Big data measurement asset use portrait and demand prediction allocation method

Abstract

The invention discloses a big data measurement asset use portrait and demand prediction allocation method, and belongs to the technical field of electric power big data and asset optimization management, and the method comprises the steps: S10, constructing a database related to a target region, the database comprising a high-demand power data group and a low-demand power data group, acquiring power consumption data of the user from the high-demand power consumption data group; and S20, distinguishing the high-demand power consumption data group into a regular power consumption user group and an irregular power consumption user group, and calculating the proportion of the regular power consumption user group and the irregular power consumption user group in the high-demand power consumption data group. According to the method, the accuracy and the response speed of power demand prediction are improved and the dynamic change of a complex power system is effectively dealt with by constructing the database, distinguishing the power consumption behavior modes of the users, dynamically monitoring the power demand change, extracting key power consumption characteristics, quantifying the load fluctuation ratio and implementing hierarchical load management.

Description

Technical Field

[0001] This invention relates to the field of power big data and asset optimization management technology, and in particular to a big data-based method for profiling asset usage and forecasting and allocating demand. Background Technology

[0002] Metering assets are the collective term for equipment used for electricity metering and management in a power system. In the power industry, effective management and accurate forecasting of power resources are crucial for ensuring the safe and stable operation of the power grid, as well as improving operational efficiency and reducing costs.

[0003] In this area of ​​research, application CN202411449645.3 provides a deep learning-based method and system for predicting power material demand. This technical solution uses sample power equipment trajectories and sample material trajectories as input data. By extracting trajectory vectors and introducing them into a first prediction unit and a second prediction unit, sample power material knowledge vectors and sample power plant knowledge vectors are generated, respectively. Furthermore, by calculating the feature distance between the two as the target training cost parameter, the basic prediction neural network undergoes fine-tuning and optimization, thereby generating a more accurate target prediction neural network, providing timely and accurate power material demand forecasts for the power system.

[0004] Another application, CN202510135682.5, provides a method and system for predicting power material demand based on deep learning. This technical solution includes using a convolutional neural network to extract features from power engineering drawings to obtain unstructured feature representations of power materials; based on these unstructured feature representations and annotation information from the power engineering drawings, using natural language processing technology to construct a multi-dimensional feature vector of power engineering material demand; this technical solution achieves efficient feature extraction and multi-dimensional feature fusion of power engineering drawings through deep learning technology, thereby improving the accuracy and efficiency of power material demand prediction and enhancing the level of intelligent material management.

[0005] However, the above-mentioned technical solutions still have shortcomings. When predicting electricity demand based on the usage profile of a certain area, it is difficult to depict the dynamic usage characteristics of electricity in that area, resulting in a deviation rate of more than 15% between the profile and the actual state of electricity demand. They are unable to cope with sudden load changes and have poor adaptability to complex dynamic changes in the power system. Summary of the Invention

[0006] In view of the problems existing in the field of power big data and asset optimization management technology, the present invention is proposed.

[0007] Therefore, one of the objectives of this invention is to provide a big data-based method for profiling asset usage and forecasting and allocating demand. By constructing a database, differentiating user electricity consumption patterns, dynamically monitoring changes in electricity demand, extracting key electricity consumption characteristics, quantifying load volatility, and implementing tiered load management, this method significantly improves the accuracy and response speed of electricity demand forecasting, enhances the safe and stable operation of the power grid, optimizes the allocation of power resources, effectively addresses the dynamic changes in complex power systems, and provides strong support for the precise management and efficient operation of the power industry.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] A big data-driven method for profiling asset usage and forecasting demand allocation includes:

[0010] S10: Construct a database about the target area, the database including a high-demand electricity consumption data group and a low-demand electricity consumption data group, and obtain the user's electricity consumption data from the high-demand electricity consumption data group;

[0011] S20: The high-demand electricity consumption data group is distinguished, including distinguishing it into regular electricity consumption user group and irregular electricity consumption user group, and the proportion of the regular electricity consumption user group and the irregular electricity consumption user group in the high-demand electricity consumption data group is calculated;

[0012] S30: Based on the calculated proportion, if the proportion of the regular electricity-consuming user group is large, then the electricity consumption data collected from the regular electricity-consuming user group will be divided into categories based on each natural day as the collection period. , , ,..., ; Indicates the first Electricity consumption data for each user;

[0013] S40: Perform feature extraction on the electricity consumption data of users, including extracting the change features of each user after the same amount of electricity consumption, and marking the same amount of electricity consumption as reference electricity consumption;

[0014] S50: Obtain the time node corresponding to the reference electricity consumption, and according to the time node, count the number of users in the regular electricity consumption user group who have reached the reference electricity consumption. If the proportion of the number is less than 50% in the regular electricity consumption user group, it is determined that the electricity demand of the target area is low; otherwise, it is not determined.

[0015] S60: When it is determined that the power demand in the target area is low, the power consumption change pattern of the irregular power consumption user group is obtained. If the power consumption changes towards the reference power consumption, it is determined that the power demand in the target area is increasing; otherwise, no determination is made.

[0016] In a preferred embodiment of the present invention, if it is determined that the power demand is increasing, the power grid load corresponding to the target area is calculated; if the power grid load is stable, the power supply is not increased; if the power grid load is on an upward trend, the power supply is increased.

[0017] In a preferred embodiment of the present invention, in S40, the reference electricity consumption includes the maximum electricity consumption of each user in the regular electricity consumption user group, and the average value calculated based on the maximum electricity consumption of each user;

[0018] Based on the reference electricity consumption, if the number of users reaching the reference electricity consumption exceeds 50% of the total number of users in the regular electricity consumption group, a monitoring mechanism is preset. Based on the monitoring mechanism, the electricity demand of the target area is predicted. The monitoring mechanism is as follows:

[0019] A preset monitoring period is given within the monitoring period. Time point, Time point and The time points are defined as a monitoring cycle of at least 10 calendar days, with each given adjacent time point spaced 10 to 20 minutes apart.

[0020] Statistics in the above The number of users whose electricity consumption reaches the reference level at a given time point;

[0021] In the aforementioned An initial time point is preset before the current time point, and the preset initial time point is at least equal to the current time point. The time points are spaced 10 to 15 minutes apart. Within the initial time point, the maximum power consumption of each user before reaching the reference power consumption is obtained according to the number of users, and the maximum power consumption is marked as the critical power consumption value.

[0022] In the aforementioned For users whose electricity consumption reaches the reference level at a given time, if the user's electricity consumption is... If the time point exceeds the critical power value, it is determined that the power demand in the target area is increasing.

[0023] In the aforementioned For users whose time point exceeds the reference power consumption, if the user... The electricity consumption at a given time point is greater than that at the time point. Time point and If a user's electricity consumption exceeds the reference electricity consumption at a given time, then the power supply in the target area will be increased.

[0024] In a preferred embodiment of the present invention, in step S40, the variation characteristics of each user after the same power consumption are extracted, and the extraction steps are as follows:

[0025] Based on historical data of high-demand electricity consumption data groups, the electricity consumption distribution of users in the electricity consumption user groups is statistically analyzed, and the electricity consumption interval with the highest frequency of occurrence is selected as the benchmark threshold from the electricity consumption distribution.

[0026] From the regular electricity user group, select users whose electricity consumption is within the benchmark threshold for at least M days within a consecutive N natural days;

[0027] For the users, calculate the time point when the electricity consumption first reaches the benchmark threshold; mark the time point as the benchmark time point; mark the electricity consumption corresponding to the benchmark time point as the benchmark electricity consumption;

[0028] Extract the electricity consumption data of each user for T hours (e.g., 2 hours) after the reference time point to form a time window;

[0029] The standardization process includes aligning each user's time window to a reference time point.

[0030] Feature extraction is performed, including calculating the time it takes for each user to reach the baseline power consumption.

[0031] The type of change characteristic is defined based on the calculated time elapsed, including rapid rise type and smooth fluctuation type.

[0032] In a preferred embodiment of the present invention, the following steps are taken: the grid load set for the target area is obtained, and the load pattern characteristics are calculated among the users corresponding to the rapidly increasing type. The load pattern characteristics include load volatility. When the grid load corresponding to the load volatility exceeds half of the set grid load, the power supply in the target area is increased; otherwise, the power supply is not increased.

[0033] In a preferred embodiment of the present invention, the load pattern characteristics of the users corresponding to the rapidly rising type are calculated according to the following formula:

[0034] ;

[0035] In the formula, Indicates absolute volatility. This represents the number of sampling points within a time window (e.g., if sampling is performed every 15 minutes within 2 hours, then T=8).

[0036] Indicates a point in time Actual electricity consumption This indicates the baseline electricity consumption.

[0037] In a preferred embodiment of the present invention, the following formula is also included:

[0038] ;

[0039] In the formula, Indicates relative volatility. This represents the number of sampling points within the time window. Indicates a point in time Actual electricity consumption This indicates the baseline electricity consumption.

[0040] In a preferred embodiment of the present invention: based on the calculated load pattern characteristics, the set grid load is divided into at least 10 sub-load segments, and the time required to reach each sub-load segment is preset, and the time is marked as a reference time; at a future time, in the electricity consumption of users corresponding to the rapidly rising type, if the grid load corresponding to the electricity consumption reaches any sub-load segment at a time lower than the reference time, it is determined that the grid load is in a rapidly rising state, and the power supply is increased in the target area; otherwise, the power supply is not increased.

[0041] A computer device includes a processor, an input interface, an output interface, and a memory, wherein the processor, input interface, output interface, and memory are interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to execute the method described above.

[0042] A computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described above.

[0043] Beneficial effects:

[0044] 1. This invention, through a preset monitoring mechanism, including the setting of monitoring cycle, time point and critical power value, can monitor changes in power demand in real time. Once an increase in power demand is detected, the power grid load is calculated and it is decided whether to increase power supply, thereby effectively responding to sudden load changes.

[0045] 2. By calculating load characteristics (such as load fluctuation rate) and comparing them with the set grid load, the system can promptly identify abnormal changes in grid load. When load fluctuations exceed the safe range, the system will automatically increase power supply to ensure the safe and stable operation of the grid.

[0046] 3. This invention divides the power grid load into multiple sub-load segments and presets the time required to reach each sub-load segment. This hierarchical management method helps optimize the allocation of power resources and improve the efficiency of power supply. At the same time, by monitoring the changes in electricity consumption of rapidly increasing users, the upward trend of the power grid load can be predicted in advance, so as to take corresponding measures to avoid power grid overload. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the process structure of an embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0049] Because existing technologies struggle to depict the dynamic characteristics of electricity use in a region when predicting electricity demand based on a region's usage profile, the deviation rate between the profile and the actual state of electricity demand exceeds 15%, making it unable to cope with sudden load changes and poorly adaptable to complex dynamic changes in the power system.

[0050] Based on this, the present invention proposes a big data-based method for profiling the use of metered assets and forecasting and allocating demand. By constructing a database, distinguishing user electricity consumption patterns, dynamically monitoring changes in electricity demand, extracting key electricity consumption characteristics, quantifying load volatility, and implementing tiered load management, the method improves the accuracy and response speed of electricity demand forecasting and effectively addresses the dynamic changes of complex power systems.

[0051] The present solution will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0052] Reference Figures 1 to 2 This is one embodiment of the present invention, which provides a method for big data metering asset usage profiling and demand forecasting and allocation, including:

[0053] S10: Construct a database about the target area, which includes a high-demand electricity consumption data group and a low-demand electricity consumption data group, and obtain the user's electricity consumption data from the high-demand electricity consumption data group;

[0054] It should be noted that the electricity consumption of the high-demand electricity data group is more than 30% of the total daily electricity consumption of the target area;

[0055] In this embodiment, by distinguishing between high / low demand data groups, a structured data foundation is provided for subsequent analysis, reducing interference from irrelevant data and improving prediction efficiency;

[0056] Meanwhile, focusing on data from high-demand users can more accurately capture electricity consumption behavior characteristics, providing a basis for optimizing resource allocation;

[0057] S20: Differentiate high-demand electricity consumption data groups, including distinguishing them into regular electricity consumption user groups and irregular electricity consumption user groups, and calculate the proportion of regular electricity consumption user groups and irregular electricity consumption user groups in the high-demand electricity consumption data groups;

[0058] In this embodiment, the high-demand electricity consumption data group is further subdivided into a regular electricity consumption user group and an irregular electricity consumption user group, and the proportion of the two groups is calculated.

[0059] By classifying users, differentiated strategies can be developed for different electricity consumption patterns;

[0060] Regular user analysis helps predict base load, while irregular user analysis can capture sudden demand and improve the flexibility of demand response.

[0061] S30: Based on the calculated proportion, if the proportion of regular electricity users is large, then the electricity consumption data collected from the regular electricity users group will be divided into categories based on each natural day as the collection period. , , ,..., ; Indicates the first Electricity consumption data for each user;

[0062] In this embodiment, if the proportion of regular electricity users is high, their electricity consumption data is divided into cycles based on natural days.

[0063] Dividing data by calendar day can match actual business cycles (such as daily load forecasting), providing time dimension support for subsequent feature extraction and enhancing the timeliness of forecasts;

[0064] S40: Perform feature extraction on the electricity consumption data of users, including extracting the change features of each user after the same amount of electricity consumption, marking the same amount of electricity consumption as reference electricity consumption, the reference electricity consumption including the maximum electricity consumption of each user in the regular electricity consumption user group, and the average value calculated based on the maximum electricity consumption of each user;

[0065] In this embodiment, by standardizing the reference electricity consumption, the impact of differences in user electricity consumption on the analysis is eliminated, making feature extraction more comparable;

[0066] At the same time, the average value setting takes into account the overall electricity consumption level and improves the representativeness of the characteristics;

[0067] S50: Obtain the time node corresponding to the reference electricity consumption. Based on the time node, count the number of users in the regular electricity consumption user group who have reached the reference electricity consumption. If the proportion of the number is less than 50% in the regular electricity consumption user group, it is determined that the electricity demand in the target area is low. Otherwise, it is not determined.

[0068] In this embodiment, the demand status is objectively determined by quantitative indicators (proportion), avoiding subjective judgment errors;

[0069] Determining low demand can provide a basis for resource allocation and reduce oversupply;

[0070] S60: When the power demand in the target area is determined to be low, obtain the change pattern of power consumption of irregular power consumption user groups. If the power consumption changes towards the reference power consumption, it is determined that the power demand in the target area is increasing; otherwise, no determination is made.

[0071] In this embodiment, by monitoring irregular user behavior, potential demand growth signals are captured, and early warnings of load increase trends are provided to support dynamic resource allocation.

[0072] Meanwhile, if it is determined that the electricity demand is increasing, the power grid load corresponding to the target area is calculated. If the power grid load is stable, the power supply is not increased; if the power grid load is on an upward trend, the power supply is increased.

[0073] In S40, based on reference electricity consumption, if the number of users reaching the reference electricity consumption exceeds 50% of the total number of users in the regular electricity consumption user group, a monitoring mechanism is preset. Based on the monitoring mechanism, the electricity demand of the target area is predicted. The monitoring mechanism is as follows:

[0074] The monitoring period is preset, and a value is given within the monitoring period. Time point, Time point and The time points are defined as a monitoring cycle of at least 10 calendar days, with each given adjacent time point spaced 10 to 20 minutes apart.

[0075] Statistics The number of users who reach the reference electricity consumption level at a given time.

[0076] At An initial time point is preset before the current time point, and the initial time point is set to be at least equal to the previous time point. The time points are spaced 10 to 15 minutes apart. Within the initial time point, the maximum power consumption of each user before reaching the reference power consumption is obtained according to the number of users, and the maximum power consumption is marked as the critical power consumption value.

[0077] At For users whose electricity consumption reaches the reference level at a given time, if the user's electricity consumption is... If the time point exceeds the critical power consumption value, it is determined that the power demand in the target area is increasing.

[0078] At For users whose electricity consumption exceeds the reference limit at a certain time, if the user... The electricity consumption at a given time point is greater than that at the time point. Time point and If a user's electricity consumption exceeds the reference electricity consumption time, the power supply in the target area will be increased.

[0079] In this embodiment, a monitoring cycle, time point and critical power value are preset, the number of qualified users is counted and the power consumption is compared to determine whether demand increases or supply is increased.

[0080] By employing a refined monitoring mechanism, we can track changes in demand in real time and improve the accuracy of forecasts.

[0081] Meanwhile, setting a critical charge value can filter out noisy data and enhance the reliability of the judgment.

[0082] In S40, the variation characteristics of each user after the same amount of electricity consumption are extracted. The extraction steps are as follows:

[0083] Based on historical data (including electricity consumption data within the past six months) of high-demand electricity consumption data groups, the electricity consumption distribution of users in the electricity consumption user groups is statistically analyzed, and the electricity consumption range with the highest frequency (such as 50-60 kWh per day) is selected as the benchmark threshold.

[0084] From the regular electricity user group, select users whose electricity consumption is within the benchmark threshold for at least M days (e.g., 5 days) within a consecutive N natural days (e.g., 7 days);

[0085] For each user, calculate the time point when the electricity consumption first reaches the baseline threshold (e.g., the electricity consumption is 55kWh at 14:00 on the 3rd day); mark this time point as the baseline time point; and mark the electricity consumption corresponding to the baseline time point as the baseline electricity consumption.

[0086] Extract the electricity consumption data of each user for T hours (e.g., 2 hours after the base time point) to form a time window; (for example, if user B reaches 55kWh at 14:00, then extract the electricity consumption curve from 14:00 to 16:00)

[0087] Standardization is performed, which includes aligning each user's time window to a reference time point (e.g., 14:00 is time 0).

[0088] Feature extraction is performed, including calculating the time it takes for each user to reach a baseline electricity consumption (e.g., 2 hours from 0 to 55 kWh).

[0089] The type of change characteristic is defined based on the calculated time elapsed, including rapid rise type and steady fluctuation type;

[0090] In this embodiment, the benchmark threshold setting can focus on typical electricity consumption behaviors, and the screening of users who continuously meet the standards can exclude occasional data and improve the stability of the analysis.

[0091] The calculation of the baseline time point provides a time reference for subsequent feature extraction.

[0092] Obtain the grid load set for the target area (including the load obtained from the power supply bureau in the target area), and calculate the load pattern characteristics among users corresponding to the rapid increase type. The load pattern characteristics include load volatility. When the grid load corresponding to the load volatility exceeds half of the set grid load, the power supply in the target area is increased; otherwise, the power supply is not increased.

[0093] The load pattern characteristics of users corresponding to the rapid increase type are calculated using the following formula:

[0094] ;

[0095] In the formula, Indicates absolute volatility. This represents the number of sampling points within a time window (e.g., if sampling is performed every 15 minutes within 2 hours, then T=8).

[0096] Indicates a point in time Actual electricity consumption This indicates the baseline electricity consumption.

[0097] Example:

[0098] User A's electricity consumption sequence from 14:00 to 16:00 is [55, 56, 54, 57, 55, 53, 54, 55] kWh, based on the baseline value. =55kWh, then:

[0099] =1 / 8(|55-55|+|56-55|+...+|55-55|)=1+0+2+1+0+2+1+0 / 8=0.875kWh

[0100] It also includes calculations based on the following formula:

[0101] ;

[0102] In the formula, Indicates relative volatility. This represents the number of sampling points within the time window. Indicates a point in time Actual electricity consumption This indicates the baseline electricity consumption.

[0103] Example:

[0104] Using user A's data, the baseline value =55kWh, then:

[0105] ;

[0106] Based on the calculated load characteristics, the set grid load is divided into at least 10 sub-load segments, and the time required to reach each sub-load segment is preset and marked as the reference time. In the future, if the grid load corresponding to the electricity consumption of users with a rapidly rising load reaches any sub-load segment with a time lower than the reference time, it is determined that the grid load is in a rapidly rising state, and the power supply in the target area is increased; otherwise, the power supply is not increased.

[0107] In this embodiment, hierarchical management can refine load control strategies and improve resource utilization efficiency. Rapid load surge detection can respond promptly to sudden load changes, ensuring grid security.

[0108] A computer device includes a processor, an input interface, an output interface, and a memory, wherein the processor, input interface, output interface, and memory are interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to execute the method described above.

[0109] A computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described above.

[0110] In summary, this application significantly improves the accuracy and response speed of electricity demand forecasting by constructing a database, differentiating user electricity consumption behavior patterns, dynamically monitoring changes in electricity demand, extracting key electricity consumption characteristics, quantifying load volatility, and implementing tiered load management. It also enhances the safe and stable operation capability of the power grid, optimizes the allocation of power resources, effectively responds to the dynamic changes of complex power systems, and provides strong support for the precise management and efficient operation of the power industry.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for big data-driven asset usage profiling and demand forecasting and allocation, characterized in that, include: S10: Construct a database about the target area, the database including a high-demand electricity consumption data group and a low-demand electricity consumption data group, and obtain the user's electricity consumption data from the high-demand electricity consumption data group; S20: The high-demand electricity consumption data group is distinguished, including distinguishing it into regular electricity consumption user group and irregular electricity consumption user group, and the proportion of the regular electricity consumption user group and the irregular electricity consumption user group in the high-demand electricity consumption data group is calculated; S30: Based on the calculated proportion, if the proportion of the regular electricity-consuming user group is large, then the electricity consumption data collected from the regular electricity-consuming user group will be divided into categories based on each natural day as the collection period. , , ,..., ; Indicates the first Electricity consumption data for each user; S40: Perform feature extraction on the electricity consumption data of users, including extracting the change features of each user after the same amount of electricity consumption, and marking the same amount of electricity consumption as reference electricity consumption; S50: Obtain the time node corresponding to the reference electricity consumption, and according to the time node, count the number of users in the regular electricity consumption user group who have reached the reference electricity consumption. If the proportion of the number is less than 50% in the regular electricity consumption user group, it is determined that the electricity demand of the target area is low; otherwise, it is not determined. S60: When it is determined that the power demand in the target area is low, the power consumption change pattern of the irregular power consumption user group is obtained. If the power consumption changes towards the reference power consumption, it is determined that the power demand in the target area is increasing; otherwise, no determination is made.

2. The method for big data-based asset usage profiling and demand forecasting and allocation as described in claim 1, characterized in that, If it is determined that the electricity demand is increasing, the grid load corresponding to the target area is calculated. If the grid load is stable, the power supply is not increased. If the grid load is on an upward trend, then increase the power supply.

3. The method for big data-driven asset usage profiling and demand forecasting and allocation as described in claim 1, characterized in that, In S40, the reference electricity consumption includes the maximum electricity consumption of each user in the regular electricity consumption user group, and the average value calculated based on the maximum electricity consumption of each user; Based on the reference electricity consumption, if the number of users reaching the reference electricity consumption exceeds 50% of the total number of users in the regular electricity consumption group, a monitoring mechanism is preset. Based on the monitoring mechanism, the electricity demand of the target area is predicted. The monitoring mechanism is as follows: A preset monitoring period is given within the monitoring period. Time point, Time point and The time points are defined as a monitoring cycle of at least 10 calendar days, with each given adjacent time point spaced 10 to 20 minutes apart. Statistics in the above The number of users whose electricity consumption reaches the reference level at a given time point; In the aforementioned An initial time point is preset before the current time point, and the preset initial time point is at least equal to the current time point. The time points are spaced 10 to 15 minutes apart. Within the initial time point, the maximum power consumption of each user before reaching the reference power consumption is obtained according to the number of users, and the maximum power consumption is marked as the critical power consumption value. In the aforementioned For users whose electricity consumption reaches the reference level at a given time, if the user's electricity consumption is... If the time point exceeds the critical power value, it is determined that the power demand in the target area is increasing. In the aforementioned For users whose time point exceeds the reference power consumption, if the user... The electricity consumption at a given time point is greater than that at the time point. Time point and If a user's electricity consumption exceeds the reference electricity consumption at a given time, then the power supply in the target area will be increased.

4. The method for big data-driven asset usage profiling and demand forecasting and allocation as described in claim 1, characterized in that, In step S40, the variation characteristics of each user after the same amount of electricity consumption are extracted. The extraction steps are as follows: Based on historical data of high-demand electricity consumption data groups, the electricity consumption distribution of users in the electricity consumption user groups is statistically analyzed, and the electricity consumption interval with the highest frequency of occurrence is selected as the benchmark threshold from the electricity consumption distribution. From the regular electricity user group, select users whose electricity consumption is within the benchmark threshold for at least M days within a consecutive N natural days; For the users mentioned above, calculate the time point at which the electricity consumption first reaches the benchmark threshold. Mark the aforementioned time point as the baseline time point; The electricity consumption corresponding to the benchmark time point is marked as the benchmark electricity consumption; Extract the electricity consumption data of each user for T hours (e.g., 2 hours) after the reference time point to form a time window; The standardization process includes aligning each user's time window to a reference time point. Feature extraction is performed, including calculating the time it takes for each user to reach the baseline power consumption. The type of change characteristic is defined based on the calculated time elapsed, including rapid rise type and smooth fluctuation type.

5. The method for big data-based asset usage profiling and demand forecasting and allocation as described in claim 4, characterized in that, Obtain the grid load set for the target area, calculate the load pattern characteristics among the users corresponding to the rapid increase type, the load pattern characteristics include load volatility, and when the grid load corresponding to the load volatility exceeds half of the set grid load, increase the power supply in the target area; Conversely, the power supply will not be increased.

6. The method for big data-based asset usage profiling and demand forecasting and allocation as described in claim 5, characterized in that, The load pattern characteristics of users corresponding to the rapidly rising type are calculated according to the following formula: ； In the formula, Indicates absolute volatility. This represents the number of sampling points within a time window (e.g., if sampling is performed every 15 minutes within 2 hours, then T=8). Indicates a point in time Actual electricity consumption This indicates the baseline electricity consumption.

7. The method for big data-based asset usage profiling and demand forecasting and allocation as described in claim 6, characterized in that, It also includes calculations based on the following formula: ； In the formula, Indicates relative volatility. This represents the number of sampling points within the time window. Indicates a point in time Actual electricity consumption This indicates the baseline electricity consumption.

8. A method for big data metering asset usage profiling and demand forecasting and allocation as described in any one of claims 6 to 7, characterized in that, Based on the calculated load characteristics, the set grid load is divided into at least 10 sub-load segments, and the time required to reach each sub-load segment is preset, and the time is marked as a reference time. In the future, if the grid load corresponding to the electricity consumption of the user corresponding to the rapid increase type reaches any sub-load segment at a time lower than the reference time, it is determined that the grid load is in a rapid increase state, and the power supply in the target area is increased; otherwise, the power supply is not increased.

9. A computer device, characterized in that, The system includes a processor, an input interface, an output interface, and a memory, which are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 8.

Images