User side energy storage identification method and device
By extracting and standardizing user load data features and using convolutional neural network model, the precise identification of user-side energy storage is achieved, solving the problem that user-side energy storage identification method is difficult to promote in the existing technology, reducing costs and improving applicability.
Patent Information
- Application Number
- CN202411691889.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-25
- Publication Date
- 2025-03-14
AI Technical Summary
In the prior art, the user-side energy storage identification method is difficult to promote and apply on a large scale, mainly due to the need for high-frequency data acquisition, which leads to higher cost of non-invasive devices.
By selecting user feature data based on user load data, performing standardized processing, it is input into the convolutional neural network model to identify the user-side energy storage.
It realizes accurate identification of user-side energy storage, reduces the requirements for data acquisition, avoids high-cost non-invasive devices, and is highly applicable and has low cost.
Smart Images
Figure CN119622468A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of energy storage technology, and in particular to a method and device for identifying energy storage at a user side. Background Art
[0002] The leapfrog development of energy storage technology, especially the rapid growth of household energy storage systems, has significantly changed the way energy is stored and used. In recent years, the installation of energy storage equipment has increased dramatically. Although it has promoted the popularization of green energy and household energy self-sufficiency, it has also brought many challenges. First, large-scale energy storage equipment has put forward new requirements for the operation and regulation of the power grid. The power grid must be able to adapt to and effectively manage the input and output of a large number of energy storage devices to ensure the stability and reliability of power supply. Secondly, with the increase in the number of energy storage devices, safety hazards are gradually emerging. For example, if energy storage equipment is not installed and maintained in accordance with standards, it may cause safety problems such as fire. Therefore, in the absence of effective supervision, energy storage equipment may bring operational and safety hazards while providing convenience. To this end, it is urgent to strengthen the control of energy storage equipment and formulate more stringent safety standards and detection mechanisms to ensure that it serves users safely and efficiently.
[0003] User-side energy storage refers to the installation of energy storage equipment on the user side. Energy storage equipment usually includes battery packs and related management equipment to store electricity when electricity demand is low and release electricity when demand peaks. User-side energy storage can help reduce electricity bills, improve power supply reliability, and support the integration of renewable energy. Through intelligent control and optimization, users can maximize energy utilization efficiency and reduce dependence on the power grid.
[0004] Various types of electrical equipment or power generation equipment have different transient characteristics when connected / exited, and energy storage equipment also has its own transient characteristics when switched on and off. The user-side energy storage identification provided by the relevant technology usually enters the transient database of the energy storage equipment when the energy storage equipment is switched on and off in a non-invasive device, and compares the user's voltage / current transient curve collected in real time with the transient characteristic library, thereby effectively identifying whether the energy storage equipment is installed on the user side. Due to the need to collect data at a high frequency, that is, the requirements for data collection are very high, the required non-invasive device is relatively expensive and difficult to promote and apply on a large scale. Summary of the invention
[0005] In order to solve the problem that the existing technology is difficult to promote and apply on a large scale, the present application provides a method for identifying user-side energy storage, which may include: selecting user feature data according to user load data. Standardizing the user feature data to obtain standardized user feature data. Inputting the standardized user feature data into a convolutional neural network model to identify user-side energy storage.
[0006] In some possible implementations, selecting user characteristic data according to user load data includes: preprocessing the user load data to obtain preprocessed user load data; and selecting user characteristic data from the preprocessed user load data.
[0007] Exemplarily, the user characteristic data includes at least one of electricity consumption during peak hours, electricity consumption during off-peak hours, daily average electricity consumption, peak hours, off-peak hours, peak duration, off-peak duration, and daily peak-valley electricity consumption difference.
[0008] In some other possible implementations, the user load data is preprocessed, including:
[0009] Interpolate user load data with missing values. Specifically, interpolate user load data with missing values. When there are many missing values, especially when there is a lack of key data at the peak, flat, and valley moments of the power grid, it is difficult to restore its true power consumption characteristics, so it needs to be discarded during data processing. When there are fewer missing values, interpolation can be performed through regression interpolation.
[0010] User load data with duplicate values or user load data with abnormal values are eliminated. Specifically, in order to eliminate duplicate values and reduce interference in the user load data mining process, thereby weakening the impact on feature extraction and classification effects, it is necessary to eliminate duplicate values. For user load data with abnormal values, the elimination process usually requires a more complex method, selecting the 3σ principle for analysis, obtaining a probability distribution diagram of the data, and then obtaining abnormal values for elimination.
[0011] In some other possible implementations, the standardized user feature data is input into a convolutional neural network model to identify the user-side energy storage, including:
[0012] The standardized user feature data is divided into a training set and a test set.
[0013] A training set indicating the category of user load data is input into the first convolutional neural network model for training to obtain the number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model.
[0014] The number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model are input into the extreme learning machine, and the artificial bee colony algorithm and the genetic algorithm are used to adjust the extreme learning machine to obtain the second convolutional neural network model.
[0015] The test set is input into the second neural network model for testing to obtain the categories of user feature data in the test set, wherein the categories of user feature data include a first category and a second category, the first category is used to indicate that the user side has installed an energy storage device, and the second category is used to indicate that the user side has not installed an energy storage device.
[0016] On the other hand, the present application provides a user-side energy storage identification device, comprising:
[0017] The selection module is used to select user characteristic data according to user load data.
[0018] The processing module is used to perform standardization processing on the user characteristic data to obtain the standardized user characteristic data.
[0019] The identification module is used to input the standardized user feature data into the convolutional neural network model to identify the user-side energy storage.
[0020] In a possible implementation, the selection module is specifically used for:
[0021] The user load data is preprocessed to obtain preprocessed user load data.
[0022] User characteristic data is selected from the preprocessed user load data.
[0023] Optionally, the user characteristic data includes at least one of electricity consumption during peak hours, electricity consumption during off-peak hours, daily average electricity consumption, peak hours, off-peak hours, peak duration, off-peak duration and daily peak-valley electricity consumption difference.
[0024] In another possible implementation, the selection module is specifically used for:
[0025] Interpolate user load data with missing values. Specifically, interpolate user load data with missing values. When there are many missing values, especially when there is a lack of key data at the peak, flat, and valley moments of the power grid, it is difficult to restore its true power consumption characteristics, so it needs to be discarded during data processing. When there are fewer missing values, interpolation can be performed through regression interpolation.
[0026] User load data with duplicate values or user load data with abnormal values are eliminated. Specifically, in order to eliminate duplicate values and reduce interference in the user load data mining process, thereby weakening the impact on feature extraction and classification effects, it is necessary to eliminate duplicate values. For user load data with abnormal values, the elimination process usually requires a more complex method, selecting the 3σ principle for analysis, obtaining a probability distribution diagram of the data, and then obtaining abnormal values for elimination.
[0027] In yet another possible implementation, the identification module is specifically used to:
[0028] The standardized user feature data is divided into a training set and a test set.
[0029] A training set indicating the category of user load data is input into the first convolutional neural network model for training to obtain the number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model.
[0030] The number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model are input into the extreme learning machine, and the artificial bee colony algorithm and the genetic algorithm are used to adjust the extreme learning machine to obtain the second convolutional neural network model.
[0031] The test set is input into the second neural network model for testing to obtain the category of the user feature data in the test set; wherein the category of the user feature data includes a first category and a second category, the first category is used to indicate that the energy storage device is installed on the user side, and the second category is used to indicate that the energy storage device is not installed on the user side.
[0032] On the other hand, the present application also provides a computer device, including: one or more processors.
[0033] A processor is used to execute one or more programs.
[0034] When one or more programs are executed by one or more processors, the above-mentioned identification method is implemented.
[0035] In another aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the above-mentioned identification method is implemented.
[0036] Compared with the prior art, the beneficial effects of this application are:
[0037] In the user-side energy storage identification method provided by the present application, user characteristic data is selected according to user load data, the user characteristic data is standardized, the standardized user characteristic data is obtained, and the standardized user characteristic data is input into the convolutional neural network model to identify the user-side energy storage. It can be seen that the present application finally realizes the identification of user-side energy storage through user load data and convolutional neural network model, has low requirements for the collection of user load data, does not need to set up non-invasive devices, and can be promoted and used on a large scale. In other words, the identification method provided by the present application is low in cost and strong in applicability.
[0038] The user-side energy storage identification method provided in this application can accurately identify whether energy storage equipment is installed on the user side, provide technical support for regional power and electricity balance regulation, and guide the safe and orderly development of user-side energy storage.
[0039] The user-side energy storage identification method provided in the present application has a wide range of applications. It can set representative user characteristic data in a region according to the specific electricity load data of users in different regions, and more accurately identify whether energy storage equipment is installed on the user side. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0041] Figure 1 A schematic flow chart of a method for identifying energy storage at a user side in an embodiment of the present application;
[0042] Figure 2 A schematic diagram of a user load curve of a first user in an embodiment of the present application;
[0043] Figure 3 Another schematic diagram of a user load curve of a first user in an embodiment of the present application;
[0044] Figure 4 A schematic diagram of a user load curve of a second user in an embodiment of the present application;
[0045] Figure 5 Another schematic diagram of a user load curve of a second user in an embodiment of the present application;
[0046] Figure 6 A schematic diagram of a user load curve of a third user in an embodiment of the present application;
[0047] Figure 7 Another schematic diagram of a user load curve of a third user in an embodiment of the present application;
[0048] Figure 8 A schematic structural diagram of a user-side energy storage identification device in an embodiment of the present application; DETAILED DESCRIPTION
[0049] The technical solution in this application will be described below in conjunction with the accompanying drawings.
[0050] The terms "first", "second", etc. in the specification embodiments, claims, and drawings of the present application are only used for the purpose of distinguishing descriptions, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, including a series of steps or units. The method, system, product, or device is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.
[0051] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.
[0052] Embodiment 1:
[0053] The present application embodiment provides a method for identifying energy storage at a user side, such as Figure 1 As shown. The following steps are included:
[0054] Step S1: Select user characteristic data according to user load data.
[0055] Step S2: performing standardization processing on the user characteristic data to obtain standardized user characteristic data.
[0056] Step S3: Input the standardized user feature data into the convolutional neural network model to identify the user-side energy storage.
[0057] In some possible implementations, selecting user characteristic data according to user load data in step S1 includes: preprocessing the user load data to obtain preprocessed user load data, and selecting user characteristic data from the preprocessed user load data.
[0058] Since there are many user load data, it is necessary to select user characteristic data. Since the electricity consumption behavior of users who have installed energy storage equipment will vary greatly, and the electricity consumption data of users in different regions are also different, users who have installed energy storage equipment consume different amounts of electricity at different times of the day, week, and year. By understanding the production and living characteristics of different categories of users, and then observing the electricity consumption curves of some users, such as the reduction in electricity consumption during the peak of the power grid, the duration of peaks and troughs, etc., there are obvious peak and valley characteristics. Therefore, appropriate user characteristic data should be selected to quantify the characteristics of electricity consumption behavior of users in different regions.
[0059] Exemplarily, the user characteristic data includes at least one of electricity consumption during peak hours, electricity consumption during off-peak hours, daily average electricity consumption, peak hours, off-peak hours, peak duration, off-peak duration, and daily peak-valley electricity consumption difference.
[0060] The power consumption during the peak hours refers to the power consumption of the user at each time point during the peak hours of the power grid. The power consumption during the valley hours refers to the power consumption of the user at each time point during the valley hours of the power grid.
[0061] For users who have installed energy storage equipment, the load curve of the day will be significantly different from their own historical data curve during the peak and valley periods of the power grid, and there will also be a certain difference in electricity consumption.
[0062] Furthermore, the user load data is preprocessed, including:
[0063] Interpolate user load data with missing values. Specifically, interpolate user load data with missing values. When there are many missing values, especially when there is a lack of key data at the peak, flat, and valley moments of the power grid, it is difficult to restore its true power consumption characteristics, so it needs to be discarded during data processing. When there are fewer missing values, interpolation can be performed through regression interpolation.
[0064] User load data with duplicate values or user load data with abnormal values are eliminated. Specifically, in order to eliminate duplicate values and reduce interference in the user load data mining process, thereby weakening the impact on feature extraction and classification effects, it is necessary to eliminate duplicate values. For user load data with abnormal values, the elimination process usually requires a more complex method, selecting the 3σ principle for analysis, obtaining a probability distribution diagram of the data, and then obtaining abnormal values for elimination.
[0065] Standardization can map user load data to a smaller feature interval through certain rules to eliminate the impact of dimensions between different user load data and make the user load data more comparable. Therefore, in step S2, the user feature data can be standardized by using the Min-Max Normalization (MMN) method to obtain the standardized user feature data.
[0066] In some other possible implementations, in step S3, the standardized user feature data is input into a convolutional neural network model to identify the user-side energy storage, including:
[0067] The standardized user feature data is divided into a training set and a test set.
[0068] The training set with the category of user load data is input into the first convolutional neural network model (CNN) for training to obtain the number of network layers, number of nodes and weights of nodes of the first convolutional neural network model. Among them, CNN is a special deep neural network model with strong feature extraction capabilities. CNN can learn and extract local and hierarchical features of input data, and effectively capture the abnormal characteristic information of power consumption behavior hidden in user load data through convolution and pooling operations.
[0069] The number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model are input into the extreme learning machine (ELM), and the artificial bee colony (ABC) algorithm (abbreviated as ABC algorithm) and the genetic algorithm (GA) are used to adjust the extreme learning machine to obtain the second convolutional neural network model.
[0070] The test set is input into the second neural network model for testing to obtain the categories of user feature data in the test set, wherein the categories of user feature data include a first category and a second category, the first category is used to indicate that the user side has installed an energy storage device, and the second category is used to indicate that the user side has not installed an energy storage device.
[0071] The effectiveness of the identification method provided in the embodiment of the present application is verified below. Select several user load data, the sampling time interval is 15 minutes, and take the user with energy storage equipment installed as an example for detailed introduction.
[0072] The user load data has obvious peak and valley phenomena. Combined with the time-of-use electricity price and industrial production activities, the peak electricity consumption period is selected as 8:00-12:00, 13:00-17:00; the valley electricity consumption period is 0:00-8:00. The user load data is preprocessed, and then the user feature data is selected. The extracted user feature data can be expressed as:
[0073] D=[t1,t2,…,t 48 ,t 52 ,t 53 ,…,t 68 ,α1,α2,β]
[0074] Where D represents the set of extracted user feature data, t m represents the user load data at time m (m = 1, 2, ..., 95, 96), t1~t 24 is the user load data at each time point during the low electricity consumption period (0:00-8:00), t 25 ~t 48 and t 52 ~t 68 It is the user load data at each time point during the peak power consumption period (8:00-12:00, 13:00-17:00). α1 represents the total load of users during the peak power consumption period (8:00-12:00), α2 represents the total load of users during the peak power consumption period (13:00-17:00), and β represents the total load of users during the off-peak period.
[0075] After the extracted user feature data is standardized, it is divided into a training set and a test set. The category to which the user load data belongs is marked in the training set, and the category is divided into 9 categories according to the local classification principle. The classification principle is set according to Table 1, and finally the obtained categories are divided into two categories according to Table 2, and finally it is determined whether the energy storage device has been installed on the user side. The training set load data is input into the first convolutional neural network model for training to obtain the number of network layers, the number of nodes, and the weights of the nodes of the first convolutional neural network model. Then the number of network layers, the number of nodes, and the weights of the nodes of the first convolutional neural network model are input into the extreme learning machine, and the artificial bee colony algorithm and the genetic algorithm are used to adjust the extreme learning machine to obtain the second convolutional neural network model.
[0076] Table 1
[0077]
[0078]
[0079] The peak and valley characteristics of the user load curve are classified according to the above 9 types. The identification results of different types of user load curves are shown in Table 2:
[0080] Table 2
[0081]
[0082] For the first user, the user load curves on different dates are as follows: Figure 2 and Figure 3 Based on the identification method provided in the embodiment of the present application, Figure 2 and Figure 3 The user load data in is identified, and the load curve type is obtained as Type 4, which determines that the first user has installed an energy storage device.
[0083] For the second user, the load curves on different dates are as follows: Figure 4 and Figure 5 As shown, based on the identification method provided in the embodiment of the present application Figure 4 and Figure 5 The user load data in is identified, and the load curve type is changed from Type 3 to Type 4, which indicates that the second user has installed energy storage equipment.
[0084] For the third user, the load curves at different times are as follows: Figure 6 and Figure 7 As shown, based on the identification method provided in the embodiment of the present application Figure 6 and Figure 7 The user load data in is identified, and the load curve type is always Type 4, which means that the third user has installed energy storage equipment.
[0085] Embodiment 2:
[0086] Based on the same inventive concept, the embodiment of the present application also provides a user-side energy storage identification device. Figure 8 As shown, the identification device 200 includes:
[0087] The selection module 21 is used to select user characteristic data according to user load data.
[0088] The processing module 22 is used to perform standardization processing on the user characteristic data to obtain the standardized user characteristic data.
[0089] The identification module 23 is used to input the standardized user feature data into the convolutional neural network model to identify the user-side energy storage.
[0090] In a possible implementation, the selection module 21 is specifically used for:
[0091] The user load data is preprocessed to obtain preprocessed user load data.
[0092] User characteristic data is selected from the preprocessed user load data.
[0093] Since there are many user load data, it is necessary to select user characteristic data. Since the electricity consumption behavior of users who have installed energy storage equipment will vary greatly, and the electricity consumption data of users in different regions are also different, users who have installed energy storage equipment consume different amounts of electricity at different times of the day, week, and year. By understanding the production and living characteristics of different categories of users, and then observing the electricity consumption curves of some users, such as the reduction in electricity consumption during the peak of the power grid, the duration of peaks and troughs, etc., there are obvious peak and valley characteristics. Therefore, appropriate user characteristic data should be selected to quantify the characteristics of electricity consumption behavior of users in different regions.
[0094] Optionally, the user characteristic data includes at least one of electricity consumption during peak hours, electricity consumption during off-peak hours, daily average electricity consumption, peak hours, off-peak hours, peak duration, off-peak duration and daily peak-valley electricity consumption difference.
[0095] The power consumption during the peak hours refers to the power consumption of the user at each time point during the peak hours of the power grid. The power consumption during the valley hours refers to the power consumption of the user at each time point during the valley hours of the power grid.
[0096] For users who have installed energy storage equipment, the load curve of the day will be significantly different from their own historical data curve during the peak and valley periods of the power grid, and there will also be a certain difference in electricity consumption.
[0097] In another possible implementation, the selection module 21 is specifically used for:
[0098] Interpolate user load data with missing values. Specifically, interpolate user load data with missing values. When there are many missing values, especially when there is a lack of key data at the peak, flat, and valley moments of the power grid, it is difficult to restore its true power consumption characteristics, so it needs to be discarded during data processing. When there are fewer missing values, interpolation can be performed through regression interpolation.
[0099] User load data with duplicate values or user load data with abnormal values are eliminated. Specifically, in order to eliminate duplicate values and reduce interference in the user load data mining process, thereby weakening the impact on feature extraction and classification effects, it is necessary to eliminate duplicate values. For user load data with abnormal values, the elimination process usually requires a more complex method, selecting the 3σ principle for analysis, obtaining a probability distribution diagram of the data, and then obtaining abnormal values for elimination.
[0100] Standardization can map user load data to a smaller feature interval through certain rules to eliminate the impact of dimensions between different user load data and make the user load data more comparable. Therefore, the processing module 22 can use the Min-Max Normalization (MMN) method to standardize the user feature data to obtain the standardized user feature data.
[0101] In another possible implementation, the identification module 23 is specifically used for:
[0102] The standardized user feature data is divided into a training set and a test set.
[0103] The training set with the category of user load data is input into the first convolutional neural network model (CNN) for training to obtain the number of network layers, number of nodes and weights of nodes of the first convolutional neural network model. Among them, CNN is a special deep neural network model with strong feature extraction capabilities. CNN can learn and extract local and hierarchical features of input data, and effectively capture the abnormal characteristic information of power consumption behavior hidden in user load data through convolution and pooling operations.
[0104] The number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model are input into the extreme learning machine (ELM), and the artificial bee colony (ABC) algorithm (abbreviated as ABC algorithm) and the genetic algorithm (GA) are used to adjust the extreme learning machine to obtain the second convolutional neural network model.
[0105] The test set is input into the second neural network model for testing to obtain the categories of user feature data in the test set, wherein the categories of user feature data include a first category and a second category, the first category is used to indicate that the user side has installed an energy storage device, and the second category is used to indicate that the user side has not installed an energy storage device.
[0106] Embodiment 3:
[0107] Based on the same inventive concept, the embodiment of the present application also provides a computer device, which includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding functions, so as to implement the steps of the identification method provided in the above embodiment.
[0108] Embodiment 4:
[0109] Based on the same inventive concept, the embodiment of the present application also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of the identification method provided in the above embodiment.
[0110] Those skilled in the art will appreciate that the embodiments of the application may be provided as methods, systems, or computer program products. Therefore, the application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0111] The application is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 generate 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.
[0112] 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 including 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.
[0113] 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 in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0114] The above are merely embodiments of the application and are not intended to limit the application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the application are included in the scope of the claims of the pending application.
Claims
1. A method for identifying energy storage at a user side, characterized in that: include: Selecting user characteristic data according to user load data; Performing standardization processing on the user characteristic data to obtain standardized user characteristic data; The standardized user feature data is input into a convolutional neural network model to identify user-side energy storage.
2. The identification method according to claim 1, characterized in that: The selecting of user characteristic data according to user load data includes: Preprocessing the user load data to obtain preprocessed user load data; User characteristic data is selected from the preprocessed user load data.
3. The identification method according to claim 2, characterized in that: The preprocessing of the user load data includes: Interpolation of user load data with missing values; User load data with duplicate values or user load data with abnormal values are eliminated.
4. The identification method according to claim 1, characterized in that: The user characteristic data includes at least one of the electricity consumption during peak hours, electricity consumption during off-peak hours, daily average electricity consumption, peak hours, off-peak hours, peak duration, off-peak duration and daily peak-to-valley electricity consumption difference.
5. The identification method according to claim 1, characterized in that: The step of inputting the standardized user feature data into a convolutional neural network model to identify user-side energy storage includes: Dividing the standardized user feature data into a training set and a test set; Inputting a training set indicating the category of user load data into a first convolutional neural network model for training, and obtaining the number of network layers, the number of nodes, and the weights of the nodes of the first convolutional neural network model; Inputting the number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model into an extreme learning machine, and adjusting the extreme learning machine using an artificial bee colony algorithm and a genetic algorithm to obtain a second convolutional neural network model; The test set is input into the second neural network model for testing to obtain the category of the user feature data in the test set; wherein the category of the user feature data includes a first category and a second category, the first category is used to indicate that the energy storage device is installed on the user side, and the second category is used to indicate that the energy storage device is not installed on the user side.
6. A user-side energy storage identification device, characterized in that: include: A selection module, used for selecting user characteristic data according to user load data; A processing module, used for performing standardization processing on the user characteristic data to obtain standardized user characteristic data; The identification module is used to input the standardized user feature data into a convolutional neural network model to identify the user-side energy storage.
7. The identification device according to claim 6, characterized in that: The selection module is specifically used for: Preprocessing the user load data to obtain preprocessed user load data; User characteristic data is selected from the preprocessed user load data.
8. The identification device according to claim 7, characterized in that: The selection module is specifically used for: Interpolation of user load data with missing values; User load data with duplicate values or user load data with abnormal values are eliminated.
9. The identification device according to claim 6, characterized in that: The user characteristic data includes at least one of the electricity consumption during peak hours, electricity consumption during off-peak hours, daily average electricity consumption, peak hours, off-peak hours, peak duration, off-peak duration and daily peak-to-valley electricity consumption difference.
10. The identification device according to claim 6, characterized in that: The identification module is specifically used for: Dividing the standardized user feature data into a training set and a test set; Inputting a training set indicating the category of user load data into a first convolutional neural network model for training, and obtaining the number of network layers, the number of nodes, and the weights of the nodes of the first convolutional neural network model; Inputting the number of network layers, the number of nodes and the weights of the nodes of the first convolutional neural network model into an extreme learning machine, and adjusting the extreme learning machine using an artificial bee colony algorithm and a genetic algorithm to obtain a second convolutional neural network model; The test set is input into the second neural network model for testing to obtain the category of the user feature data in the test set; wherein the category of the user feature data includes a first category and a second category, the first category is used to indicate that the energy storage device is installed on the user side, and the second category is used to indicate that the energy storage device is not installed on the user side.
11. A computer device, characterized in that: include: one or more processors; The processor is used to store one or more programs; When the one or more programs are executed by the one or more processors, the identification method according to any one of claims 1 to 5 is implemented.
12. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the identification method according to any one of claims 1 to 5 is implemented.