Industrial user power load decomposition method and device and electronic equipment

Abstract

The application provides an industrial user power load decomposition method and device and electronic equipment. After determining the power management configuration of an industrial user, the industrial user is classified according to the power management configuration, and multi-element completion of the user load is performed according to the classification result to obtain original power consumption data. After preprocessing the original power consumption data, load characteristic analysis is performed according to the power consumption load data to obtain a load curve for each industrial user. The power consumption behavior habit of the industrial user can be seen through the load curve. Finally, based on the load curve, power load decomposition is performed through low-frequency data processing to determine the working state of each power equipment. The low-frequency data processing extracts the effective working period of each equipment load to generate one-dimensional single load power, thereby realizing load decomposition of some power consumption equipment with similar structures and working principles.

Description

Technical Field

[0001] This application relates to the field of load splitting technology, and in particular to a method, apparatus and electronic equipment for splitting power load for industrial users. Background Technology

[0002] In the research on load decomposition technology, various energy consumption scenarios were constructed for different load types. The main idea is to improve the ability to extract equipment state characteristics and build a highly economical load decomposition model. Addressing the regulation needs of new power systems, load decomposition of industrial users can provide a clearer picture of user electricity consumption behavior from the perspective of power suppliers, facilitating overall planning and improving load management capabilities. This is of great significance for power suppliers to conduct demand-side management and understand user electricity consumption behavior. Therefore, from the perspective of power suppliers, it is necessary to solve the problem that some electrical devices with similar structures and operating principles are often difficult to distinguish. Summary of the Invention

[0003] In view of this, the purpose of this application is to propose a method, apparatus and electronic equipment for decomposing the power load of industrial users, in order to solve the problem that it is usually difficult to distinguish some electrical equipment with similar structure and working principle.

[0004] To achieve the above objectives, the first aspect of this application provides a method for decomposing the power load of industrial users, comprising:

[0005] Determine the power management configuration of industrial users and classify the industrial users according to the power management configuration;

[0006] Based on the classification results, multi-dimensional supplementation of user load is performed to obtain the raw electricity consumption data;

[0007] The raw electricity consumption data is preprocessed to obtain electricity load data;

[0008] Based on the electricity load data, load characteristic analysis is performed to obtain a load curve for each industrial user.

[0009] Based on the load curve, the power load is decomposed through low-frequency data processing to determine the operating status of each power device.

[0010] Optionally, classifying the industrial users according to the power management configuration includes:

[0011] In response to the fact that the power management configuration of any of the industrial users includes a marketing system or an electricity data acquisition system, the industrial user is identified as a first type of industrial user;

[0012] In response to the fact that the power management configuration of any of the industrial users does not include the marketing system or the electricity data acquisition system, the industrial user is identified as a second type of industrial user.

[0013] Optionally, the step of performing multivariate completion of user load based on the classification results to obtain raw electricity consumption data includes:

[0014] In response to the industrial user being the first type of industrial user, the raw electricity consumption data is obtained from the marketing system or the electricity consumption data acquisition system;

[0015] In response to the fact that the industrial user is the second type of industrial user, a data multi-element completion command is issued;

[0016] In response to receiving the industrial user feedback data, the original electricity consumption data is determined based on the feedback data;

[0017] The feedback data refers to the set of information fed back by the industrial user after receiving the data completion instruction.

[0018] Optionally, the step of performing load characteristic analysis based on the electricity load data to obtain a load curve for each industrial user includes:

[0019] The electricity load data is sampled according to a preset time interval to obtain sample data;

[0020] Based on the sample data, a load curve for each industrial user is constructed according to the principle of equal area.

[0021] Optionally, the step of decomposing the power load based on the load curve using low-frequency data processing to determine the operating status of each power device includes:

[0022] Modal decomposition is performed based on the load curve to obtain low-frequency subsequence, mid-frequency subsequence and high-frequency subsequence;

[0023] Clustering is performed based on the low-frequency subsequences to determine the types of power loads;

[0024] The operating status of the power equipment is determined based on the single-machine load capacity of the individual equipment corresponding to the type of power load.

[0025] Optionally, the preprocessing of the original electricity consumption data to obtain electricity load data includes:

[0026] The original electricity consumption data is cleaned to obtain cleaned data;

[0027] The cleaning data is standardized to obtain standard data;

[0028] The principal components of the standard data were extracted using principal component analysis to obtain the electricity load data.

[0029] Optionally, determining the operating status of the power equipment based on the single-unit load capacity of the individual equipment corresponding to the power load type includes:

[0030] The total power at any sampling moment is determined based on the load curve.

[0031] Determine the ratio of the total power to the single-unit load capacity;

[0032] The ratio is rounded up to obtain the first number of operations, and the first operating power is calculated based on the first number of operations.

[0033] The ratio is rounded down to obtain the second operating quantity, and the second operating power is calculated based on the second operating quantity.

[0034] The actual number of operations is determined based on the first operating power, the second operating power, and the total power.

[0035] Optionally, determining the actual number of operations based on the first operating power, the second operating power, and the total power includes:

[0036] Calculate the first difference between the first operating power and the total power;

[0037] Calculate the second difference between the second operating power and the total power;

[0038] In response to the first difference being less than or equal to the second difference, the first running quantity is determined as the actual running quantity;

[0039] In response to the first difference being greater than the second difference, the second running quantity is determined as the actual running quantity.

[0040] A second aspect of this application provides an industrial user power load splitting device, characterized in that it comprises:

[0041] The classification module is configured to: determine the power management configuration of industrial users and classify the industrial users according to the power management configuration;

[0042] The data acquisition module is configured to: perform multi-dimensional completion of user load based on the classification results to obtain raw electricity consumption data;

[0043] The preprocessing module is configured to preprocess the raw electricity consumption data to obtain electricity load data;

[0044] The load analysis module is configured to perform load characteristic analysis based on the electricity load data to obtain a load curve for each industrial user.

[0045] The load decomposition module is configured to: decompose the power load based on the load curve through low-frequency data processing to determine the operating status of each power device.

[0046] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method provided in the first aspect of this application.

[0047] As can be seen from the above, the industrial user power load decomposition method, apparatus, and electronic equipment provided in this application, after determining the power management configuration of industrial users, classify industrial users according to the power management configuration, and perform multi-dimensional load completion based on the classification results to obtain original power consumption data. By adopting different multi-dimensional load completion methods for different categories of industrial users, more comprehensive original power consumption data is obtained, improving the accuracy of power load decomposition. The original power consumption data is preprocessed to obtain power load data. The preprocessing process can improve the robustness of the data and reduce data redundancy, which is conducive to more efficient load decomposition in the future. Then, load characteristic analysis is performed based on the power load data to obtain the load curve for each industrial user. The load curve can reveal the power consumption behavior of industrial users. Finally, based on the load curve, power load decomposition is performed through low-frequency data processing to determine the working status of each power device. Low-frequency data processing extracts the effective working time of each device load and generates a one-dimensional single load power, thereby realizing the load decomposition of some power devices with similar structures and working principles. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart of the industrial user power load decomposition method according to an embodiment of this application;

[0050] Figure 2 This is a bar chart showing the industry categories and operating capacity of embodiments of this application;

[0051] Figure 3 This is a schematic diagram illustrating data multi-factor completion in an embodiment of this application;

[0052] Figure 4 A schematic diagram of the load curve for an industrial user manufacturing electric vehicles according to an embodiment of this application;

[0053] Figure 5 A flowchart illustrating user classification in the embodiments of this application;

[0054] Figure 6 This is a flowchart illustrating how the original electricity consumption data was obtained in an embodiment of this application.

[0055] Figure 7 This is a flowchart illustrating how the load curve is obtained in an embodiment of this application;

[0056] Figure 8a This is a flowchart illustrating the load decomposition process in an embodiment of this application.

[0057] Figure 8b This is a schematic diagram illustrating the low-frequency clustering results of industrial users manufacturing electric vehicles, as described in an embodiment of this application.

[0058] Figure 8c A schematic diagram of low-frequency clustering results for industrial users performing electro-hydrogen production according to embodiments of this application;

[0059] Figure 8d This is a schematic diagram of the low-frequency clustering results of industrial users in the data center according to an embodiment of this application;

[0060] Figure 9 A flowchart for determining the operating state of power equipment in an embodiment of this application;

[0061] Figure 10 This is a schematic diagram of the structure of the industrial user power load splitting device according to an embodiment of this application;

[0062] Figure 11 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0064] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0065] As shown in the background section, power grid intelligence is an inevitable trend and choice for power grid development. Demand-side management plays a crucial role in smart grids and is one of the important manifestations of power grid intelligence. With the recent surge in artificial intelligence, the concept of smart grids has been proposed and has developed rapidly. In the future, more and more diverse electricity loads will be connected to the grid, which places higher demands on the intelligent development of the power system, and power management will face greater and more severe challenges.

[0066] Load splitting is a core issue in power management, and its implementation is a win-win situation for both power suppliers and consumers. For power suppliers, load splitting technology allows them to better understand the detailed electrical parameters of different types of load equipment, grasp the specific load composition and usage of industrial users, formulate more rational dispatch strategies, and conduct more detailed and comprehensive analysis of the monitored data. This facilitates more rational planning of the power system by power suppliers, ensuring the entire power grid operates efficiently, safely, and economically. For industrial users, from an economic perspective, those who respond to load splitting can receive subsidies and incentives from power suppliers. Simultaneously, influenced by the deepening concept of energy conservation and emission reduction, electricity consumers are more enthusiastic about understanding their own electricity consumption and related electrical quantity information. Industrial users can make corresponding changes and measures to reduce energy consumption based on the load splitting information fed back by power suppliers, and rationally adjust the load operation status. Two typical methods for energy conservation by industrial users are increasing electricity efficiency and reducing unnecessary electricity consumption. For example, reducing unnecessary energy waste, replacing old equipment with high-efficiency and energy-saving equipment, and changing the usage time of high-power equipment to off-peak hours can further improve the efficiency of electricity utilization, thereby achieving energy conservation and rational electricity use.

[0067] Load splitting is broadly categorized into two types: invasive load splitting and non-invasive load splitting. Invasive load splitting involves installing additional energy metering sensor chips or modules on each load to be metered. These devices can accurately record detailed operating status and specific electrical parameters of the load. However, the additional sensor metering equipment has drawbacks such as relatively high cost, complex installation and operation, and difficult daily maintenance, significantly limiting its widespread industrial application. Non-invasive load identification and splitting, on the other hand, does not require additional sensor metering equipment. It only needs to acquire electrical quantity information such as total voltage, current, power, and power factor, and then process and analyze this data to use as features for load splitting. Compared to invasive load identification methods, this method offers simpler and more convenient communication, relatively lower hardware costs, and allows for real-time online monitoring. It is easier to implement and apply in actual industrial production activities, making it a key research and development direction for load splitting.

[0068] Load splitting is essentially a decomposition of energy, breaking down total electricity consumption into the electricity consumption of individual loads. Implementing load splitting requires not only total electricity consumption data but also the metering of individual sub-load consumption data. Therefore, from the perspective of electricity suppliers, to achieve more rational and precise planning and management of the power system, it is essential to address the problem of difficulty in distinguishing between electrical devices with similar structures and operating principles, and this also places high demands on the completeness and comprehensiveness of the data source.

[0069] The industrial user power load decomposition method provided in this application, after determining the power management configuration of industrial users, classifies industrial users according to the power management configuration, and obtains more comprehensive and complete original power consumption data by adopting different user load multivariate completion methods for different categories of industrial users, thereby improving the accuracy of power load decomposition. The original power consumption data is preprocessed to obtain power load data. The preprocessing process can improve the robustness of the data and reduce data redundancy, which is conducive to more efficient load decomposition in the future. Then, load characteristic analysis is performed based on the power load data to obtain a load curve for each industrial user. The load curve reveals the power consumption behavior habits of industrial users. Finally, based on the load curve, power load decomposition is performed through low-frequency data processing to determine the working status of each power device. Low-frequency data processing can extract the effective working time of each device load and generate a one-dimensional single load power, thereby realizing the load decomposition of some power devices with similar structures and working principles. The following embodiments will illustrate this in detail.

[0070] In some embodiments, such as Figure 1 As shown, the method for decomposing the power load of industrial users includes:

[0071] Step 100: Determine the power management configuration of industrial users and classify them according to the power management configuration.

[0072] In some optional embodiments, before determining the power management configuration for industrial users, the industry type and operating capacity of the industrial users can be determined first. Different industry categories of industrial users have different operating capacities. For industrial users belonging to industry categories with large operating capacities, their equipment is diverse and their power load is numerous and varied, making them the primary targets for load decomposition. For industrial users belonging to industry categories with small operating capacities, their equipment may be numerous, but because their total power consumption is small, from the perspective of the power supplier, in order to better plan the entire power system more rationally, load decomposition for this type of user can be temporarily suspended when resources are scarce. That is, the management configuration of industrial users under industry categories with larger operating capacities can be prioritized. For example, as shown... Figure 2 As shown, when the industrial user's industry category is chemical raw materials and chemical products manufacturing, the operating capacity of industrial users in this category during normal operation is 160 million kVA. This category of industrial users requires special attention and has a high priority in power system planning. When the industrial user's industry category is tobacco products manufacturing, the operating capacity of industrial users in this category during normal operation is very low, and their impact on power system planning is relatively small. Therefore, their planning priority can be reduced after load decomposition.

[0073] In some alternative embodiments, power suppliers generally have sufficient resources to distribute the load among various industrial users. However, industrial user power load data is enormous, with different types of loads applied in different scenarios at different times, generating tens of billions of data points. Industrial power data mainly originates from power values, voltage values, current values, etc., at various nodes. With technological advancements, more and more power data acquisition and marketing systems are emerging in new power systems, capable of collecting, storing, and managing the massive amounts of power data under their jurisdiction. Therefore, before collecting and acquiring industrial user power load data, it is necessary to conduct initial screening of industrial users' information to determine their power management configuration and whether they possess a complete power data acquisition or marketing system. Industrial users with complete electricity data acquisition or marketing systems can directly extract the collected and stored data as known raw electricity data. Therefore, industrial users whose management configuration includes a marketing system or electricity data acquisition system are identified as the first type of industrial users. Industrial users without complete energy management data or marketing systems need to undergo targeted complementary and multi-dimensional surveys to supplement the user load of raw electricity data. Therefore, industrial users whose management configuration does not include a marketing system or electricity data acquisition system are identified as the second type of industrial users.

[0074] Step 200: Perform multi-dimensional completion of user load based on the classification results to obtain the original electricity consumption data.

[0075] In some alternative embodiments, since industrial power loads involve a wide range of fields and cover many enterprises, it is not possible to simply obtain the electricity consumption data of all flexible loads. It is necessary to conduct a more extensive multi-dimensional load census to supplement the multi-dimensional data of user loads. Considering that most industrial users do not have the conditions to directly collect data from external sources, it is proposed to further distinguish the internal branch loads and collection conditions of industrial users and to formulate detailed census plans for different categories of industrial users so that industrial users can provide more comprehensive and accurate feedback data.

[0076] In some alternative embodiments, detailed census plans are developed for different categories of users, such as... Figure 3 As shown, based on whether an industrial user data acquisition system or marketing system is included and whether there is a clear wiring relationship, a targeted industrial user survey method is designed:

[0077] For industrial users who have electricity data acquisition systems or marketing systems, they belong to the first category of industrial users. They can collect data from the marketing system or their own electricity data acquisition system to obtain the raw electricity data of the industry.

[0078] For industrial users without electricity data acquisition or marketing systems, falling under the second category, a comprehensive survey needs to be conducted using data multi-source completion instructions. This includes: first, determining whether the industrial user has a dedicated meter; for those with branch meters, clarifying whether there is a clear wiring relationship; if so, data can be directly obtained from the nameplates of each branch device; if there is no clear wiring relationship, the load decomposition technology proposed in this project needs to be used for analysis. For industrial loads without branch meters, if the wiring relationship is relatively clear, data can be read from the nameplates of all equipment; if there is no clear wiring relationship, the industrial user needs to proactively provide feedback data such as industry production operation characteristics (e.g., production planning and adjustment capabilities, key electrical equipment), and daily load data. This yields a wide range of raw electricity data from industrial users, including equipment type, single-machine load capacity, load operation status in the production process, and production energy consumption habits.

[0079] Step 300: Preprocess the raw electricity consumption data to obtain the electricity load data.

[0080] In some optional embodiments, the raw electricity consumption data is preprocessed to obtain electricity load data, including:

[0081] The raw electricity consumption data is cleaned to obtain cleaned data.

[0082] The cleaned data is standardized to obtain standard data;

[0083] Principal component analysis was used to extract the principal components from the standard data to obtain the electricity load data.

[0084] The process begins with cleaning the raw electricity consumption data. This involves observing the characteristics of the raw data, checking for duplicate rows, outliers, and missing values, deleting duplicates and outliers, and using simple mean imputation to fill in missing values, resulting in cleaned data. Next, the cleaned data is standardized. For example, z-score standardization can be used to obtain further processed standard data, which improves the robustness of the data. Finally, considering data requirements, principal component analysis can be used to extract the principal components of the standard data, obtaining data that correlates with the processed electricity load data, thereby reducing data redundancy and uncovering the inherent relationships within the data.

[0085] Step 400: Perform load characteristic analysis based on the electricity load data to obtain the load curve for each industrial user.

[0086] In some alternative embodiments, load characteristic analysis is a crucial aspect of smart grid development. It helps to understand users' electricity consumption habits, facilitating research such as electricity consumption pattern analysis, social attribute identification, and non-intrusive load monitoring, thus providing conditions for load decomposition. For example, taking an industrial user manufacturing electric vehicles as an example, electricity load data is sampled at 15-minute intervals. Using 24-hour electricity load data as an example, data from 96 sampling points within a day is obtained. Long-term measurements are then performed to obtain data such as... Figure 4 The figure shows the load curve of an industrial user. It can be observed that this industrial user's load has two prominent peaks, mainly concentrated during midnight and midday. During these two periods, the electricity consumption of industrial users manufacturing electric vehicles is relatively concentrated, reflecting a high demand for electricity during these times. The two peaks at midnight and midday indicate concentrated charging during these periods, which is related to the industrial user's production habits. During midday, other users are generally resting, allowing the electric vehicle manufacturing industrial user to effectively manage concentrated electricity consumption during this idle period, improving power efficiency. Midnight is a low-demand period for other users. Concentrating the load of electric vehicle manufacturing industrial users during this period can alleviate the large gap between peak and off-peak electricity consumption, effectively regulating the peak-valley difference in electricity load and playing a role in maintaining the supply and demand balance of the power system. During off-peak hours, when other users are working, the power consumption of electric vehicle manufacturing industrial users is relatively low. In contrast, the load of industrial users engaged in electro-hydrogen production has a prominent peak (not shown in the figure), mainly concentrated around midday working hours, indicating a high and relatively concentrated demand for electricity during this period. The variance of the load data of industrial users in the data center is very small, and the load power fluctuates around a specific value in each time period (not shown in the figure), indicating that the power consumption behavior of industrial users in the data center is weakly correlated with time, reflecting the strong stability of power consumption of data center users.

[0087] Step 500: Based on the load curve, decompose the power load through low-frequency data processing to determine the operating status of each power device.

[0088] In some optional embodiments, load decomposition based on the load curve first requires performing Empirical Mode Decomposition (EMD) on the load curve to obtain modal feature components of different frequencies. This includes using low-frequency subsequences as the main feature components of the load curve, mid-frequency subsequences as detailed feature components, and high-frequency subsequences as random feature components. Then, clustering is performed based on the low-frequency subsequences to obtain the types of industrial user power loads. Each type of power load generates a different one-dimensional individual load power, thus achieving load decomposition for electrical equipment with similar structures and operating principles. Finally, the single-unit load capacity of equipment in a non-zero stable operating state (non-zero means the equipment is in a working state; zero means the equipment is off and does not generate a response load; stable means the equipment is in a stable operating state with minimal load fluctuations) is used as the independent variable. This minimizes the deviation between the actual power value on the load curve and the power value obtained from the load decomposition, thus obtaining the operating state of each electrical device and achieving the purpose of load decomposition.

[0089] In summary, the industrial user power load decomposition method provided in this application, after determining the power management configuration of industrial users, classifies them according to the power management configuration, and performs multivariate load completion based on the classification results to obtain raw power consumption data. By adopting different multivariate load completion methods for different categories of industrial users, more comprehensive raw power consumption data is obtained, improving the accuracy of power load decomposition. Preprocessing the raw power consumption data to obtain power load data improves the robustness of the data and reduces data redundancy, which is beneficial for more efficient subsequent load decomposition. Then, load characteristic analysis is performed based on the power load data to obtain… For each industrial user, a load curve is generated, which reveals their electricity consumption habits. Based on this load curve, low-frequency data processing is used to decompose the power load, determining the operating status of each electrical device. The low-frequency data processing extracts the effective operating time of each device's load, generating a one-dimensional single load power. This allows for load decomposition of electrical devices with similar structures and operating principles. Finally, using the single-machine load capacity of each device in its non-zero stable operating state from the power load data as the independent variable, the deviation between the actual power value on the load curve and the power value obtained from the load decomposition is minimized, thus obtaining the operating status of each electrical device and achieving the purpose of load decomposition.

[0090] In some embodiments, such as Figure 5 As shown, industrial users are classified according to their power management configuration, including:

[0091] Step 110: In response to any industrial user's power management configuration including a marketing system or an electricity data acquisition system, the user is identified as a Class I industrial user.

[0092] In some alternative embodiments, with technological advancements, new power systems are seeing an increasing number of electricity data acquisition and marketing systems. In these systems, additional energy metering sensor chips or modules are installed on each load that needs to be metered, acting as monitoring devices. These devices can very accurately record the detailed operating status of the load and specific electrical quantity parameters, enabling industrial users to collect, store, and manage massive amounts of electricity data. Therefore, before collecting and acquiring industrial user power load data, it is necessary to conduct initial screening of industrial users to determine their power management configuration. This configuration is used to determine whether they possess a complete electricity data acquisition or marketing system. Industrial users with a complete electricity data acquisition or marketing system can directly extract the collected and stored data as known raw electricity data. Therefore, industrial users whose management configuration includes a marketing system or electricity data acquisition system are classified as Category I industrial users. It should be noted that any industrial user whose electricity load data can be directly obtained can be classified as Category I industrial users; therefore, no restrictions are placed on the specific monitoring devices included in the industrial user's power management configuration.

[0093] Step 120: In response to the fact that the power management configuration of any industrial user does not include a marketing system or a power consumption data acquisition system, the user is identified as a second-class industrial user.

[0094] In some alternative embodiments, for industrial users who do not have complete energy management data or marketing systems and cannot directly obtain known raw electricity consumption data, it is necessary to conduct a targeted complementary multi-dimensional survey to supplement the user load of the raw electricity consumption data. Therefore, industrial users whose management configuration does not include a marketing system or electricity consumption data collection system are identified as the second type of industrial users.

[0095] In some embodiments, such as Figure 6 As shown, based on the classification results, multivariate completion of user load is performed to obtain the original electricity consumption data, including:

[0096] Step 210: In response to the industrial user being classified as a first-class industrial user, obtain raw electricity consumption data from the marketing system or electricity data acquisition system.

[0097] In some alternative embodiments, after an industrial user is identified as a first-class industrial user, since its power management configuration has a complete power consumption data acquisition system or marketing system, the data information stored by the industrial user can be directly extracted as known raw power consumption data.

[0098] Step 220: In response to the industrial user being a second-class industrial user, issue a data multi-factor completion command.

[0099] In some optional embodiments, for example, after determining that an industrial user is a second-class industrial user, since its power management configuration lacks a complete electricity data acquisition system or marketing system, a data multi-factor completion instruction is issued to the industrial user. Based on this instruction, the system first checks whether a branch meter is available. If a branch meter is available and has a clear connection relationship, the nameplate information of each device within that connection relationship is determined, and this nameplate information is summarized to obtain feedback data. If a branch meter is available but has no clear connection relationship, only the information recorded by the meter is summarized, and then the branch meter data is supplemented through methods such as issuing questionnaires to obtain feedback data. If a branch meter is not available but has a clear connection relationship, a questionnaire is issued based on that connection relationship to supplement the connection relationship data, thus obtaining feedback data. If neither a branch meter nor a clear connection relationship is available, a questionnaire is directly issued to investigate the required data, thus obtaining feedback data. The feedback data is the set of information provided by the industrial user after receiving the data multi-factor completion instruction. It should be noted that for the second type of industrial users, as long as they can provide data feedback after receiving the data multi-source completion instruction, there are no restrictions on the method of data acquisition.

[0100] Step 230: In response to receiving feedback data from industrial users, determine the original power consumption data based on the feedback data.

[0101] In some alternative embodiments, more attention is paid to the industry production operation characteristics of industrial loads (such as production planning and adjustment capabilities, key electrical equipment, etc.) and daily load data during the load decomposition process. Therefore, after receiving feedback data from industrial users, the original electricity consumption data is determined based on the equipment type, the stable operating power value of a single device (single machine load capacity), the loads included in the production process (electricity load types), production energy consumption habits, and other data in the feedback data.

[0102] In some embodiments, such as Figure 7 As shown, load characteristic analysis is performed based on electricity load data to obtain load curves for each industrial user, including:

[0103] Step 410: Sample the power load data according to the preset time interval to obtain sample data.

[0104] In some alternative embodiments, for example, starting at midnight of the day, the electricity load data is sampled at 15-minute intervals. Taking the electricity load data within a 24-hour period of a day as an example, sample data from 96 sampling points within a day is obtained.

[0105] Step 420: Based on the sample data, construct the load curve for each industrial user according to the principle of equal area.

[0106] In some alternative embodiments, since the sampling interval of the sample data is once every 15 minutes, it is difficult to perceive the load state changes that occur within the sampling interval based solely on the discrete data obtained from the sampling points. It is necessary to restore the discrete data points into a continuous curve. Using the principle of equal area, and based on the method that the power consumption value between two sampling points is equal to the power consumption value of the restored continuous curve, data points within the sampling interval are supplemented to restore the characteristics of the data and the trend of the curve, thereby obtaining the load curve for each industrial user.

[0107] In some embodiments, such as Figure 8a As shown, based on the load curve, power load is decomposed through low-frequency data processing to determine the operating status of each power device, including:

[0108] Step 510: Perform modal decomposition based on the load curve to obtain low-frequency subsequence, mid-frequency subsequence, and high-frequency subsequence.

[0109] In some alternative embodiments, empirical mode decomposition (EMD) can be used to perform mode decomposition on the load curve. The decomposition process is as follows: find all the maxima of the original data sequence (load curve) X(t) and fit them with a cubic spline interpolation function to form the upper envelope of the original data; similarly, find all the minima and fit them with a cubic spline interpolation function to form the lower envelope of the data. The mean of the upper and lower envelopes is denoted as ml. Subtract this mean envelope ml from the original data sequence X(t) to obtain a new data sequence hl.

[0110] X(t)-ml=hl

[0111] If the new data obtained by subtracting the envelope average from the original data still contains negative local maxima and positive local minima, it indicates that this is not an intrinsic mode function and further "filtering" is needed until low-frequency subsequences, mid-frequency subsequences, and high-frequency subsequences are obtained.

[0112] Step 520: Cluster the low-frequency subsequences to determine the types of power loads.

[0113] In some alternative embodiments, cluster analysis groups data objects based on information describing the objects and their relationships found in the data. For example... Figure 8bThe low-frequency clustering results shown are for industrial users engaged in electric vehicle manufacturing, such as Figure 8c The results shown are low-frequency clustering results for industrial users engaged in electro-hydrogen production, and such as Figure 8d Taking the low-frequency clustering results of industrial users in the data center as an example, it can be seen that... Figure 8b The three low-frequency subsequence curves in the data form a group. Figure 8c The three low-frequency subsequence curves in the data form a group. Figure 8d Five low-frequency subsequence curves are grouped together. Objects within a group are similar (related), while objects in different groups are different (unrelated). Higher intra-group similarity and greater inter-group discrepancy indicate better clustering. The goal of clustering is to achieve high intra-cluster similarity and low inter-cluster similarity, maximizing the distance between clusters and minimizing the distance between samples within a cluster and the cluster center. Therefore, low-frequency subsequences are used for clustering to group loads of electrical equipment with similar structures and operating principles, preparing for subsequent detailed differentiation. Low-frequency subsequences are chosen because the total power fluctuation of industrial users' equipment is relatively small during stable operation, so the time periods corresponding to low-frequency subsequences better reflect the actual power load of industrial users during operation.

[0114] Step 530: Determine the operating status of the power equipment based on the single-unit load capacity of the individual equipment corresponding to the type of power load.

[0115] In some optional embodiments, for multiple load types in the clustered low-frequency subsequences, each low-frequency subsequence undergoes a working state determination with all power load types (i.e., using the single-machine load capacity of each device in the non-zero stable working state in the power load data as the independent variable, calculating the deviation between the actual power value and the power value obtained from load decomposition for each power load type on the load curve). The power load type with the smallest deviation is determined as the power load type corresponding to that low-frequency subsequence. When all low-frequency subsequences have found corresponding load types, the actual number of devices corresponding to each power load type can also be determined during the working state determination process, thus achieving the purpose of load decomposition. The process of determining the working state of any low-frequency subsequence with any power load type is as follows: The working state of the power equipment is determined based on the single-machine load capacity of the individual device corresponding to the power load type, such as... Figure 9 As shown, it includes:

[0116] Step 531: Determine the total power corresponding to any sampling time based on the load curve.

[0117] In this step, the total power of industrial users at a certain sampling time is determined on the load curve. To be more representative, sampling times in more representative time periods can be selected first. Taking industrial users who manufacture electric vehicles as an example, the total power corresponding to the sampling time at midnight or noon is more representative.

[0118] Step 532: Determine the ratio of total power to single-unit load capacity.

[0119] In this step, the ratio of total power to the single-unit load capacity of each device is calculated. This ratio can generally be equated to the number of devices that are operating stably.

[0120] Step 533: Round up the ratio to obtain the first number of operations, and calculate the first operating power based on the first number of operations.

[0121] In this step, due to interference from other factors, the calculated ratio is generally not an integer. At this time, the ratio is rounded up to obtain the first number of runs, and the first running power is calculated based on the first number of runs, which is used to compare with the total power to determine the magnitude of the error.

[0122] Step 534: Round down the ratio to obtain the second number of operations, and calculate the second operating power based on the second number of operations.

[0123] In this step, due to interference from other factors, the calculated ratio is generally not an integer. At this time, the ratio is rounded down to obtain the second running quantity, and the second running power is calculated based on the second running quantity to compare with the total power and determine the magnitude of the error.

[0124] Step 535: Determine the actual number of operations based on the first operating power, the second operating power, and the total power.

[0125] In some optional embodiments, determining the actual number of operations based on the first operating power, the second operating power, and the total power includes:

[0126] Calculate the first difference between the first operating power and the total power;

[0127] Calculate the second difference between the second operating power and the total power;

[0128] In response to the first difference being less than or equal to the second difference, the first running quantity is determined as the actual running quantity;

[0129] In response to the first difference being greater than the second difference, the second running quantity is determined as the actual running quantity.

[0130] The first operating power is calculated as the total power of a first number of devices operating stably at the same time. Since the first number of devices is obtained by rounding up, the calculated first operating power will be greater than the total power. The first difference between the first operating power and the total power represents the power error caused by rounding up. The second operating power is calculated as the total power of a second number of devices operating stably at the same time. Since the second number of devices is obtained by rounding up or down, the calculated first operating power will be less than the total power. The second difference between the second operating power and the total power represents the power error caused by rounding down. When the first difference is less than or equal to the second difference, it means that the first number of devices is closer to the actual number of devices operating, and the first number of devices is determined as the actual number of devices operating. When the first difference is greater than the second difference, it means that the second number of devices operating is closer to the actual number of devices operating, and the second number of devices is determined as the actual number of devices operating.

[0131] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0132] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0133] Based on the same inventive concept, and corresponding to any of the above embodiments, this application also provides an industrial user power load decomposition device.

[0134] refer to Figure 10 The industrial user power load splitting device includes

[0135] Classification module 10 is configured to: determine the power management configuration of industrial users and classify industrial users according to the power management configuration;

[0136] The data acquisition module 20 is configured to: perform multi-dimensional completion of user load based on the classification results to obtain raw electricity consumption data;

[0137] The preprocessing module 30 is configured to preprocess the raw electricity consumption data to obtain electricity load data;

[0138] The load analysis module 40 is configured to perform load characteristic analysis based on electricity load data to obtain a load curve for each industrial user.

[0139] The load decomposition module 50 is configured to: decompose the power load based on the load curve and through low-frequency data processing to determine the operating status of each power device.

[0140] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0141] The apparatus described above is used to implement the corresponding industrial user power load decomposition method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0142] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the industrial user power load decomposition method described in any of the above embodiments.

[0143] Figure 11 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0144] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0145] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0146] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0147] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0148] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0149] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0150] The electronic devices described above are used to implement the corresponding industrial user power load decomposition method in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0151] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a computer-readable storage medium storing computer instructions for causing the computer to execute the industrial user power load decomposition method as described in any of the above embodiments.

[0152] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. 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 technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0153] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the industrial user power load decomposition method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0154] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0155] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0156] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0157] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A method for decomposing the power load of industrial users, characterized in that, include: Determine the power management configuration of industrial users and classify the industrial users according to the power management configuration; Based on the classification results, multi-dimensional supplementation of user load is performed to obtain the raw electricity consumption data; The raw electricity consumption data is preprocessed to obtain electricity load data; Based on the electricity load data, load characteristic analysis is performed to obtain a load curve for each industrial user. Based on the load curve, the power load is decomposed through low-frequency data processing to determine the operating status of each power device. Based on the load curve, power load is decomposed through low-frequency data processing to determine the operating status of each power device, including: The total power at any sampling moment is determined based on the load curve. Determine the ratio of the total power to the single-unit load capacity; wherein the single-unit load capacity is the load capacity of a single device corresponding to the low-frequency subsequence load type of the load curve; The ratio is rounded up to obtain the first number of operations, and the first operating power is calculated based on the first number of operations. The ratio is rounded down to obtain the second operating quantity, and the second operating power is calculated based on the second operating quantity. The actual number of operations is determined based on the first operating power, the second operating power, and the total power.

2. The method according to claim 1, characterized in that, The classification of industrial users according to the power management configuration includes: In response to the fact that the power management configuration of any of the industrial users includes a marketing system or an electricity data acquisition system, the industrial user is identified as a first type of industrial user; In response to the fact that the power management configuration of any of the industrial users does not include the marketing system or the electricity data acquisition system, the industrial user is identified as a second type of industrial user.

3. The method according to claim 2, characterized in that, The process of performing multi-dimensional completion of user load based on the classification results to obtain raw electricity consumption data includes: In response to the industrial user being the first type of industrial user, the raw electricity consumption data is obtained from the marketing system or the electricity consumption data acquisition system; In response to the fact that the industrial user is the second type of industrial user, a data multi-element completion command is issued; In response to receiving the industrial user feedback data, the original electricity consumption data is determined based on the feedback data; The feedback data refers to the set of information fed back by the industrial user after receiving the data completion instruction.

4. The method according to claim 1, characterized in that, The step of performing load characteristic analysis based on the electricity load data to obtain a load curve for each industrial user includes: The electricity load data is sampled according to a preset time interval to obtain sample data; Based on the sample data, a load curve for each industrial user is constructed according to the principle of equal area.

5. The method according to claim 1, characterized in that, The process of decomposing the power load based on the load curve using low-frequency data processing to determine the operating status of each power device includes: Modal decomposition is performed based on the load curve to obtain low-frequency subsequence, mid-frequency subsequence and high-frequency subsequence; Clustering is performed based on the low-frequency subsequences to determine the types of power loads; The operating status of the power equipment is determined based on the single-machine load capacity of the individual equipment corresponding to the type of power load.

6. The method according to claim 1, characterized in that, The preprocessing of the raw electricity consumption data to obtain electricity load data includes: The original electricity consumption data is cleaned to obtain cleaned data; The cleaning data is standardized to obtain standard data; The principal components of the standard data were extracted using principal component analysis to obtain the electricity load data.

7. The method according to claim 1, characterized in that, The step of determining the actual number of operations based on the first operating power, the second operating power, and the total power includes: Calculate the first difference between the first operating power and the total power; Calculate the second difference between the second operating power and the total power; In response to the first difference being less than or equal to the second difference, the first running quantity is determined as the actual running quantity; In response to the first difference being greater than the second difference, the second running quantity is determined as the actual running quantity.

8. An industrial user power load splitting device, characterized in that, include: The classification module is configured to: determine the power management configuration of industrial users and classify the industrial users according to the power management configuration; The data acquisition module is configured to: perform multi-dimensional completion of user load based on the classification results to obtain raw electricity consumption data; The preprocessing module is configured to preprocess the raw electricity consumption data to obtain electricity load data; The load analysis module is configured to perform load characteristic analysis based on the electricity load data to obtain a load curve for each industrial user. The load decomposition module is configured to: decompose the power load based on the load curve through low-frequency data processing to determine the operating status of each power device; Based on the load curve, power load is decomposed through low-frequency data processing to determine the operating status of each power device, including: The total power at any sampling moment is determined based on the load curve. Determine the ratio of the total power to the single-unit load capacity; wherein the single-unit load capacity is the load capacity of a single device corresponding to the low-frequency subsequence load type of the load curve; The ratio is rounded up to obtain the first number of operations, and the first operating power is calculated based on the first number of operations. The ratio is rounded down to obtain the second operating quantity, and the second operating power is calculated based on the second operating quantity. The actual number of operations is determined based on the first operating power, the second operating power, and the total power.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 7.

Images