Classification method and device based on power utilization side customer load characteristics and storage medium

Abstract

The invention relates to a classification method and device based on power consumption side customer load characteristics and a storage medium, which are applied to the technical field of power market transaction.The method comprises the steps that on the basis of actual power consumption data of power market electricity selling side customers, which analysis method the power consumption customers are suitable for is judged, and for the customers with different power consumption load characteristics, analysis is performed; different analysis methods are adopted to finely classify electricity customers, analysis and classification management are carried out on the customers from an electricity load level, and an electricity selling company is helped to better carry out classification management and electricity transaction on the users.

Description

Technical Field

[0001] The present invention relates to the technical field of power market transactions, and particularly relates to a classification method, device, and storage medium based on the load characteristics of electricity customers on the power consumption side. Background Art

[0002] Due to the current classification management of electricity customers on the power consumption side in the power industry, electricity sales companies can only classify and manage different customers from the aspects of industry and the contract type of electricity customers, and cannot analyze the load characteristics of customers from a more microscopic perspective of customer load electricity consumption. Lack of effective analysis and classification management leads to chaotic user management, unclear user types, and inconsistent trading decisions in actual transactions. It is impossible to make trading judgments based on the electricity consumption characteristics of users and make scientific and efficient trading decisions.

[0003] Briefly speaking, the existing technology mainly relies on manual classification of customer types based on the contract type signed by customers, the industry to which customers belong, etc. The way of classifying and managing customers is relatively single and cannot make more detailed classifications according to the electricity consumption characteristics of customers. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a classification method, device, and storage medium based on the load characteristics of electricity customers on the power consumption side, so as to solve the problems in the existing technology that rely on manual classification of customer types based on the contract type signed by customers, the industry to which customers belong, etc., the way of classifying and managing customers is relatively single, and cannot make more detailed classifications according to the electricity consumption characteristics of customers.

[0005] According to the first aspect of the embodiments of the present invention, a classification method based on the load characteristics of electricity customers on the power consumption side is provided. The method includes: Obtain the daily time-of-use electricity consumption of electricity customers released by the trading center, obtain the four time periods of daily peak, valley, flat, and peak released by the trading center, obtain the time-of-use average curve and the typical decomposition curve released by the trading center, obtain the meteorological temperature curve of each city, and obtain the sub-T time period released by the trading center and the electricity consumption of electricity customers within each sub-T time period; According to the daily time-of-use electricity consumption of the electricity customers and the four time periods of daily peak, valley, flat, and peak, obtain the electricity consumption of the electricity customers in the four time periods of peak, valley, flat, and peak within the target time limit range. If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is greater than or equal to the preset peak-valley difference threshold, then use the peak-valley-flat classification method to classify the electricity customers; If the difference between the electricity consumption during the peak period and the electricity consumption during the valley period is less than the preset peak-valley difference threshold, then obtain the total time-of-use electricity consumption of the electricity customer within the target time limit according to the daily time-of-use electricity consumption of the electricity customer, obtain the time-of-use electricity consumption curve according to the total time-of-use electricity consumption, and obtain the correlation between the time-of-use electricity consumption curve and the meteorological temperature curves of each city, or, the time-period average curve and the typical decomposition curve. If there is any correlation greater than the preset correlation threshold, then classify the electricity customer by using the correlation analysis method; If there is no correlation greater than the preset correlation threshold, then determine whether there is a specified-period electricity consumption in the electricity consumption of the electricity customer during each sub-T period. If there is, then classify the electricity customer by using the sub-T electricity consumption characteristic analysis method; If there is no specified-period electricity consumption, then classify the electricity customer by using the clustering analysis method.

[0006] Preferably, The obtaining of the correlation between the time-of-use electricity consumption curve and the meteorological temperature curves of each city, or, the time-period average curve and the typical decomposition curve, and if there is any correlation greater than the preset correlation threshold, then classifying the electricity customer by using the correlation analysis method includes: Respectively obtain the correlations between the time-of-use electricity consumption curve and the time-period average curve and the typical decomposition curve published by the trading center. If there is any correlation greater than the preset correlation threshold, then classify the electricity customer by using the decomposition curve correlation analysis method; If there is no correlation greater than the preset correlation threshold, obtain the meteorological temperature curves of each city within the target time limit according to the meteorological temperature curves of each city, respectively obtain the correlations between the time-of-use electricity consumption curve and the meteorological temperature curves of each city within the target time limit. If there is any correlation greater than the preset correlation threshold, then classify the electricity customer by using the temperature correlation analysis method.

[0007] Preferably, The classifying of the electricity customer by using the peak, flat, valley and spike analysis method includes: According to the daily time-of-use electricity consumption of the electricity customer and the four periods of peak, flat, valley and spike per day, obtain the electricity consumption of the electricity customer during the four periods of peak, flat, valley and spike per day. Add up the electricity consumption during the four periods of peak, flat, valley and spike per day within the target time limit according to the same period type to obtain the electricity consumption of the electricity customer during the four periods of peak, flat, valley and spike within the target time limit; According to the electricity consumption of the electricity customer during the four periods of peak, flat, valley and spike within the target time limit, classify the electricity customer as a peak customer, a high-peak customer, a flat-period customer or a low-valley customer according to the period with the largest electricity consumption ratio.

[0008] Preferably, The classification of electricity customers using the decomposition curve correlation analysis method includes: Using the Pearson correlation analysis method, taking the time-of-use electricity consumption curve of electricity customers as the ordinate and the equally divided curve or typical decomposition curve by time period published by the trading center as the abscissa, obtaining the linear correlation values between electricity customers and each curve, and taking the equally divided curve or typical decomposition curve corresponding to the maximum correlation value as the curve type of electricity customers.

[0009] Preferably, The classification of electricity customers using the temperature correlation analysis method includes: Using the Pearson correlation analysis method, taking the time-of-use electricity consumption curve of electricity customers as the ordinate and the meteorological temperature curve of each city within the target time limit as the abscissa, obtaining the linear correlation values between electricity customers and each meteorological temperature curve, and classifying electricity customers into the city with the maximum correlation.

[0010] Preferably, The classification of electricity customers using the sub-T electricity consumption characteristic analysis method includes: Obtaining the classification results of electricity customers by the peak-valley-flat classification method; For all peak users, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each peak user. If not, it is a peak regular user; if so, it is a peak specified user; For all high-peak users, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each high-peak user. If not, it is a high-peak regular user; if so, it is a high-peak specified user; For all flat-section users, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each flat-section user. If not, it is a flat-section regular user; if so, it is a flat-section specified user; For all low-valley users, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each low-valley user. If not, it is a low-valley regular user; if so, it is a low-valley specified user.

[0011] Preferably, The classification of electricity customers using the clustering analysis method includes: Obtaining the total time-of-use electricity consumption of electricity customers within the target time limit; taking the total time-of-use electricity consumption as a data set; Determining the clustering K value and randomly selecting initial centroids for each cluster in the data set; Clustering the data set according to the minimum distance principle; Using the sample mean of each updated cluster as the new centroid of each cluster; Repeat the clustering process until the difference between the new centroid of each cluster and the previous centroid is less than a preset threshold or the preset number of iterations is reached, obtaining the final centroid and K cluster partitions, and classifying electricity-consuming customers into different clusters.

[0012] According to the second aspect of the embodiments of the present invention, there is provided a classification device based on the load characteristics of electricity-consuming customers on the power consumption side, and the device includes: A data acquisition module: configured to acquire the daily time-of-use power consumption of electricity-consuming customers published by the trading center, acquire the four time periods of peak, valley, flat, and peak of each day published by the trading center, acquire the time-of-use average curve and the typical decomposition curve published by the trading center, acquire the meteorological temperature curve of each city, and acquire the sub-T time periods published by the trading center and the power consumption of electricity-consuming customers in each sub-T time period; A peak evaluation and classification module: configured to acquire the power consumption of electricity-consuming customers in the four time periods of peak, valley, flat, and peak within the target time limit according to the daily time-of-use power consumption of the electricity-consuming customers and the four time periods of peak, valley, flat, and peak of each day. If the difference between the power consumption in the peak period and the power consumption in the valley period is greater than or equal to a preset peak-valley difference threshold, then use the peak-valley-flat classification method to classify the electricity-consuming customers; A correlation classification module: configured to, if the difference between the power consumption in the peak period and the power consumption in the valley period is less than a preset peak-valley difference threshold, acquire the total time-of-use power consumption of the electricity-consuming customers within the target time limit according to the daily time-of-use power consumption of the electricity-consuming customers, acquire the time-of-use power consumption curve according to the total time-of-use power consumption, acquire the correlation between the time-of-use power consumption curve and the meteorological temperature curve of each city, or the time-of-use average curve and the typical decomposition curve. If there is any correlation greater than a preset correlation threshold, then use the correlation analysis method to classify the electricity-consuming customers; A sub-T classification module: configured to, if there is no correlation greater than a preset correlation threshold, determine whether there is a specified time period power consumption in the power consumption of the electricity-consuming customers in each sub-T time period. If so, use the sub-T power consumption characteristic analysis method to classify the electricity-consuming customers; A clustering classification module: configured to, if there is no specified time period power consumption, use the clustering analysis method to classify the electricity-consuming customers.

[0013] According to the third aspect of the embodiments of the present invention, there is provided a storage medium storing a computer program, and when the computer program is executed by a main controller, each step in the above method is implemented.

[0014] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: Based on the actual power consumption data of electricity customers on the selling side of the power market, this application determines which analysis method is applicable to the electricity customers. For customers with different electricity load characteristics, different analysis methods are used to conduct a detailed classification of the electricity customers, and the customers are analyzed and classified from the electricity load level to help the electricity selling company better classify and manage users and conduct electricity transactions.

[0015] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present invention and used together with the specification to explain the principles of the present invention.

[0017] Figure 1 is a schematic flow chart of a classification method based on the load characteristics of electricity customers on the electricity consumption side shown according to an exemplary embodiment; Figure 2 is a schematic system diagram of a classification device based on the load characteristics of electricity customers on the electricity consumption side shown according to another exemplary embodiment; In the drawings: 1 - data acquisition module, 2 - peak evaluation classification module, 3 - correlation classification module, 4 - sub-T classification module, 5 - clustering classification module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are only examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0019] Embodiment 1 Figure 1 is a schematic flow chart of a classification method based on the load characteristics of electricity customers on the electricity consumption side shown according to an exemplary embodiment, as Figure 1 shown, the method includes: S1, obtaining the daily time-of-use electricity consumption of electricity customers released by the trading center, obtaining the four time periods of peak, high, flat, and valley released by the trading center every day, obtaining the evenly divided curve by time period and the typical decomposition curve released by the trading center, obtaining the meteorological temperature curve of each city, and obtaining the sub-T time period released by the trading center and the electricity consumption of electricity customers in each sub-T time period; S2. Based on the daily time-of-use electricity consumption of the electricity customers and the four periods of peak, valley, flat, and off-peak in a day, obtain the electricity consumption of the electricity customers in the four periods of peak, valley, flat, and off-peak within the target time limit. If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is greater than or equal to the preset peak-valley difference threshold, then use the peak-valley-flat-valley type analysis method to classify the electricity customers; S3. If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is less than the preset peak-valley difference threshold, then obtain the total time-of-use electricity consumption of the electricity customers within the target time limit based on the daily time-of-use electricity consumption of the electricity customers. Obtain the time-of-use electricity consumption curve based on the total time-of-use electricity consumption. Obtain the correlation between the time-of-use electricity consumption curve and the meteorological temperature curve of each city, or the equally divided curve by time period and the typical decomposition curve. If there is any correlation greater than the preset correlation threshold, then use the correlation analysis method to classify the electricity customers; S4. If there is no correlation greater than the preset correlation threshold, then determine whether there is a specified period electricity consumption in the electricity consumption of the electricity customers in each sub-T period. If there is, then use the sub-T electricity consumption characteristic analysis method to classify the electricity customers; S5. If there is no specified period electricity consumption, then use the clustering analysis method to classify the electricity customers; It can be understood that the data collection includes: Customer time-of-use electricity consumption: The daily time-of-use electricity consumption of the electricity customers, which is announced by the trading center; Peak-valley-flat-valley periods: The power trading center divides the electricity consumption at 24:00 every day into four periods of peak, valley, flat, and off-peak. The type of period for each month is announced by the trading center; Decomposition curve: The medium- and long-term electricity trading volume in the electricity market is decomposed to the load side on a daily, monthly, and hourly basis according to the decomposition curve function given by the trading center. The formation methods of the decomposition curve include the equally divided curve by time period and the typical decomposition curve. These two types of decomposition curves are announced by the trading center; Temperature: The temperature of the trading province of the electricity sales company. Meteorological temperature is an important influencing factor in the electricity market trading. Analyzing whether there is a correlation between the load of electricity customers on the power consumption side and the temperature can help understand the electricity consumption habits of customers. The data source is generated independently by each participating entity; Customer's time-of-use electricity consumption: Obtain the electricity consumption at the metering point of the user each month. This electricity consumption is displayed in time period types of T1, T2, T3, T4, T5, and T6, and this data is announced by the trading center. It should be emphasized that generally, the time period types of T1, T2, T3, and T4 are regular time periods, which means that the 24-hour day is divided into 4 regular types. The number of time periods within each type is subject to the results announced by the trading center. The regular time periods do not overlap with each other. The time period types of T5 and T6 are designated time periods. For example, T5 can be the first 12 hours of a day, and T6 is the last 12 hours of a day. Obviously, T5 or T6 may overlap with the time period types of T1, T2, T3, and T4. After simply organizing and preliminarily classifying the obtained customer electricity consumption data, different analysis methods can be adopted according to the results for more detailed classification, specifically including: When the time-of-use electricity consumption of customers in a province has obvious peak-valley characteristics (for example, more electricity consumption during peak periods and less during valley periods, that is, the peak-valley difference is greater than the preset peak-valley difference threshold), the peak-valley classification analysis method is adopted: Allocate the electricity consumption of customers at 24 o'clock to the peak, flat, and valley periods according to the rules of the power trading center, calculate the total electricity consumption in each period, and then divide the electricity customers into peak customers, high-peak customers, flat-period customers, and low-valley customers according to the classification rules of the power sales company itself. The specific implementation steps of the above algorithm: Step 1: Obtain the time-of-use electricity consumption of electricity customers and the time points included in the peak, high-peak, flat, and valley period types of the current month. Step 2: Obtain the time-of-use electricity consumption of the target customer within the date range and sum it up according to the time points. Step 3: Set classification rules. Generally, the period with the most electricity consumption among the peak, high-peak, flat, and valley periods is used as the classification result of the electricity customer, that is, peak customers, high-peak customers, flat-period customers, and low-valley customers.

[0020] If the time-of-use electricity consumption of electricity customers does not have obvious peak-valley characteristics, then judge whether there is a certain degree of correlation between the time-of-use electricity consumption of customers and the various decomposition curves announced by the trading center. That is, obtain the total time-of-use electricity consumption of electricity customers within the target time range according to the daily time-of-use electricity consumption of electricity customers, obtain the time-of-use electricity curve according to the total time-of-use electricity consumption, and then obtain the correlations between the time-of-use electricity curve and the evenly divided time-period curve and the typical decomposition curve respectively. If any of the correlations is greater than the preset correlation threshold, the decomposition curve correlation analysis method is used to classify the electricity customers. Analysis of the correlation of decomposition curves: Select the decomposition curves, fit the time-of-use electricity consumption within the target customer's date range to the decomposition curves, and use the Pearson correlation coefficient analysis method. Classify the customers with a high degree of fit between the time-of-use electricity consumption and the decomposition curves into the same group, so as to manage customers by classification through the decomposition curves; Specific implementation steps of the above algorithm: Step 1: Obtain the target decomposition curve within the target date range of the customer; Step 2: Obtain the time-of-use electricity consumption within the target customer's date range and sum it up by time point to get the total time-of-use electricity consumption of the electricity customer within the target period, that is, the total electricity consumption in 24 hours of a day within the target period, corresponding to 24 electricity consumption data. Connect the 24 electricity consumption data in sequence to obtain the time-of-use electricity curve; Step 3: Use the Pearson correlation analysis method, with the customer as the ordinate and the decomposition curve as the abscissa, to measure the linear correlation between the customer and each curve data. Its value range is from -1 to 1, where 1 represents a perfect positive correlation, -1 represents a perfect negative correlation, and 0 represents no linear correlation. Calculate the correlation value between the customer and the decomposition curve, and take the curve with the largest correlation as the curve type of the customer; If the correlation values of the decomposition curves are all less than the preset correlation threshold, then obtain the meteorological temperature curves of each city within the target period according to the meteorological temperature curves of each city, and respectively obtain the correlations between the time-of-use electricity curve and the meteorological temperature curves of each city within the target period. If there is any correlation greater than the preset correlation threshold, use the temperature correlation analysis method to classify the electricity customers; Temperature correlation analysis: Select the time-of-use electricity consumption of the target customer on the target date, select the temperatures of each city in the province as a reference, conduct a Pearson correlation analysis on the temperatures of the target cities in the province and the time-of-use electricity, analyze the relationship between the customer's load electricity and the temperatures of each city, and classify them; Specific implementation steps of the above algorithm: Step 1: Obtain the time-of-use electricity curve within the target customer's date range; Step 2: Select the cities for which the temperature correlation analysis is to be carried out; Step 3: Use the Pearson correlation analysis method, with the customer as the ordinate and the temperatures of each city as the abscissa, to measure the linear correlation between the customer and the temperature data of each city. Its value range is from -1 to 1, where 1 represents a perfect positive correlation, -1 represents a perfect negative correlation, and 0 represents no linear correlation. Classify the customer into the city with the largest correlation; If there is no correlation greater than the preset correlation threshold, then judge whether there is a specified period of electricity consumption (that is, the T5 period type or the T6 period type) in the electricity consumption of the electricity customer in each sub-T period. If so, use the sub-T electricity consumption characteristic analysis method to classify the electricity customers; Analysis method of electricity consumption characteristics for classification: Step 1: First, conduct a preliminary classification in combination with Analysis Method 1; Step 2: Then, determine whether each type of electricity-consuming customer (peak customers, high-peak customers, flat-period customers, low-valley customers) only contains regular-period electricity (that is, the electricity consumption of the electricity-consuming customer during a day is concentrated in T1, T2, T3, or T4, and does not include the electricity consumption during the specified period. For example, if the electricity consumption of an electricity-consuming customer is from 8 to 12 in a day, concentrated in the T2 type period, and the specified period T5 is from 13 to 18, and the specified period T6 is from 19 to 24, then the electricity consumption type of this electricity-consuming customer does not include the electricity consumption during the specified period). Such customers are regular customers. If the electricity consumption data of the electricity-consuming customer includes the T5 or T6 period, then they are specified customers. Of course, for specified customers, further subdivision can be carried out, that is, whether they are single-specified-period customers or multi-specified-period customers; When there is no obvious pattern in the electricity consumption demands among most customers within a province and it is difficult to directly classify the customers simply, kmeans clustering analysis can be used to classify the electricity-consuming customers; Select the time-of-use electricity consumption of customers within the target date range, select the number of clusters for data partitioning, and through an iterative method, partition the data set into the clusters represented by the closest cluster center points. Then, recalculate the center point (take the average value) of each cluster based on all the points within each cluster, and continuously repeat this process until the change in the cluster center point is very small or reaches the specified number of iterations, and finally divide the customers into different clusters; Specific implementation steps of the above algorithm: Step 1: Obtain the time-of-use electricity consumption of customers within the target date range; Step 2: Determine the clustering k value; Step 3: Randomly select initial centroids for K clusters in the data set; Step 4: Cluster the data set according to the principle of minimum distance; Step 5: Iteratively update the centroids using the sample means of the K clusters; Step 6: Repeat Steps 3 and 4 until the centroids are stable and no longer change; Step 7: Output the final centroids and the partitioning of the K clusters.

[0021] Embodiment 2: Figure 2 It is a system schematic diagram of a classification device based on the load characteristics of electricity-consuming customers on the power consumption side shown in another exemplary embodiment. The device includes: Data acquisition module 1: It is used to obtain the daily time-of-use electricity consumption of electricity customers released by the trading center, obtain the four time periods of peak, valley, flat, and valley released by the trading center every day, obtain the evenly distributed curve by time period and the typical decomposition curve released by the trading center, obtain the meteorological temperature curve of each city, and obtain the sub-T time period released by the trading center and the electricity consumption of electricity customers within each sub-T time period; Peak evaluation classification module 2: It is used to obtain the electricity consumption of electricity customers in the four time periods of peak, valley, flat, and valley within the target time limit according to the daily time-of-use electricity consumption of the electricity customers and the four time periods of peak, valley, flat, and valley every day. If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is greater than or equal to the preset peak-valley difference threshold, the peak-valley flat valley analysis method is used to classify the electricity customers; Correlation classification module 3: If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is less than the preset peak-valley difference threshold, it is used to obtain the total time-of-use electricity consumption of the electricity customers within the target time limit according to the daily time-of-use electricity consumption of the electricity customers, obtain the time-of-use electricity curve according to the total time-of-use electricity consumption, obtain the correlation between the time-of-use electricity curve and the meteorological temperature curve of each city, or the evenly distributed curve by time period and the typical decomposition curve. If there is any correlation greater than the preset correlation threshold, the correlation analysis method is used to classify the electricity customers; Sub-T classification module 4: If there is no correlation greater than the preset correlation threshold, it is used to judge whether there is a specified time period electricity consumption in the electricity consumption of the electricity customers within each sub-T time period. If so, the sub-T electricity consumption characteristic analysis method is used to classify the electricity customers; Clustering classification module 5: If there is no specified time period electricity consumption, the clustering analysis method is used to classify the electricity customers.

[0022] Embodiment 3: This embodiment provides a storage medium. The storage medium stores a computer program. When the computer program is executed by the main controller, each step in the above method is implemented; It can be understood that the above-mentioned storage medium can be a read-only memory, a magnetic disk or an optical disc, etc.

[0023] It can be understood that the same or similar parts in the above embodiments can be referred to each other. The content not detailed in some embodiments can be seen in the same or similar content in other embodiments.

[0024] It should be noted that in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise stated, the meaning of "a small number of sparsely distributed" means at least two.

[0025] Any process or method description depicted in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a specific logical function or process step, and the scope of the preferred embodiments of the present invention includes additional implementations where functions may be performed not in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the relevant functions, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0026] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, a small number of sparsely distributed steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0027] Those of ordinary skill in the art of the present technology can understand that all or part of the steps carried by the method of implementing the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0028] In addition, in each embodiment of the present invention, the functional units can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0029] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, etc.

[0030] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or a small number of sparsely distributed embodiments or examples.

[0031] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A classification method based on the load characteristics of electricity customers on the consumption side, characterized in that The method includes: Obtaining the daily time-of-use electricity consumption of electricity customers released by the trading center, obtaining the four time periods of peak, valley, flat, and peak of each day released by the trading center, obtaining the equally divided curve by time period and the typical decomposition curve released by the trading center, obtaining the meteorological temperature curve of each city, and obtaining the sub-T time periods released by the trading center and the electricity consumption of electricity customers within each sub-T time period; According to the daily time-of-use electricity consumption of the electricity customer and the four time periods of peak, valley, flat, and peak of each day, obtain the electricity consumption of the electricity customer in the four time periods of peak, valley, flat, and peak within the target time limit. If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is greater than or equal to the preset peak-valley difference threshold, then use the peak-valley-flat-valley type analysis method to classify the electricity customers; If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is less than the preset peak-valley difference threshold, obtain the total time-of-use electricity consumption of the electricity customer within the target time limit according to the daily time-of-use electricity consumption of the electricity customer, obtain the time-of-use electricity curve according to the total time-of-use electricity consumption, and obtain the correlation between the time-of-use electricity curve and the meteorological temperature curve of each city, or the equally divided curve by time period and the typical decomposition curve. If there is any correlation greater than the preset correlation threshold, then use the correlation analysis method to classify the electricity customers; If there is no correlation greater than the preset correlation threshold, determine whether there is a specified time period electricity consumption in the electricity consumption of the electricity customer within each sub-T time period. If so, use the sub-T electricity consumption characteristic analysis method to classify the electricity customers; If there is no specified time period electricity consumption, use the clustering analysis method to classify the electricity customers.

2. The method according to claim 1, wherein The obtaining the correlation between the time-of-use electricity curve and the meteorological temperature curve of each city, or the equally divided curve by time period and the typical decomposition curve, and if there is any correlation greater than the preset correlation threshold, then using the correlation analysis method to classify the electricity customers includes: Respectively obtain the correlation between the time-of-use electricity curve and the equally divided curve by time period and the typical decomposition curve released by the trading center. If there is any correlation greater than the preset correlation threshold, then use the decomposition curve correlation analysis method to classify the electricity customers; If there is no correlation greater than the preset correlation threshold, obtain the meteorological temperature curve of each city within the target time limit according to the meteorological temperature curve of each city, and respectively obtain the correlation between the time-of-use electricity curve and the meteorological temperature curve of each city within the target time limit. If there is any correlation greater than the preset correlation threshold, then use the temperature correlation analysis method to classify the electricity customers.

3. The method according to claim 2, wherein The using the peak-valley-flat-valley type analysis method to classify the electricity customers includes: According to the daily time-of-use electricity consumption of the electricity customer and the four time periods of peak, valley, flat, and peak of each day, obtain the electricity consumption of the electricity customer in the four time periods of peak, valley, flat, and peak of each day, and add the electricity consumption in the four time periods of peak, valley, flat, and peak within the target time limit according to the same time period type to obtain the electricity consumption of the electricity customer in the four time periods of peak, valley, flat, and peak within the target time limit; According to the electricity consumption of the electricity customers in the four time periods of peak, high, flat, and valley within the target time limit, the electricity customers are classified into peak customers, high customers, flat customers, or valley customers according to the time period with the largest electricity consumption ratio.

4. The method according to claim 2, wherein the classification of electricity customers by using the decomposition curve correlation analysis method includes: Using the Pearson correlation analysis method, taking the time-of-use electricity consumption curve of the electricity customers as the ordinate and the equally divided curve or typical decomposition curve of each time period published by the trading center as the abscissa, obtaining the linear correlation value between the electricity customers and each curve, and taking the equally divided curve or typical decomposition curve corresponding to the maximum correlation value as the curve type of the electricity customers.

5. The method according to claim 2, wherein the classification of electricity customers by using the temperature correlation analysis method includes: Using the Pearson correlation analysis method, taking the time-of-use electricity consumption curve of the electricity customers as the ordinate and the meteorological temperature curve of each city within the target time limit as the abscissa, obtaining the linear correlation value between the electricity customers and each meteorological temperature curve, and classifying the electricity customers into the city with the largest correlation.

6. The method according to claim 5, wherein the classification of electricity customers by using the sub-T electricity consumption characteristic analysis method includes: Obtaining the classification result of the peak, flat, and valley analysis method for electricity customers; For all peak customers, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each peak customer. If not, it is a peak regular customer. If so, it is a peak specified customer; For all high customers, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each high customer. If not, it is a high regular customer. If so, it is a high specified customer; For all flat customers, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each flat customer. If not, it is a flat regular customer. If so, it is a flat specified customer; For all valley customers, judge whether the specified time period type is included according to the daily time-of-use electricity consumption of each valley customer. If not, it is a valley regular customer. If so, it is a valley specified customer.

7. The method according to claim 6, wherein the classification of electricity customers by using the clustering analysis method includes: Obtaining the total time-of-use electricity consumption of the electricity customers within the target time limit; taking the total time-of-use electricity consumption as a data set; Determining the clustering K value, and randomly selecting an initial centroid for each cluster in the data set; Clustering the data set according to the minimum distance principle; Using the sample mean of each updated cluster as the new centroid of each cluster; Repeating the clustering process until the difference between the new centroid of each cluster and the previous centroid is less than the preset threshold or reaches the preset number of iterations, obtaining the final centroid and K clustering partitions, and classifying the electricity customers into different clusters.

8. Classification device based on the load characteristics of electricity customers on the power consumption side, characterized in that The device includes: Data acquisition module: It is used to acquire the daily time-of-use electricity consumption of electricity customers released by the trading center, acquire the four time periods of peak, valley, flat, and valley released by the trading center every day, acquire the time-of-use average curve and the typical decomposition curve released by the trading center, acquire the meteorological temperature curve of each city, and acquire the sub-T time periods released by the trading center and the electricity consumption of electricity customers within each sub-T time period; Peak evaluation classification module: It is used to acquire the electricity consumption of the electricity customer in the four time periods of peak, valley, flat, and valley within the target time limit according to the daily time-of-use electricity consumption of the electricity customer and the four time periods of peak, valley, flat, and valley every day. If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is greater than or equal to the preset peak-valley difference threshold, the peak-valley flat-valley analysis method is used to classify the electricity customer; Correlation classification module: If the difference between the electricity consumption in the peak period and the electricity consumption in the valley period is less than the preset peak-valley difference threshold, it is used to acquire the total time-of-use electricity consumption of the electricity customer within the target time limit according to the daily time-of-use electricity consumption of the electricity customer, acquire the time-of-use electricity curve according to the total time-of-use electricity consumption, acquire the correlation between the time-of-use electricity curve and the meteorological temperature curve of each city, or the time-of-use average curve and the typical decomposition curve. If there is any correlation greater than the preset correlation threshold, the correlation analysis method is used to classify the electricity customer; Sub-T classification module: If there is no correlation greater than the preset correlation threshold, it is used to judge whether there is a specified period of electricity consumption in the electricity consumption of the electricity customer in each sub-T time period. If so, the sub-T electricity consumption characteristic analysis method is used to classify the electricity customer; Clustering classification module: If there is no specified period of electricity consumption, the clustering analysis method is used to classify the electricity customer.

9. A storage medium, characterized in that, The storage medium stores a computer program, and when the computer program is executed by the main controller, it realizes each step in the classification method based on the load characteristics of electricity customers on the power consumption side described in any one of claims 1-7.

Images