A centralized management method for charging power banks based on users' electricity consumption needs

Abstract

The present invention discloses a centralized management method for charging power banks based on users' electricity consumption demands, which relates to the field of centralized management of power bank charging and includes: collecting multi-dimensional data information of the power banks under centralized management during the period from lending to return, and analyzing the electricity consumption demand patterns of users; customizing charging management strategies according to the characteristics of user electricity consumption mode tags to achieve rapid rotation of power banks and targeted optimized allocation of charging resources; when the user electricity consumption mode tag is the rapid turnover mode, real-time monitoring the heat generation situation of each power bank and making secondary adjustments to the charging strategy based on the heat generation situation; analyzing the historical demand peak time period data of the site, setting up peak buffer devices in advance, and adopting dynamic hierarchical charging management for the buffer devices, thus achieving efficient, accurate and intelligent centralized charging management of power banks.

Description

Technical Field

[0001] The present invention relates to the field of centralized management of power bank charging, and in particular to a centralized management method of power bank charging based on user power demand. Background Art

[0002] Traditional charging management models face many challenges, especially in terms of user demand fluctuations, equipment utilization optimization, and energy efficiency improvement. Existing systems can usually only simply record basic information such as the time and place of lending and returning power banks, and lack the detailed collection and management of multi-dimensional data during the use of power banks. This makes it difficult to accurately identify users' personalized electricity needs, and charging strategies lack refinement and pertinence, resulting in low equipment utilization efficiency, uneven resource allocation, and inability to achieve efficient charging management.

[0003] This solution forms a personalized charging strategy through intelligent analysis, thereby optimizing the rapid rotation of power banks and the allocation of charging resources. At the same time, in order to deal with temperature issues under high-frequency usage modes, the system should have the ability to monitor heat in real time and dynamically adjust charging strategies to ensure equipment safety and extend battery life. In addition, the system needs to analyze the historical peak demand data of the site, configure buffer devices in advance, and manage these devices dynamically in layers to ensure that user needs can be quickly responded to during peak hours and reduce resource waste during non-peak hours, thereby achieving efficient, accurate and intelligent charging management. Summary of the invention

[0004] In order to solve the above technical problems, a centralized management method for charging power banks based on user electricity demand is provided. This technical solution solves the problems raised in the above background technology.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A centralized management method for charging power banks based on user power demand, comprising:

[0007] Collect multi-dimensional data information from the lending to the return of power banks for centralized management, and analyze the user's electricity demand pattern;

[0008] Customize charging management strategies based on user power usage pattern label characteristics to achieve rapid rotation of power banks and targeted optimization of charging resources;

[0009] When the user's power usage mode label is fast turnover mode, the heating condition of each power bank is monitored in real time, and the charging strategy is adjusted again based on the heating condition;

[0010] Analyze the historical peak demand time period data of the site, set up peak buffer equipment in advance, and adopt dynamic tiered charging management for the buffer equipment.

[0011] Preferably, the collecting of multi-dimensional data information of the power bank from the time of lending to the time of returning for centralized management and the analysis of the user's electricity demand pattern specifically include:

[0012] Count the number of times power banks are lent out within a set time window, and calculate the lending frequency to judge the intensity of users' demand for power banks;

[0013] Accurately track the return of all borrowed power banks, record the actual usage time of each user's rented power bank and the time interval between the final return, and analyze the time characteristics of users returning the power bank after borrowing;

[0014] According to the GPS location information of the power bank, the flow path of the power bank between different locations is recorded to form the user's site characteristics;

[0015] Based on the collected data on the lending frequency, time period characteristics and site features of power banks used by users, users' electricity demand patterns are identified through multi-layer clustering analysis.

[0016] Preferably, the identifying the user's electricity demand pattern through multi-layer cluster analysis based on the collected lending frequency, time period characteristics and site characteristics of the user's power bank specifically includes:

[0017] After the collected data is preliminarily cleaned, the first-level cluster analysis is performed on the data, using the borrowing frequency and return interval as the target features to identify the user's initial electricity demand pattern, and the initial electricity demand pattern is divided into two categories: high-frequency borrowing and low-frequency borrowing;

[0018] Based on the first-level clustering, the second-level clustering analysis is performed on the user's usage time, and the demand pattern is combined with the first-level clustering analysis results to identify the user's specific electricity demand pattern;

[0019] High-frequency and low-frequency borrowing users are further subdivided into short-term and long-term users, and the mode scenarios are combined into short-term high-demand scenarios and long-term low-frequency demand scenarios;

[0020] Based on the second-level clustering results, combined with the user's geographical location and time characteristics, a third clustering analysis is conducted on the electricity demand patterns of different sites to form a specific identification result of the site user's dynamic electricity demand;

[0021] Based on the results of the three-layer clustering analysis, specific user electricity usage pattern labels are generated for each site, including rapid turnover mode and resource optimization mode.

[0022] Preferably, the customization of charging management strategy according to the characteristics of the user's power usage mode label to achieve rapid rotation of power banks and targeted optimization allocation of charging resources specifically includes:

[0023] Differentiate and analyze the user power usage pattern labels at each site and plan charging strategies;

[0024] The specific steps of distinguishing and analyzing the user power consumption mode labels and planning the charging strategy for each site are as follows:

[0025] When the user's power usage mode label is fast turnover mode, within the specification range of the charging equipment, a fast charging mode with a current greater than the normal charging current is set, and charging priority is assigned in real time according to the return order and power status of the power bank;

[0026] Obtain all power banks that are being charged, set the target power of the power banks according to the current user charging demand, monitor the power banks that are being charged in real time, and when there is a power bank whose real-time power reaches the target power, remove the rental restrictions on the power bank and put it back into use;

[0027] When the user's power usage mode label is resource optimization mode, the returned power bank is charged slowly with low current, and the minimum power of the stage charging target is set to full power;

[0028] The returned power banks are evaluated for health by testing battery parameters, and the charging priorities of the power banks to be charged are sorted based on the health evaluation results;

[0029] Based on the distribution of electricity costs and historical usage demand, off-peak charging is arranged through electricity price forecasts, and charging tasks are scheduled during periods when electricity prices are lower.

[0030] Preferably, when the user's power usage mode label is a fast turnover mode, real-time monitoring of the heating condition of each power bank and secondary adjustment of the charging strategy based on the heating condition specifically include:

[0031] A temperature sensing device is installed inside or on the surface of each power bank to monitor the temperature changes during charging in real time, remove noise from the collected temperature data, and filter out the impact of short-term temperature fluctuations;

[0032] Obtain the safety specification standard temperature and the highest acceptable charging temperature of each power bank, calculate the pulse charging cycle based on the adaptive algorithm of temperature feedback, and dynamically adjust the charging time and the charging pause time according to the difference between the current temperature of the power bank and the safe temperature;

[0033] Based on the real-time deviation between the safe temperature and the actual temperature of the power bank during charging, the charging current is dynamically controlled.

[0034] Preferably, the analyzing of the historical demand peak time period data of the site, setting up the peak buffer equipment in advance, and adopting dynamic hierarchical charging management for the buffer equipment specifically includes:

[0035] Obtain the historical peak demand time period of the site, set up peak buffer equipment in the overlapping time interval of all historical peak time periods of the site according to the demand fluctuation range, and fully charge the buffer equipment and put it in standby mode;

[0036] A tiered charging plan is adopted for buffer devices, which is divided into three tiers: fast charging tier, standard charging tier and standby tier;

[0037] Set the corresponding relationship between the remaining power gradient interval of the buffer device and the charging plan layer, where the first gradient interval corresponds to the fast charging layer, the second gradient interval corresponds to the standard charging layer, and the third gradient interval corresponds to the standby layer;

[0038] The buffer devices whose remaining power is in the corresponding gradient when returned are placed in the corresponding charging layer, and the real-time power of the buffer devices is monitored, and the charging layer to which the buffer devices belong is dynamically adjusted based on the real-time power;

[0039] Set the demand fluctuation assessment time interval, evaluate the demand fluctuation amplitude in the previous time interval at a fixed frequency, and shut down all buffer devices when the fluctuation amplitude tends to be stable and the current time exceeds the peak demand period.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] Through intelligent analysis, a personalized charging strategy is formed to optimize the rapid rotation of power banks and the allocation of charging resources. At the same time, in order to deal with temperature problems in high-frequency usage modes, the system should have the ability to monitor heat in real time and dynamically adjust charging strategies to ensure equipment safety and extend battery life. In addition, the system needs to analyze the historical peak demand data of the site, configure buffer devices in advance, and manage these devices dynamically in layers to ensure that user needs can be quickly responded to during peak hours and reduce resource waste during non-peak hours, thereby achieving efficient, accurate and intelligent charging management. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flow chart of a centralized management method for charging power banks based on user power demand of the present invention;

[0043] Figure 2 A flowchart for analyzing the power demand pattern of users by collecting and centrally managing multi-dimensional data information of power banks from lending to returning;

[0044] Figure 3This is a flow chart of identifying the user's electricity demand pattern through multi-layer clustering analysis based on the collected user's lending frequency, time period characteristics and site characteristics of the power bank;

[0045] Figure 4 This is a flow chart of the invention that customizes the charging management strategy according to the characteristics of the user's power usage pattern label to achieve rapid rotation of power banks and targeted optimization allocation of charging resources;

[0046] Figure 5 When the user's power usage mode label is a fast turnover mode, the heating condition of each power bank is monitored in real time, and a flow chart of secondary adjustment of the charging strategy based on the heating condition is provided;

[0047] Figure 6 The present invention analyzes the historical peak demand time period data of the site, sets the peak buffer equipment in advance, and adopts a dynamic hierarchical charging management flow chart for the buffer equipment. DETAILED DESCRIPTION

[0048] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0049] Reference Figure 1 As shown, a centralized management method for charging a power bank based on the power demand of users includes:

[0050] Collect multi-dimensional data information from the lending to the return of power banks for centralized management, and analyze the user's electricity demand pattern;

[0051] Customize charging management strategies based on user power usage pattern label characteristics to achieve rapid rotation of power banks and targeted optimization of charging resources;

[0052] When the user's power usage mode label is fast turnover mode, the heating condition of each power bank is monitored in real time, and the charging strategy is adjusted again based on the heating condition;

[0053] Analyze the historical peak demand time period data of the site, set up peak buffer equipment in advance, and adopt dynamic tiered charging management for the buffer equipment.

[0054] Reference Figure 2 As shown in the figure, the multi-dimensional data information of the power bank from lending to returning for centralized management is collected, and the user's electricity demand pattern is analyzed, including:

[0055] The number of times power banks are loaned out from smart power bank cabinets or other centralized power bank management devices is counted within a set time window, and the lending frequency is calculated to judge the intensity of users' demand for power banks. The length of the time window can be set according to the needs of different scenarios. For example, the lending situation can be counted in a 7-day cycle, or high-frequency statistics can be set within 24 hours according to site requirements, providing data support for subsequent charging management strategies.

[0056] Accurately track the return of all borrowed power banks, record the actual usage time of each user's rented power bank and the time interval between the final return, and analyze the time characteristics of users returning the power bank after borrowing;

[0057] According to the GPS location information of the power bank, the flow path of the power bank between different locations is recorded, and the user's power bank usage location is divided into different venues, such as home, office, shopping mall and other places, to form the user's venue characteristics.

[0058] Through the analysis of the above multi-dimensional data, based on the collected frequency, time period characteristics and site characteristics of the power banks used by users, multi-layer clustering analysis is used to identify the users' electricity demand patterns.

[0059] Reference Figure 3 As shown in the figure, based on the collected user's lending frequency, time characteristics and site characteristics of power banks, the user's electricity demand pattern is identified through multi-layer clustering analysis, including:

[0060] After preliminary cleaning of the collected data, the first level of cluster analysis is performed on the data, using the borrowing frequency and return interval as target features to identify the initial electricity demand pattern of users;

[0061] The K-means clustering algorithm is used to perform preliminary clustering of user data based on rental frequency and return interval. In the K-means clustering, two initial cluster centers are defined, representing the "high-frequency borrowing" and "low-frequency borrowing" categories. The distance formula between the user data point and the cluster center is: ;

[0062] In the formula, is the distance between the user data point and the cluster center, and They represent the rental frequency and return interval of the i-th user respectively, and is the rental frequency and return interval of the cluster center.

[0063] By iteratively updating the cluster centers until the distance converges, two clustering results are finally output: high rental frequency and short return interval, which usually represent the demand scenario of batch borrowing and fast return; low rental frequency and long return interval, which usually represent the demand scenario of dispersed borrowing and slow return.

[0064] Based on the first-level clustering, the second-level clustering analysis is performed on the user's usage time, and the demand pattern is combined with the first-level clustering analysis results to identify the user's specific electricity demand pattern;

[0065] The Gaussian mixture model (GMM) is used to cluster the usage time. First, the usage time is standardized, and then the data is analyzed by probability density through GMM. The Gaussian distribution density function is: ;

[0066] In the formula, for The corresponding probability density value is, The user's usage time, is the mean of the Gaussian distribution, is the standard deviation.

[0067] High-frequency and low-frequency borrowing users are further divided into short-term users (usually in airports, stations and other places, with more short-term use) and long-term users (usually in hotels, libraries and other places, with more long-term use). The mode scenarios are combined into short-term high-demand scenarios and long-term low-frequency demand scenarios;

[0068] Based on the second-level clustering results, the third clustering analysis is conducted on the electricity demand patterns of different venues in combination with the users' geographical locations and time characteristics. The density-based DBSCAN algorithm is used to cluster the venue location and time data. By comparing the spatial instances and density radius between users, the specific identification results of the dynamic electricity demand of venue users are formed.

[0069] Based on the results of the three-layer clustering analysis, specific user power consumption pattern labels are generated for each venue. The labels include fast turnover mode (such as peak hours at airports, scenarios with a large number of users) and resource optimization mode (such as off-peak hours at hotels, scenarios with less user demand).

[0070] Through the implementation of multi-layer clustering analysis, the system can accurately identify the user's power consumption pattern, simplify complex user demand scenarios into manageable scenario labels, and support the dynamic adjustment of subsequent charging strategies. This method combines K-means clustering, GMM and DBSCAN, and uses mathematical formulas to accurately calculate user behavior characteristics to achieve efficient clustering analysis.

[0071] Reference Figure 4 As shown in the figure, the charging management strategy is customized according to the characteristics of the user's power consumption mode label to achieve rapid rotation of power banks and targeted optimization allocation of charging resources. Specifically, it includes:

[0072] Differentiate and analyze the user power usage pattern labels at each site and plan charging strategies;

[0073] The specific steps of distinguishing and analyzing the user power consumption mode labels of each site and planning the charging strategy are as follows:

[0074] When the user's power usage mode label is fast turnover mode, within the specification range of the charging equipment, set the fast charging mode with a current greater than the normal charging current. At the same time, according to the return order and power status of the power bank, the charging priority is assigned in real time. Assuming that the current power of the power bank is , the power state allocation priority is calculated as follows: ;

[0075] In the formula, is the maximum power of the power bank, The order of return is that the earlier the power bank is returned, the higher the priority. and is a weight parameter used to adjust the impact of power and return order on priority. The result of the priority calculation.

[0076] Get all the power banks that are being charged, set the target power of the power banks according to the current user charging demand (labels corresponding to the three-layer cluster analysis results), monitor the power banks that are being charged in real time, and when there is a power bank whose real-time power reaches the target power, open the rental restrictions of the power bank and put it into use again;

[0077] When the user's power usage mode label is resource optimization mode, the returned power bank is charged slowly with low current (lower than the normal charging current charging mode), and the minimum power of the stage charging target is set to full power;

[0078] By testing the battery parameters, the health of the returned power bank is evaluated. The health of the battery parameters can be obtained by weighting and comprehensively operating them. The charging priority of the power banks to be charged is sorted based on the health evaluation results.

[0079] Based on the distribution of electricity costs and historical usage demand, off-peak charging is arranged through electricity price forecasts, and charging tasks are scheduled during periods when electricity prices are lower.

[0080] Reference Figure 5 As shown in the figure, when the user's power usage mode label is the fast turnover mode, the heating condition of each power bank is monitored in real time, and the charging strategy is adjusted secondary based on the heating condition, including:

[0081] A temperature sensing device is installed inside or on the surface of each power bank to monitor the temperature changes during the charging process in real time, and the collected temperature data is processed to remove noise. The sliding average method is used to remove noise and filter out the impact of short-term temperature fluctuations.

[0082] Obtain the safety specification standard temperature and the highest acceptable charging temperature of each power bank, calculate the pulse charging cycle based on the adaptive algorithm of temperature feedback, and dynamically adjust the charging time and the charging suspension time according to the difference between the current temperature of the power bank and the safe temperature. Specifically:

[0083] Set charging time and pause time The adaptive adjustment formula dynamically adjusts the charging cycle according to the difference between the real-time temperature of the power bank and the safe temperature. The adaptive calculation formula for the pulse charging time is as follows:

[0084] ;

[0085] In the formula, They are the maximum continuous charging time, the minimum and maximum charging pause time. They are the real-time temperature, safety temperature and maximum chargeable temperature of the power bank. The following is a specific calculation example:

[0086] Assuming that the safe charging temperature of the power bank is 40 degrees Celsius, the maximum charging temperature is 45 degrees Celsius, the real-time temperature is 43 degrees Celsius, the maximum continuous charging time is 10 minutes, and the minimum and maximum pause times are 2 minutes and 5 minutes, then the charging time and pause time calculation process is: ,

[0087] In this case, the charging time and the pause time are and .

[0088] During intermittent charging, the system will continuously monitor the temperature change curve and determine whether the device needs a longer pause time based on the temperature data. If the device temperature continues to rise, the system will extend the pause time to ensure that the device cools down before continuing to charge until the power bank is within a safe temperature during the charging process.

[0089] Based on the real-time deviation between the safe temperature and the actual temperature of the power bank during charging, the charging current is dynamically controlled and the formula for dynamically adjusting the charging current is as follows: ;

[0090] In the formula, is the adjusted charging current, is the maximum charging current.

[0091] By dynamically adjusting the charging current, the system can reduce the charging current when the temperature is high, thereby reducing the risk of heating and ensuring the safety of the charging process.

[0092] Reference Figure 6As shown, the historical peak demand time period data of the site is analyzed, peak buffer equipment is set in advance, and dynamic hierarchical charging management is adopted for the buffer equipment, including:

[0093] Obtain the historical peak demand time period of the venue (i.e., the user demand density for power banks is greater than the set density), and set up peak buffer devices in the overlapping time interval of all historical peak time periods of the venue according to the demand fluctuation range. The buffer devices are only open for use during the peak demand period. The buffer devices are fully charged and in standby mode. Specifically:

[0094] The demand fluctuation range is defined as , ;

[0095] In the formula, is the loan amount at each time t in the historical data, The time interval coincides with the historical peak period, according to the fluctuation range of demand And the product of the initial buffer device quantity is set to determine the final arrangement buffer device quantity.

[0096] A tiered charging plan is adopted for buffer devices, which is divided into three tiers: fast charging tier, standard charging tier and standby tier;

[0097] Set the corresponding relationship between the remaining power gradient interval of the buffer device and the charging plan layer, where the first gradient interval corresponds to the fast charging layer, the second gradient interval corresponds to the standard charging layer, and the third gradient interval corresponds to the standby layer;

[0098] The buffer devices whose remaining power is in the corresponding gradient when returned are placed in the corresponding charging layer, and the real-time power of the buffer devices is monitored, and the charging layer to which the buffer devices belong is dynamically adjusted based on the real-time power;

[0099] Set the demand fluctuation assessment time interval, and assess the demand fluctuation amplitude in the previous time interval at a fixed frequency. The value of is acceptable) and the current time exceeds the peak demand period ( ), close all buffer devices.

[0100] The above implementation steps can effectively identify and analyze the overlap intervals of historical demand peak time periods, and calculate the number of peak buffer devices based on the demand fluctuation amplitude and buffer coefficient. This solution ensures sufficient supply of power banks during peak periods, while reducing resource waste during non-peak periods, thereby improving the service quality and resource utilization efficiency of charging stations.

[0101] Furthermore, the present solution also proposes a storage medium for a centralized management method for charging power banks based on user electricity demand, on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned centralized management method for charging power banks based on user electricity demand is executed.

[0102] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0103] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A centralized management method for charging power banks based on user power demand, characterized in that: include: Collect multi-dimensional data information from the lending to the return of power banks for centralized management, and analyze the user's electricity demand pattern; Customize charging management strategies based on user power usage pattern label characteristics to achieve rapid rotation of power banks and targeted optimization of charging resources; When the user's power usage mode label is fast turnover mode, the heating condition of each power bank is monitored in real time, and the charging strategy is adjusted again based on the heating condition; Analyze the historical peak demand time period data of the site, set up peak buffer equipment in advance, and adopt dynamic hierarchical charging management for the buffer equipment; The method of customizing the charging management strategy according to the characteristics of the user's power usage pattern label to achieve rapid rotation of power banks and targeted optimization allocation of charging resources specifically includes: Differentiate and analyze the user power usage pattern labels at each site and plan charging strategies; The specific steps of distinguishing and analyzing the user power consumption mode labels and planning the charging strategy for each site are as follows: When the user's power usage mode label is fast turnover mode, within the specification range of the charging equipment, a fast charging mode with a current greater than the normal charging current is set, and charging priority is assigned in real time according to the return order and power status of the power bank; Obtain all power banks that are being charged, set the target power of the power banks according to the current user charging demand, monitor the power banks that are being charged in real time, and when there is a power bank whose real-time power reaches the target power, remove the rental restrictions on the power bank and put it back into use; When the user's power usage mode label is resource optimization mode, the returned power bank is charged slowly with low current, and the minimum power of the stage charging target is set to full power; The returned power banks are evaluated for health by testing battery parameters, and the charging priorities of the power banks to be charged are sorted based on the health evaluation results; According to the electricity cost and the distribution of historical usage demand, peak-shift charging is arranged through electricity price forecast, and charging tasks are scheduled during periods with lower electricity prices; When the user's power usage mode label is a fast turnover mode, real-time monitoring of the heating condition of each power bank and secondary adjustment of the charging strategy based on the heating condition specifically include: A temperature sensing device is installed inside or on the surface of each power bank to monitor the temperature changes during charging in real time, remove noise from the collected temperature data, and filter out the impact of short-term temperature fluctuations; Obtain the safety specification standard temperature and the highest acceptable charging temperature of each power bank, calculate the pulse charging cycle based on the adaptive algorithm of temperature feedback, and dynamically adjust the charging time and the charging pause time according to the difference between the current temperature of the power bank and the safe temperature; Based on the real-time deviation between the safe temperature and the actual temperature of the power bank during charging, the charging current is dynamically controlled.

2. According to claim 1, a centralized management method for charging power banks based on user power demand, characterized in that: The multi-dimensional data information of the power bank collected and centrally managed from the time of lending to the time of returning and analyzing the user's electricity demand pattern specifically includes: Count the number of times power banks are lent out within a set time window, and calculate the lending frequency to judge the intensity of users' demand for power banks; Accurately track the return of all borrowed power banks, record the actual usage time of each user's rented power bank and the time interval between the final return, and analyze the time characteristics of users returning the power bank after borrowing; According to the GPS location information of the power bank, the flow path of the power bank between different locations is recorded to form the user's site characteristics; Based on the collected data on the lending frequency, time period characteristics and site features of power banks used by users, users' electricity demand patterns are identified through multi-layer clustering analysis.

3. A centralized management method for charging power banks based on user power demand according to claim 2, characterized in that: The frequency, time period and venue characteristics of the power bank used by users are collected. Identifying the user's electricity demand pattern through multi-layer cluster analysis specifically includes: After the collected data is preliminarily cleaned, the first-level cluster analysis is performed on the data, using the borrowing frequency and return interval as the target features to identify the user's initial electricity demand pattern, and the initial electricity demand pattern is divided into two categories: high-frequency borrowing and low-frequency borrowing; Based on the first-level clustering, the second-level clustering analysis is performed on the user's usage time, and the demand pattern is combined with the first-level clustering analysis results to identify the user's specific electricity demand pattern; High-frequency and low-frequency borrowing users are further subdivided into short-term and long-term users, and the mode scenarios are combined into short-term high-demand scenarios and long-term low-frequency demand scenarios; Based on the second-level clustering results, combined with the user's geographical location and time characteristics, a third clustering analysis is conducted on the electricity demand patterns of different sites to form a specific identification result of the site user's dynamic electricity demand; Based on the results of the three-layer clustering analysis, specific user electricity usage pattern labels are generated for each site, including rapid turnover mode and resource optimization mode.

4. A centralized management method for charging power banks based on user power demand according to claim 3, characterized in that: The analysis of the historical peak demand time period data of the site, setting up peak buffer equipment in advance, and adopting dynamic hierarchical charging management for the buffer equipment specifically includes: Obtain the historical peak demand time period of the site, set up peak buffer equipment in the overlapping time interval of all historical peak time periods of the site according to the demand fluctuation range, and fully charge the buffer equipment and put it in standby mode; A tiered charging plan is adopted for buffer devices, which is divided into three tiers: fast charging tier, standard charging tier and standby tier; Set the corresponding relationship between the remaining power gradient interval of the buffer device and the charging plan layer, where the first gradient interval corresponds to the fast charging layer, the second gradient interval corresponds to the standard charging layer, and the third gradient interval corresponds to the standby layer; The buffer devices whose remaining power is in the corresponding gradient when returned are placed in the corresponding charging layer, and the real-time power of the buffer devices is monitored, and the charging layer to which the buffer devices belong is dynamically adjusted based on the real-time power; Set the demand fluctuation assessment time interval, evaluate the demand fluctuation amplitude in the previous time interval at a fixed frequency, and shut down all buffer devices when the fluctuation amplitude tends to be stable and the current time exceeds the peak demand period.

Images