actual energy consumption prediction process.

[0004] Therefore, the current energy consumption data prediction technology mainly has the disadvantages of over-reliance on complete data sets and poor ability to process nonlinear data. Summary of the invention

[0005] The embodiments of the present invention provide a method, system and device for predicting enterprise energy consumption data, which are used to solve the following technical problems: the current energy consumption data prediction technology mainly has the disadvantages of over-reliance on complete data sets and poor ability to process nonlinear data.

[0006] The embodiment of the present invention adopts the following technical solutions:

[0007] On the one hand, an embodiment of the present invention provides a method for predicting enterprise energy consumption data, the method comprising: preprocessing pre-collected enterprise energy consumption data based on a K-means algorithm;

[0008] Extracting features from the pre-processed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least include time features and statistical features;

[0009] Constructing an energy consumption prediction model, and training the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data features;

[0010] During the training process, the optimal parameters of the model are determined based on the grid search method to obtain the optimized energy consumption prediction model;

[0011] The latest energy consumption data of the target enterprise is input into the optimized energy consumption prediction model to obtain the energy consumption prediction result within a preset time period in the future.

[0012] In a feasible implementation, the pre-collected enterprise energy consumption data is pre-processed based on the K-means algorithm, specifically including:

[0013] Normalize the pre-collected enterprise energy consumption data to normalize the enterprise energy consumption data collected from different enterprises to the same scale;

[0014] The K-means algorithm is used to perform cluster analysis on the normalized enterprise energy consumption data, and the missing values ​​are filled to obtain a complete data set.

[0015] In a feasible implementation, the K-means algorithm is used to perform cluster analysis on the normalized enterprise energy consumption data and fill in the missing values ​​to obtain a complete data set, which specifically includes:

[0016] Determine data points containing missing values ​​in the normalized enterprise energy consumption data, and fill the missing values ​​with temporary values ​​to obtain a temporary data set;

[0017] Clustering the temporary data set using a K-means algorithm to divide all data points into K clusters;

[0018] Similar data points are searched in the cluster where the data point with missing values ​​is located, and the missing values ​​are calculated based on the similar data points, and the temporary values ​​filled in are replaced to obtain the complete data set.

[0019] In a feasible implementation, feature extraction is performed on the pre-processed enterprise energy consumption data to obtain energy consumption data features, specifically including:

[0020] Extracting the time feature from the pre-processed enterprise energy consumption data; wherein the time feature at least includes year, month, day, hour and week;

[0021] The statistical features are extracted from the pre-processed enterprise energy consumption data; wherein the statistical features at least include: the mean and variance of the enterprise energy consumption data.

[0022] In a feasible implementation, an energy consumption prediction model is constructed, and the energy consumption prediction model is trained by using the preprocessed enterprise energy consumption data and the energy consumption data features, specifically including:

[0023] Based on the support vector machine SVM model, the energy consumption prediction model is constructed; the energy consumption prediction model uses a radial basis function as a kernel function;

[0024] Aggregating the preprocessed enterprise energy consumption data and the energy consumption data features into a model training data set, and dividing the model training data set into N sub-data sets;

[0025] The energy consumption prediction model is trained using the N subset data sets.

[0026] In a feasible implementation manner, training the energy consumption prediction model by using the N subset data sets specifically includes:

[0027] Traversing the N sub-data sets, taking the first one as a test set and the remaining N-1 parts as training sets, training and testing the energy consumption prediction model to obtain a first test accuracy rate;

[0028] Continue to take the second part as the test set and the remaining N-1 parts as the training set, train and test the energy consumption prediction model, and so on, until the Nth test accuracy is obtained;

[0029] The obtained N test accuracy rates are averaged to obtain the algorithm accuracy value of the trained energy consumption prediction model;

[0030] If the algorithm accuracy value does not reach the preset threshold, the model parameters are adjusted and the model training is performed again.

[0031] In a feasible implementation, during the training process, the optimal parameters of the model are determined based on the grid search method to obtain an optimized energy consumption prediction model, which specifically includes:

[0032] During the training process of the energy consumption prediction model, if the algorithm accuracy value does not reach the preset threshold, the optimal model parameters of the energy consumption prediction model are found through a grid search optimization method; wherein the optimal model parameters at least include an optimal penalty coefficient and an optimal kernel function parameter;

[0033] Substitute the optimal parameters of the model into the energy consumption prediction model to obtain the optimized energy consumption prediction model.

[0034] In a feasible implementation manner, the loss function of the optimized energy consumption prediction model is:

[0035]

[0036] Among them, ω is the weight vector; C is the optimal penalty coefficient, which is used to balance the loss term and the regularization term; ξ i and is the slack variable, representing the error; b is the bias term.

[0037] On the other hand, an embodiment of the present invention further provides an enterprise energy consumption data prediction system, the system comprising:

[0038] A data preprocessing unit is used to preprocess the pre-collected enterprise energy consumption data based on the K-means algorithm;

[0039] A feature extraction unit is used to extract features from the pre-processed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least include time features and statistical features;

[0040] A model training unit is used to construct an energy consumption prediction model and train the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data characteristics; during the training process, based on the grid search method, the optimal parameters of the model are determined to obtain an optimized energy consumption prediction model;

[0041] The model prediction unit is used to input the latest energy consumption data of the target enterprise into the optimized energy consumption prediction model to obtain the energy consumption prediction result within a preset time period in the future.

[0042] Finally, an embodiment of the present invention further provides an enterprise energy consumption data prediction device, the device comprising:

[0043] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor so that the at least one processor can execute the enterprise energy consumption data prediction method.

[0044] Compared with the prior art, the enterprise energy consumption data prediction method, system and device provided by the embodiment of the present invention have the following beneficial effects:

[0045] 1. The present invention uses the K-means algorithm to cluster enterprise energy consumption data and uses similar data points to fill in missing values, which can more accurately restore the potential pattern of the data, improve the integrity and effectiveness of the data set, and solve the problem that the existing energy consumption data prediction scheme is overly dependent on the complete data set.

[0046] 2. The present invention particularly adopts the support vector machine (SVM) model to predict energy consumption data, uses the radial basis function kernel to process nonlinear data, and can process complex nonlinear relationships by calculating the similarity of the high-dimensional feature space between input samples. In the model prediction stage, the trained SVM model is used to predict the trend of future energy consumption data. By inputting the latest energy consumption data, the model can output the energy consumption forecast results for a period of time in the future, provide data support for the company's decision-making, and help the company achieve the goals of energy conservation, emission reduction and cost reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0048] Figure 1 A flow chart of a method for predicting enterprise energy consumption data provided by an embodiment of the present invention;

[0049] Figure 2 A schematic diagram of the structure of an enterprise energy consumption data prediction system provided by an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of the structure of an enterprise energy consumption data prediction device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0052] The embodiment of the present invention provides a method for predicting energy consumption data of an enterprise, such as Figure 1 As shown, the enterprise energy consumption data prediction method specifically includes steps S101-S105:

[0053] S101. Preprocess the pre-collected enterprise energy consumption data based on the K-means algorithm.

[0054] Specifically, firstly, a large amount of enterprise energy consumption data is pre-collected in different enterprise systems, and then the enterprise energy consumption data is normalized to normalize the enterprise energy consumption data collected from different enterprises to the same scale.

[0055] Furthermore, the K-means algorithm is used to perform cluster analysis on the normalized enterprise energy consumption data and fill in the missing values ​​to obtain a complete data set. The specific implementation method is as follows:

[0056] First, the data points with missing values ​​in the normalized enterprise energy consumption data are determined, and the missing values ​​are filled with temporary values ​​to obtain a temporary data set. Then, the temporary data set is clustered using the K-means algorithm to divide all data points into K clusters. Similar data points are found in the clusters where the data points with missing values ​​are located, and the missing values ​​are calculated based on the similar data points, and the filled temporary values ​​are replaced to obtain a complete data set.

[0057] As a feasible implementation method, in the present invention, the missing data is completed using the K-means algorithm, which mainly includes the following steps:

[0058] 1. Data preprocessing: Fill the missing values ​​in the dataset with temporary values ​​(such as mean or median) so that the K-means algorithm can run properly.

[0059] 2. K-means clustering: Perform K-means clustering on the padded data set to obtain K clusters and their centers.

[0060] 3. Find similar data points: For each data point with missing values, find similar data points that belong to the same cluster.

[0061] 4. Missing value completion: Use the feature mean or other statistical features of similar data points to fill in missing values.

[0062] The goal of the K-means algorithm is to minimize the objective function, which is Among them, C i is the i-th cluster, x is the cluster belonging to cluster C i The data point, u i is the center of the ith cluster, ‖xu i ‖ 2 Represents data point x and cluster center u i The Euclidean distance between .

[0063] By iteratively optimizing the objective function, the data points are assigned to the nearest cluster center and the cluster center position is updated. The main algorithm process is as follows:

[0064] 1. Initialization: Randomly select K data points as the initial cluster centers.

[0065] 2. Assign data points: Assign each data point to the cluster center closest to it.

[0066] 3. Update cluster centers: Recalculate the center of each cluster as the new cluster center.

[0067] 4. Iteration: Repeat steps 2-3 until the cluster center no longer changes or the change is less than the preset threshold.

[0068] The present invention uses the K-means algorithm to complete missing values, thereby restoring the potential pattern of the data and solving the problem that the existing energy consumption data prediction scheme is overly dependent on the complete data set.

[0069] S102. Extract features from the pre-processed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least include time features and statistical features.

[0070] Specifically, in the pre-processed enterprise energy consumption data, time features are extracted; wherein the time features at least include year, month, day, hour and week. In the pre-processed enterprise energy consumption data, statistical features are extracted; wherein the statistical features at least include: mean and variance of the enterprise energy consumption data.

[0071] These features can provide important information about the time dimension and statistical distribution of the data. The extracted features will be used as parameters for model training to enhance the model's predictive ability.

[0072] S103, constructing an energy consumption prediction model, and training the energy consumption prediction model through the pre-processed enterprise energy consumption data and energy consumption data features.

[0073] Specifically, an energy consumption prediction model is constructed based on a support vector machine (SVM) model; the energy consumption prediction model uses a radial basis function as a kernel function.

[0074] Furthermore, the preprocessed enterprise energy consumption data and energy consumption data features are summarized as a model training data set, and the model training data set is divided into N sub-data sets.

[0075] Furthermore, the energy consumption prediction model is trained through N-molecule datasets. The specific implementation process includes:

[0076] Traverse N sub-data sets, take the first one as the test set, and the remaining N-1 parts as the training set, train and test the energy consumption prediction model, and get the first test accuracy. Continue to take the second part as the test set, and the remaining N-1 parts as the training set, train and test the energy consumption prediction model, and so on, until the Nth test accuracy is obtained.

[0077] Then the N test accuracy rates are averaged to obtain the algorithm accuracy value of the trained energy consumption prediction model. If the algorithm accuracy value does not reach the preset threshold, the model parameters are adjusted and the model training is performed again until the algorithm accuracy value reaches the preset threshold and the training is stopped.

[0078] S104. During the training process, based on the grid search method, the optimal parameters of the model are determined to obtain an optimized energy consumption prediction model.

[0079] Specifically, during the training process of the energy consumption prediction model, if the algorithm accuracy value does not reach the preset threshold, the optimal model parameters of the energy consumption prediction model are found through the grid search optimization method; wherein the optimal model parameters at least include the optimal penalty coefficient and the optimal kernel function parameters. The optimal model parameters are substituted into the energy consumption prediction model to obtain the optimized energy consumption prediction model.

[0080] The loss function of optimizing the energy consumption prediction model is: Among them, ω is the weight vector; C is the optimal penalty coefficient, which is used to balance the loss term and the regularization term; ξ i and is the slack variable, representing the error; b is the bias term.

[0081] As a feasible implementation method, this scheme adopts the SVM model for data prediction, in which the kernel function adopts the radial basis function. The radial basis function can handle complex nonlinear relationships by calculating the similarity of the high-dimensional feature space between the input samples. SVM is a supervised learning algorithm that separates data points of different categories by finding an optimal hyperplane in high-dimensional space to achieve the purpose of classification or regression. The goal of SVM regression is to find a function f(x) so that the values ​​of all training data points that deviate from the function are within a tolerance range and make the function as smooth as possible.

[0082] When solving the objective function, the Lagrange multiplier can be introduced to obtain the Lagrangian function, and then the partial derivatives of the weight coefficient vector, bias term and slack variable in the Lagrangian function are respectively calculated, and the partial derivatives are set to 0 to obtain the objective function f(x) of the SVM.

[0083] S105: Input the latest energy consumption data of the target enterprise into the optimized energy consumption prediction model to obtain energy consumption prediction results within a preset time period in the future.

[0084] Specifically, when the target enterprise wants to obtain the energy consumption in a certain period of time in the future, it only needs to input the energy consumption data of the enterprise in the recent period into the optimized energy consumption prediction model, and the energy consumption prediction data in the future period can be output. The enterprise can adjust the subsequent energy use decision based on the energy consumption prediction data to achieve energy conservation and emission reduction.

[0085] In addition, the present invention also provides an enterprise energy consumption data prediction system. Figure 2 As shown, the enterprise energy consumption data prediction system 200 includes:

[0086] The data preprocessing unit 210 is used to preprocess the pre-collected enterprise energy consumption data based on the K-means algorithm;

[0087] The feature extraction unit 220 is used to extract features from the pre-processed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least include time features and statistical features;

[0088] The model training unit 230 is used to construct an energy consumption prediction model and train the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data features; during the training process, the optimal parameters of the model are determined based on the grid search method to obtain an optimized energy consumption prediction model;

[0089] The model prediction unit 240 is used to input the latest energy consumption data of the target enterprise into the optimized energy consumption prediction model to obtain the energy consumption prediction result within a preset time period in the future.

[0090] In addition, the embodiment of the present invention also provides an enterprise energy consumption data prediction device, such as Figure 3 As shown in the figure, the enterprise energy consumption data prediction equipment specifically includes:

[0091] at least one processor; and a memory in communication with the at least one processor; wherein,

[0092] The memory stores instructions executable by at least one processor to enable the at least one processor to perform:

[0093] Based on the K-means algorithm, the pre-collected enterprise energy consumption data is pre-processed;

[0094] Extracting features from the pre-processed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least include time features and statistical features;

[0095] Constructing an energy consumption prediction model, and training the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data features;

[0096] During the training process, the optimal parameters of the model are determined based on the grid search method to obtain the optimized energy consumption prediction model;

[0097] The latest energy consumption data of the target enterprise is input into the optimized energy consumption prediction model to obtain the energy consumption prediction result within a preset time period in the future.

[0098] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0099] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0100] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0101] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0102] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0103] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0104] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0105] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0106] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0107] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0108] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0109] Each embodiment of the present invention is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0110] The above describes specific embodiments of the present invention. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting enterprise energy consumption data, characterized in that: The method comprises: Based on the K-means algorithm, the pre-collected enterprise energy consumption data is pre-processed; Extracting features from the pre-processed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least include time features and statistical features; Constructing an energy consumption prediction model, and training the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data features; During the training process, the optimal parameters of the model are determined based on the grid search method to obtain the optimized energy consumption prediction model; The latest energy consumption data of the target enterprise is input into the optimized energy consumption prediction model to obtain the energy consumption prediction result within a preset time period in the future.

2. The method for predicting enterprise energy consumption data according to claim 1, characterized in that: Based on the K-means algorithm, the pre-collected enterprise energy consumption data is pre-processed, including: Normalize the pre-collected enterprise energy consumption data to normalize the enterprise energy consumption data collected from different enterprises to the same scale; The K-means algorithm is used to perform cluster analysis on the normalized enterprise energy consumption data, and the missing values ​​are filled to obtain a complete data set.

3. The method for predicting enterprise energy consumption data according to claim 2, characterized in that: The K-means algorithm is used to perform cluster analysis on the normalized enterprise energy consumption data and fill in the missing values ​​to obtain a complete data set, including: Determine data points containing missing values ​​in the normalized enterprise energy consumption data, and fill the missing values ​​with temporary values ​​to obtain a temporary data set; Clustering the temporary data set using a K-means algorithm to divide all data points into K clusters; Similar data points are searched in the cluster where the data point with missing values ​​is located, and the missing values ​​are calculated based on the similar data points, and the temporary values ​​filled in are replaced to obtain the complete data set.

4. The method for predicting enterprise energy consumption data according to claim 1, characterized in that: The pre-processed enterprise energy consumption data is subjected to feature extraction to obtain energy consumption data features, including: Extracting the time feature from the pre-processed enterprise energy consumption data; wherein the time feature at least includes year, month, day, hour and week; The statistical features are extracted from the pre-processed enterprise energy consumption data; wherein the statistical features at least include: the mean and variance of the enterprise energy consumption data.

5. The method for predicting enterprise energy consumption data according to claim 1, characterized in that: Constructing an energy consumption prediction model, and training the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data features, specifically including: Based on the support vector machine SVM model, the energy consumption prediction model is constructed; the energy consumption prediction model uses a radial basis function as a kernel function; Aggregating the preprocessed enterprise energy consumption data and the energy consumption data features into a model training data set, and dividing the model training data set into N sub-data sets; The energy consumption prediction model is trained using the N subset data sets.

6. The method for predicting enterprise energy consumption data according to claim 5, characterized in that: The energy consumption prediction model is trained by using the N subset data sets, specifically including: Traversing the N sub-data sets, taking the first one as a test set and the remaining N-1 parts as training sets, training and testing the energy consumption prediction model to obtain a first test accuracy rate; Continue to take the second part as the test set and the remaining N-1 parts as the training set, train and test the energy consumption prediction model, and so on, until the Nth test accuracy is obtained; The obtained N test accuracy rates are averaged to obtain the algorithm accuracy value of the trained energy consumption prediction model; If the algorithm accuracy value does not reach the preset threshold, the model parameters are adjusted and the model training is performed again.

7. The method for predicting enterprise energy consumption data according to claim 6, characterized in that: During the training process, the optimal parameters of the model are determined based on the grid search method to obtain the optimized energy consumption prediction model, which includes: During the training process of the energy consumption prediction model, if the algorithm accuracy value does not reach the preset threshold, the optimal model parameters of the energy consumption prediction model are found through a grid search optimization method; wherein the optimal model parameters at least include an optimal penalty coefficient and an optimal kernel function parameter; Substitute the optimal parameters of the model into the energy consumption prediction model to obtain the optimized energy consumption prediction model.

8. The method for predicting enterprise energy consumption data according to claim 1, characterized in that: The loss function of the optimized energy consumption prediction model is: Among them, ω is the weight vector; C is the optimal penalty coefficient, which is used to balance the loss term and the regularization term; ξ i and is the slack variable, representing the error; b is the bias term.

9. An enterprise energy consumption data prediction system, characterized in that: The system comprises: A data preprocessing unit is used to preprocess the pre-collected enterprise energy consumption data based on the K-means algorithm; A feature extraction unit is used to extract features from the pre-processed enterprise energy consumption data to obtain energy consumption data features; wherein the energy consumption data features at least include time features and statistical features; A model training unit is used to construct an energy consumption prediction model and train the energy consumption prediction model through the preprocessed enterprise energy consumption data and the energy consumption data characteristics; during the training process, based on the grid search method, the optimal parameters of the model are determined to obtain an optimized energy consumption prediction model; The model prediction unit is used to input the latest energy consumption data of the target enterprise into the optimized energy consumption prediction model to obtain the energy consumption prediction result within a preset time period in the future.

10. An enterprise energy consumption data prediction device, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, so that the at least one processor can execute the enterprise energy consumption data prediction method according to any one of claims 1-8.

Images