Method, system and device for predicting electricity consumption of rural emerging business and medium

Abstract

The invention discloses a rural emerging business electricity consumption prediction method, system and device and a medium, and relates to the field of electricity market electricity consumption demand prediction, and the method comprises the steps: obtaining the electricity consumption associated data of a target region in a prediction time period; the target area is a rural area; performing dimension reduction on the power utilization associated data by adopting a principal component analysis method, and extracting key influence factors to obtain multi-modal factor data of the target area; performing feature fusion on the multi-modal factor data of the target area to obtain multi-modal fusion feature data; on the basis of an electricity consumption prediction model, according to the multi-modal fusion feature data, determining electricity consumption data of the target area in a prediction time period; the electricity consumption prediction model is obtained by training a long-short-term memory network by adopting a training sample set in advance; the training sample set comprises historical electricity consumption data and historical multi-modal fusion feature data of the target area. According to the invention, the accuracy of predicting the electricity consumption of the rural emerging business is improved.

Description

Technical Field

[0001] The present application relates to the field of electricity market demand forecasting, and in particular to a method, system, device and medium for forecasting electricity consumption of emerging rural businesses taking into account multi-dimensional characteristics. Background Art

[0002] With the vigorous development of new rural businesses, agricultural electricity consumption continues to rise, and accurate prediction of electricity consumption is becoming increasingly important in the planning and operation of power systems. For rural areas, by analyzing and predicting the electricity consumption of new rural businesses, it is helpful for rural areas to quickly discover whether there are abnormalities in electricity consumption and provide solutions in a timely manner. For relevant power supply companies, they can flexibly adjust power supply decisions based on the predicted trend of electricity consumption, thereby improving the efficiency and reliability of power supply services, promoting the awareness of energy conservation and emission reduction, and actively building a social environment for saving electricity.

[0003] The current electricity consumption forecasting methods used in agriculture usually rely on statistical time series models, regression analysis or modeling based on physical principles. These modeling methods are usually based on single data of linear relationships. However, according to the differences in new rural business forms and models, electricity consumption forecasts are also affected by many factors. Traditional electricity consumption forecasting methods are limited by the modeling capabilities of nonlinear relationships, cannot handle complex dynamic environmental change information, and lack adaptability to large-scale multi-dimensional data, making electricity consumption forecasts inaccurate. Summary of the invention

[0004] The purpose of this application is to provide a method, system, device and medium for predicting electricity consumption of emerging rural businesses, which can improve the accuracy of electricity consumption prediction for emerging rural businesses.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for predicting electricity consumption of emerging rural businesses, including:

[0007] Obtaining electricity consumption related data of a target area within a forecast period; the target area is a rural area;

[0008] The power consumption related data are reduced in dimension by using a principal component analysis method to extract key influencing factors, thereby obtaining multimodal factor data of the target area; the power consumption related data include data of multiple modes; the multimodal factor data of the target area include key influencing factors of each mode;

[0009] Performing feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data;

[0010] Based on the electricity consumption prediction model, the electricity consumption data of the target area within the prediction period is determined according to the multimodal fusion feature data; the electricity consumption prediction model is obtained by pre-training a long short-term memory network using a training sample set; the training sample set includes historical electricity consumption data and historical multimodal fusion feature data of the target area.

[0011] Optionally, a principal component analysis method is used to reduce the dimension of the electricity consumption related data, extract key influencing factors, and obtain multimodal factor data of the target area, specifically including:

[0012] Standardize the data of each modality separately to obtain standardized data of each modality;

[0013] Principal component analysis was performed on the standardized data of each modality to extract the key influencing factors of each modality to obtain the multimodal factor data of the target area.

[0014] Optionally, the data of the multiple modalities are: economic factor data, technical factor data, policy factor data, environmental factor data and social factor data.

[0015] Optionally, the economic factor data include gross domestic product and energy prices; the technical factor data include the number of electrical appliances and the number of enterprises; the policy factor data include electricity price policy; the environmental factor data include monthly temperature; and the social factor data include the number of rural population.

[0016] Optionally, principal component analysis is performed on the standardized data of each modality to extract the key influencing factors of each modality, including:

[0017] For the standardized data of any modality, calculating the covariance matrix of the standardized data;

[0018] Calculating the eigenvalues ​​and eigenvectors of the covariance matrix;

[0019] The eigenvectors with eigenvalues ​​greater than the set threshold are taken as principal components;

[0020] The standardized data is projected onto the principal components to obtain key influencing factors of the modality.

[0021] Optionally, performing feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data specifically includes:

[0022] The multimodal factor data of the target area are sequentially cleaned, converted and normalized to obtain preprocessed multimodal data;

[0023] Performing feature extraction on the preprocessed multimodal data to obtain multimodal feature data;

[0024] Aligning the multimodal feature data in time and space to obtain aligned multimodal feature data;

[0025] The aligned multimodal feature data are fused to obtain multimodal fused feature data.

[0026] Optionally, the method for predicting electricity consumption of emerging rural services further includes:

[0027] Obtain historical electricity consumption data and historical electricity consumption related data of the target area;

[0028] The historical electricity consumption related data are reduced in dimension by using a principal component analysis method to extract key influencing factors, thereby obtaining historical multimodal factor data of the target area; the historical electricity consumption related data include historical data of multiple modes; the historical multimodal factor data of the target area include key influencing factors of each mode;

[0029] Performing feature fusion on the historical multimodal factor data of the target area to obtain historical multimodal fusion feature data;

[0030] According to the historical electricity consumption data and the historical multimodal fusion feature data, the mean square error is used as the cost function to train the long short-term memory network to obtain the electricity consumption prediction model.

[0031] In the second aspect, the present application provides a rural emerging business electricity consumption prediction system, including:

[0032] A data acquisition module, used to acquire electricity consumption related data of a target area within a forecast period; the target area is a rural area;

[0033] A data dimension reduction module, connected to the data acquisition module, for reducing the dimension of the electricity consumption related data by using a principal component analysis method, extracting key influencing factors, and obtaining multimodal factor data of the target area; the electricity consumption related data includes data of multiple modes; the multimodal factor data of the target area includes key influencing factors of each mode;

[0034] A feature fusion module, connected to the data dimension reduction module, for performing feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data;

[0035] A prediction module is connected to the feature fusion module and is used to determine the power consumption data of the target area within the prediction period based on the power consumption prediction model and the multimodal fusion feature data; the power consumption prediction model is obtained by pre-training the long short-term memory network using a training sample set; the training sample set includes historical power consumption data and historical multimodal fusion feature data of the target area.

[0036] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method for predicting electricity consumption of emerging rural businesses.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for predicting electricity consumption of emerging rural businesses.

[0038] According to the specific embodiments provided in this application, this application has the following technical effects:

[0039] The present application provides a method, system, device and medium for predicting electricity consumption for emerging rural businesses. The key influencing factors of electricity consumption-related data are extracted through the principal component analysis method to obtain multimodal factor data, and the multimodal factor data is feature fused. The long short-term memory network is used to predict electricity consumption, which improves the accuracy of electricity consumption prediction and provides effective technical support for the power grid agency electricity purchasing business. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 This is an application environment diagram of a method for predicting electricity consumption of emerging rural businesses in one embodiment of the present application;

[0042] Figure 2 A flowchart of a method for predicting electricity consumption of emerging rural businesses provided in one embodiment of the present application;

[0043] Figure 3 This is a schematic diagram of a process for predicting electricity consumption of emerging rural businesses in one embodiment of the present application;

[0044] Figure 4 This is a unit structure diagram of the long short-term memory network;

[0045] Figure 5 A schematic diagram of functional modules of a rural emerging business electricity consumption prediction system provided in one embodiment of the present application;

[0046] Figure 6 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0048] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0049] The method for predicting electricity consumption of emerging rural businesses provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the power consumption related data of the target area within the prediction period to the server 104, and the server 104 determines the power consumption data of the target area within the prediction period based on the power consumption related data of the target area within the prediction period. The server 104 can feedback the power consumption data of the target area within the prediction period to the terminal 102. In addition, in some embodiments, the power consumption prediction method for emerging rural businesses can also be implemented separately by the server 104 or the terminal 102.

[0050] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0051] In an exemplary embodiment, Figure 2 and Figure 3 As shown, a method for predicting electricity consumption of emerging rural businesses is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used for explanation, and the steps include the following steps 201 to 204.

[0052] Step 201, obtaining electricity consumption related data of a target area within a forecast period. The target area is a rural area.

[0053] In an exemplary embodiment, the electricity consumption related data includes data in multiple modes, which are: economic factor data, technical factor data, policy factor data, environmental factor data, and social factor data.

[0054] The economic factor data include gross domestic product (agricultural output value) and energy prices (monthly electricity price data). The technical factor data include the number of electrical appliances and the number of enterprises. The policy factor data include electricity price policy. The environmental factor data include monthly temperature. The social factor data include the number of rural population.

[0055] Step 202: Use principal component analysis (PCA) to reduce the dimension of the electricity consumption related data, extract key influencing factors, and obtain multimodal factor data of the target area.

[0056] In an exemplary embodiment, the multimodal factor data of the target area includes key influencing factors of each modality, namely economic modality data, social modality data, policy modality data, environmental modality data and technical modality data. Step 202 includes the following steps 301 to 302.

[0057] Step 301, standardize the data of each modality to obtain standardized data of each modality. The standardized data includes standardized economic characteristic data, standardized social characteristic data, standardized policy characteristic data, standardized environmental characteristic data and standardized technical characteristic data.

[0058] Since PCA is sensitive to the scale and dimension of the data, the data needs to be standardized so that the mean of each feature is 0 and the variance is 1, thereby ensuring that all features have the same importance in the analysis. The standardized formula is:

[0059]

[0060] Among them, z mn is the standardized value, x mn is the value of the mth sample on the nth feature, μ n is the mean of the nth feature, σ n is the standard deviation of the nth feature.

[0061] Step 302 , principal component analysis is performed on the standardized data of each modality to extract the key influencing factors of each modality to obtain multimodal factor data of the target area.

[0062] Specifically, for the standardized data of any modality, the covariance matrix of the standardized data is first calculated. The covariance matrix is ​​used to describe the correlation between various features in the data, and is obtained by calculating the covariance between the various features of the data. The formula is:

[0063]

[0064] Where ∑ is the covariance matrix, x m is the vector of the mth sample, μ is the mean vector of all samples, and M is the number of samples.

[0065] Then the eigenvalues ​​and eigenvectors of the covariance matrix are calculated. The eigenvalues ​​and eigenvectors of the covariance matrix are calculated using linear algebra methods, where the eigenvalues ​​represent the variance of the data in different directions, and the eigenvectors represent the unit vectors of the data in that direction.

[0066] Then, the eigenvectors with eigenvalues ​​greater than the set threshold are used as principal components. Specifically, the first k eigenvectors with larger eigenvalues ​​are selected as principal components, and these principal components will be used to form a new feature space.

[0067] Finally, the standardized data is projected onto the principal component to obtain the key influencing factors of the modality, thereby achieving data dimensionality reduction while retaining the original data information as much as possible.

[0068] Step 203: perform feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data.

[0069] In an exemplary embodiment, step 203 includes the following steps 401 to 404 .

[0070] Step 401, the multimodal factor data of the target area is sequentially cleaned, converted and normalized to obtain pre-processed multimodal data. By cleaning, converting and normalizing the multimodal factor data, noise, outliers and inconsistencies in the data are eliminated, providing guarantee for subsequent feature extraction and fusion processing.

[0071] Step 402: extract features from the preprocessed multimodal data to obtain multimodal feature data. Specifically, meaningful information and features are extracted from the preprocessed multimodal data for subsequent fusion and processing. Feature extraction needs to take into account the characteristics of the data and application scenarios, and select appropriate feature extraction methods and algorithms.

[0072] Step 403: align the multimodal feature data in time and space to obtain aligned multimodal feature data. Specifically, align the data from different sources in time and space to ensure consistency of the fused data.

[0073] Step 404, the aligned multimodal feature data is fused to obtain multimodal fused feature data. Considering the characteristics of different data sources and the fusion target, a reasonable fusion model is designed and optimized. Common fusion models include weighted average, neural network, support vector machine, etc.

[0074] Step 204: Based on the power consumption prediction model and the multimodal fusion feature data, determine the power consumption data of the target area within the prediction period. The power consumption prediction model is obtained by pre-training a long short-term memory network (LSTM) using a training sample set. The training sample set includes historical power consumption data and historical multimodal fusion feature data of the target area.

[0075] The electricity consumption prediction model can perform time series prediction and accurately predict the electricity consumption of the next month. Specifically, the multimodal fusion feature data of the next month is input into the electricity consumption prediction model to obtain the electricity consumption data of the next month. The electricity consumption data output by the electricity consumption prediction model is normalized data, and the normalized electricity consumption data is further denormalized to obtain the actual electricity consumption data.

[0076] In terms of constructing a power consumption prediction model, the method for predicting power consumption of emerging rural businesses also includes the following steps (1) to (4).

[0077] (1) Obtain historical electricity consumption data and historical electricity consumption related data of the target area.

[0078] (2) Using the principal component analysis method to reduce the dimension of the historical electricity consumption related data, extract key influencing factors, and obtain historical multimodal factor data of the target area. The historical electricity consumption related data includes historical data of multiple modes. The historical multimodal factor data of the target area includes key influencing factors of each mode.

[0079] (3) Performing feature fusion on the historical multimodal factor data of the target area to obtain historical multimodal fusion feature data.

[0080] (4) Based on the historical electricity consumption data and the historical multimodal fusion feature data, a long short-term memory network is trained using a mean square error as a cost function to obtain an electricity consumption prediction model.

[0081] Specifically, the historical electricity consumption data and the historical multimodal fusion feature data are divided into a training set and a test set to train and test the long short-term memory network, obtain the optimal hyperparameters of the long short-term memory network, and obtain the electricity consumption prediction model.

[0082] LSTM includes input layer, LSTM layer, fully connected layer and output layer. The input layer receives standardized time series data, usually represented in the form of a three-dimensional tensor (number of samples, number of time steps, number of features); multiple LSTM units in the LSTM layer can be stacked to form a deep LSTM network to capture more complex time series patterns. Each LSTM unit includes an input gate, a forget gate, an output gate and a unit state; the output of the LSTM layer is passed to the fully connected layer for further feature extraction and prediction; the output layer generates the final prediction result.

[0083] LSTM introduces a logic unit cell state cell to judge the validity of input data. Three gate structures control the update of data in the cell state to achieve long-term flow of data on the network. The input gate controls the data input into the memory cell, the forget gate controls the deletion of data and resets the memory cell, and the output gate controls the data output.

[0084] Input gate: Based on the input data, the input gate conditionally determines which data can be used to update the memory state. The specific formula is: i = sigmoid (W i [h t-1 , X t ]+b i ).

[0085] Forget gate: According to the input data, the forget gate conditionally determines the information required for the cell state based on the activation function. The specific formula is: f = sigmoid (W f [h t-1 , X t ]+b f ).

[0086] Output gate: According to the input data and the information stored in the current memory unit, the output information is conditionally determined. The specific formula is: o = sigmoid (W o [h t-1 , X t ]+b o ).

[0087] Among them, i, f, and o are the output data of the input gate, forget gate, and output gate respectively. f determines the forgetting ratio of the cell state, i determines the ratio of the current data input to the cell state, and o determines the ratio of the cell state output to the current state h. t is the input data at time T, h t-1 is the state at time T-1, sigmoid is the activation function, W i , W f , W o are the weight vectors of the input gate, forget gate, and output gate, respectively, and b i 、b f 、bo are the phase vectors of the input gate, forget gate, and output gate, respectively, and are parameters to be trained.

[0088] like Figure 4 As shown, the input data is composed of the input data X at time T t and the state h at time t-1 t-1 Calculated, where the tanh function limits the function value to (-1, 1) to create a new candidate state vector, the memory information C at time T t The memory information C at time T-1 after passing through the forget gate t-1 And the input information through the input gate determines: C t =f*C t-1 +i*tanh(W c [h t-1 , X t ]+b c ), where W c is the weight vector of the memory cell, b c It is the memory cell phase.

[0089] The state h at time t t The output gate o determines how to update: h t =o*tanh(C t ).

[0090] Use the training set to train LSTM, use the test set to evaluate the performance of LSTM, adjust the hyperparameters to optimize LSTM, and if LSTM reaches the accuracy, save and use it. If not, train repeatedly until the accuracy is reached.

[0091] This application combines multimodal feature data fusion and LSTM to identify the relevant factors that affect the monthly electricity consumption of different emerging agricultural businesses, and models the change functions of various factors, so as to accurately predict the future electricity consumption of agriculture, and thus provide effective technical support for the power grid agency power purchasing business.

[0092] Based on the same inventive concept, the embodiment of the present application also provides a rural emerging business electricity consumption prediction system for implementing the rural emerging business electricity consumption prediction method involved above. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more rural emerging business electricity consumption prediction system embodiments provided below can refer to the limitations of the rural emerging business electricity consumption prediction method above, and will not be repeated here.

[0093] In an exemplary embodiment, Figure 5As shown, a rural emerging business electricity consumption prediction system is provided, including: a data acquisition module 501, a data dimension reduction module 502, a feature fusion module 503 and a prediction module 504.

[0094] The data acquisition module 501 is used to acquire electricity consumption related data of a target area within a forecast period; the target area is a rural area.

[0095] The data dimension reduction module 502 is connected to the data acquisition module 501, and is used to reduce the dimension of the electricity consumption related data by using the principal component analysis method, extract key influencing factors, and obtain multimodal factor data of the target area.

[0096] The feature fusion module 503 is connected to the data dimension reduction module 502, and the feature fusion module 503 is used to perform feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data. The historical electricity consumption related data includes historical data of multiple modes. The historical multimodal factor data of the target area includes key influencing factors of each mode.

[0097] The prediction module 504 is connected to the feature fusion module 503, and the prediction module 504 is used to determine the power consumption data of the target area within the prediction period based on the power consumption prediction model and the multimodal fusion feature data. The power consumption prediction model is obtained by pre-training the long short-term memory network using a training sample set. The training sample set includes historical power consumption data and historical multimodal fusion feature data of the target area.

[0098] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store a power consumption prediction model. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting power consumption of emerging rural businesses is implemented.

[0099] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0100] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0101] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0102] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0103] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0104] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0105] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0106] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0107] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for predicting electricity consumption of emerging rural businesses, characterized in that: include: Obtaining electricity consumption related data of a target area within a forecast period; the target area is a rural area; The power consumption related data are reduced in dimension by using a principal component analysis method to extract key influencing factors, thereby obtaining multimodal factor data of the target area; the power consumption related data include data of multiple modes; the multimodal factor data of the target area include key influencing factors of each mode; Performing feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data; Based on the electricity consumption prediction model, the electricity consumption data of the target area within the prediction period is determined according to the multimodal fusion feature data; the electricity consumption prediction model is obtained by pre-training a long short-term memory network using a training sample set; the training sample set includes historical electricity consumption data and historical multimodal fusion feature data of the target area.

2. The method for predicting electricity consumption of emerging rural businesses according to claim 1 is characterized in that: The principal component analysis method is used to reduce the dimension of the electricity consumption related data, extract key influencing factors, and obtain multimodal factor data of the target area, including: Standardize the data of each modality separately to obtain standardized data of each modality; Principal component analysis was performed on the standardized data of each modality to extract the key influencing factors of each modality to obtain the multimodal factor data of the target area.

3. The method for predicting electricity consumption of emerging rural businesses according to claim 2 is characterized in that: The data of multiple modes are: economic factor data, technical factor data, policy factor data, environmental factor data and social factor data.

4. The method for predicting electricity consumption of emerging rural businesses according to claim 3 is characterized in that: The economic factor data include gross domestic product and energy prices; the technical factor data include the number of electrical appliances and the number of enterprises; the policy factor data include electricity price policy; the environmental factor data include monthly temperature; and the social factor data include the number of rural population.

5. The method for predicting electricity consumption of emerging rural businesses according to claim 2, characterized in that: The principal component analysis was performed on the standardized data of each modality to extract the key influencing factors of each modality, including: For the standardized data of any modality, calculating the covariance matrix of the standardized data; Calculating the eigenvalues ​​and eigenvectors of the covariance matrix; The eigenvectors with eigenvalues ​​greater than the set threshold are taken as principal components; The standardized data is projected onto the principal components to obtain key influencing factors of the modality.

6. The method for predicting electricity consumption of emerging rural businesses according to claim 1, characterized in that: Performing feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data specifically includes: The multimodal factor data of the target area are sequentially cleaned, converted and normalized to obtain preprocessed multimodal data; Performing feature extraction on the preprocessed multimodal data to obtain multimodal feature data; Aligning the multimodal feature data in time and space to obtain aligned multimodal feature data; The aligned multimodal feature data are fused to obtain multimodal fused feature data.

7. The method for predicting electricity consumption of emerging rural businesses according to claim 1, characterized in that: The method for predicting electricity consumption of emerging rural businesses also includes: Obtain historical electricity consumption data and historical electricity consumption related data of the target area; The historical electricity consumption related data are reduced in dimension by using a principal component analysis method to extract key influencing factors, thereby obtaining historical multimodal factor data of the target area; the historical electricity consumption related data include historical data of multiple modes; the historical multimodal factor data of the target area include key influencing factors of each mode; Performing feature fusion on the historical multimodal factor data of the target area to obtain historical multimodal fusion feature data; According to the historical electricity consumption data and the historical multimodal fusion feature data, the mean square error is used as the cost function to train the long short-term memory network to obtain the electricity consumption prediction model.

8. A rural emerging business electricity consumption prediction system, applied to the rural emerging business electricity consumption prediction method according to any one of claims 1 to 7, characterized in that: The rural emerging business electricity consumption prediction system comprises: A data acquisition module, used to acquire electricity consumption related data of a target area within a forecast period; the target area is a rural area; A data dimension reduction module, connected to the data acquisition module, for reducing the dimension of the electricity consumption related data by using a principal component analysis method, extracting key influencing factors, and obtaining multimodal factor data of the target area; the electricity consumption related data includes data of multiple modes; the multimodal factor data of the target area includes key influencing factors of each mode; A feature fusion module, connected to the data dimension reduction module, for performing feature fusion on the multimodal factor data of the target area to obtain multimodal fusion feature data; A prediction module is connected to the feature fusion module and is used to determine the power consumption data of the target area within the prediction period based on the power consumption prediction model and the multimodal fusion feature data; the power consumption prediction model is obtained by pre-training the long short-term memory network using a training sample set; the training sample set includes historical power consumption data and historical multimodal fusion feature data of the target area.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for predicting electricity consumption of emerging rural businesses according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting electricity consumption of emerging rural businesses described in any one of claims 1 to 7 is implemented.

Images