Rental management method based on shared power bank with anti-theft mechanism

By analyzing the location and historical data of the shared power bank cabinet, establishing a loan quantity model, and adjusting the loan quantity and rental price based on real-time data, the loss of shared power bank cabinets is solved, reducing operational losses and improving management efficiency.

CN119049179AInactive Publication Date: 2024-11-29SHENZHEN BEST INTERNET OF THINGS TECH CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411217076.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Due to unmanned self-service operations, shared power bank cabinets often suffer losses caused by users forgetting to return or maliciously encroachment, causing property losses to the operator. How to implement an intelligent leasing management method of anti-theft mechanism through risk assessment and data prediction.

Method used

By obtaining and analyzing the location information, historical lending data and people flow of shared power bank cabinets, establishing an optimal expected lending number model, adjusting the number of power banks that can be borrowed in real time, and assessing abnormality based on the discharge time, idle time and location changes. Risk of encroachment and dynamically adjust rental prices.

Benefits of technology

Effectively prevent theft losses caused by excessive borrowing of power banks, and reduce the losses caused by abnormal encroachment through dynamic leasing price adjustments, and improve operational efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049179A_ABST
    Figure CN119049179A_ABST
Patent Text Reader

Abstract

The invention discloses a lease management method based on a shared power bank with an anti-theft mechanism, and relates to the field of power bank lease management, and the method comprises the steps: obtaining and analyzing the position information of an operating shared power bank cabinet, and evaluating the theft risk of the shared power bank in the position cabinet; obtaining the borrowing quantity of the power banks in each time period of the historical sample day, and fitting and establishing an optimal expected borrowing quantity model of the power banks in combination with the expected theft risk index curve of the historical sample day; according to the real-time data condition of the power bank cabinet, the current optimal expected lending quantity is calculated and adjusted, and the server controls the cabinet to limit the quantity of the power banks capable of being lended in real time; analyzing the discharge duration, idle duration and position change information of the borrowed charge pal, and dynamically evaluating the risk that the charge pal is abnormally infringed and occupied; according to the abnormal invasion and occupation risk conditions of all the borrowed power banks, the lease price of each borrowed power bank is dynamically and accurately adjusted in real time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of power bank rental management, and in particular to a rental management method based on a shared power bank with an anti-theft mechanism. Background Art

[0002] Shared power banks refer to charging rental devices provided by enterprises. Users can rent a power bank by scanning the QR code on the device screen with their mobile devices and paying a deposit. After the power bank is successfully returned, the deposit can be withdrawn and returned to the account at any time. Since shared power bank cabinets are generally unmanned and self-service, shared power banks in unsupervised areas are often lost, including losses caused by users forgetting to return them after use and users maliciously occupying them, resulting in property losses to the operator.

[0003] Therefore, how to implement an intelligent rental management method with an anti-theft mechanism through risk assessment and data prediction is an urgent problem that needs to be solved. Summary of the invention

[0004] In order to solve the above technical problems, a rental management method for a shared power bank with an anti-theft mechanism 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 rental management method based on a shared power bank with an anti-theft mechanism, comprising:

[0007] Obtain and analyze the location information of shared power bank cabinets in operation, and assess the risk of theft of shared power banks in the cabinets at that location;

[0008] Obtain the number of power banks loaned out in each period of the historical sample day, analyze it in combination with the expected theft risk index curve of the historical sample day, and fit and establish a model for the optimal expected number of power banks loaned out;

[0009] According to the real-time data of the power bank cabinet, the server calculates and adjusts the current optimal expected lending quantity, and controls the cabinet in real time to limit the number of power banks that can be lent out;

[0010] Analyze the discharge time, idle time and location change information of the borrowed power bank, and dynamically assess the risk of the power bank being occupied abnormally;

[0011] According to the abnormal occupation risk of all borrowed power banks, the rental price of each borrowed power bank is adjusted dynamically and accurately in real time.

[0012] Preferably, the obtaining and analyzing the location information of the shared power bank cabinet in operation and evaluating the theft risk of the shared power bank in the cabinet at the location specifically includes:

[0013] Obtain all areas that meet the standards for the operation and deployment of shared power banks, classify the areas according to their usage, and set the risk weight for each area;

[0014] Obtain the locations of shared power bank cabinets in operation, and collect data on the regional usage characteristics and monitoring facility coverage area of ​​each cabinet location;

[0015] Set a fixed number of days N, select N days before the current date and mark them as historical sample days, set a time scale, divide a single day into time points with equal intervals according to the set time scale, mark them as monitoring time points, and collect the flow of people at each monitoring time point on the historical sample day;

[0016] According to the regional risk weight, the coverage area of ​​monitoring facilities and the historical traffic data collected at the monitoring time points on the sample day, the expected theft risk index of the power bank is calculated comprehensively. The calculation formula for the expected theft risk index is:

[0017]

[0018] Where R k is the expected theft risk index at the kth monitoring time point, N is the set number of days, M ki is the flow of people collected at the kth monitoring time point on the i-th sample day, W A is to set the risk weight of the area, S m is the monitoring coverage area of ​​the region, and S is the total area of ​​the region;

[0019] Obtain the expected theft risk index at each monitoring time point and establish an expected theft risk index curve.

[0020] Preferably, obtaining the number of power banks loaned out in each period of the historical sample day, analyzing the expected theft risk index curve of the historical sample day, and fitting and establishing a model for the optimal expected number of power banks loaned out specifically includes:

[0021] Select the first N historical sample days, calculate the expected theft risk index for each monitoring time point corresponding to all historical sample days according to the expected theft risk index calculation formula, and establish the expected theft risk index curve set A for the historical sample days;

[0022] Obtain the number of power banks loaned out at each monitoring time point on the historical sample day, and establish a curve set B of the number of loaned out on the historical sample day;

[0023] The curve sets A and B are vertically divided according to the monitoring time points to obtain the expected theft risk index set A corresponding to the monitoring time points of the historical sample days. kt and the loan quantity set B Kt ;

[0024] Expected theft risk index set A kt =[a t1 …a td …a tN ], where A kt is the expected theft risk index set at the t-th monitoring time point on the historical sample day, a td is the expected theft risk index at the t-th monitoring time point corresponding to the d-th historical sample day;

[0025] Loan quantity set B kt =[b t1 …b td …b tN ], where B kt is the set of loan quantities at the tth monitoring time point on the historical sample day, b td is the lending quantity at the tth monitoring time point corresponding to the dth historical sample day;

[0026] For the expected theft risk index set A kt and the loan quantity set B kt Perform comprehensive analysis to define the kernel function:

[0027]

[0028] In the formula, k(a,a i ) is the kernel function, a is the expected theft risk index at any set time point predicted on the day, and a i is the expected theft risk index of the same monitoring time point as a on the i-th sample day, σ 2 is the variance of the monitoring time point data of all sample days;

[0029] According to the kernel function, the kernel matrix of all sample day monitoring time points is established:

[0030] where k ij = k(a i ,a j ), m and n are the number of monitoring time points on all sample days;

[0031] Establish the covariance vector K of the expected theft risk index at the monitoring time point set on the day and the expected theft risk index at the monitoring time point set on the sample day C , K C =[k(a,a1)…k(a,a i )…k(a,a n )] T ;

[0032] According to the expected theft risk index at the sample monitoring time point of the day, combined with the kernel function fitting, the optimal expected loan quantity calculation formula is established. The optimal expected loan quantity calculation formula is:

[0033] B pi =K C T K -1 B ki ;

[0034] In the formula, B ki is the sample set vector of the number of loans at the i-th monitoring time point, B pi is the prediction result of the optimal lending quantity at the i-th monitoring time point on the day;

[0035] The variance of the prediction results is calculated to evaluate the uncertainty of the prediction results. The formula for calculating the variance of the prediction results is:

[0036] σ C 2 =-K C T K -1 K C +1;

[0037] In the formula, σ C 2 is the variance of the prediction result, when σ C 2 The larger it is, the greater the uncertainty of the prediction result is, and vice versa.

[0038] Preferably, the calculation and adjustment of the current optimal expected lending quantity according to the real-time data of the power bank cabinet, and the server controlling the cabinet in real time to limit the number of power banks that can be lent specifically include:

[0039] Obtain the expected theft risk index curve of the day and the expected theft risk index corresponding to each monitoring time point on the curve, calculate the expected loan quantity data corresponding to each monitoring time point using the optimal expected loan quantity calculation formula, and send it to the server;

[0040] After receiving the information, the server takes the current day as a separate sample and calculates the real-time theft risk index at the current monitoring time point using the expected theft risk index calculation formula. According to the deviation between the expected theft risk index at the current monitoring time point and the real-time theft risk index, the optimal expected loan quantity of the current cabinet at the monitoring time point is adjusted in the same proportion;

[0041] The number of power banks that the cabinet is allowed to lend is automatically planned and adjusted based on the adjusted optimal expected lending quantity. When the cabinet detects that the current lending quantity has reached the maximum allowed lending quantity, the subsequent rental of power banks is restricted.

[0042] Preferably, the analyzing the discharge time, idle time and location change information of the lent power bank and dynamically evaluating the risk of the power bank being abnormally occupied specifically includes:

[0043] Obtain the information of the power banks that have been borrowed from the cabinet, analyze each borrowed power bank information separately, obtain the discharge time, idle time and location change information of the power bank, and calculate the abnormal occupation risk index of the power bank by the ratio of the discharge time to the idle time of the power bank and the distance between the current borrowed power bank and the borrowing cabinet. The calculation formula of the abnormal occupation risk index is:

[0044]

[0045] In the formula, R e is the risk index of abnormal occupation of power bank, T E is the discharge time of the power bank, T L is the idle time of the power bank, and D is the distance between the power bank and the lending cabinet.

[0046] Preferably, the real-time dynamic and accurate adjustment of the rental price of each lent power bank according to the abnormal occupation risk of all lent power banks specifically includes:

[0047] Set a threshold value for the abnormal occupation risk index of the power bank in each cabinet, filter out the power banks whose abnormal occupation risk index is greater than the threshold value, and calculate the deviation ratio between the normal occupation risk index of these power banks and the threshold value of the abnormal occupation risk index;

[0048] The rental price of each lent power bank is obtained, and the server dynamically adjusts the rental price of the power bank in real time according to the deviation ratio of the abnormal risk index.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The theft risk of power banks in cabinets in each region is evaluated through historical sample data. A comprehensive analysis and fitting of the theft risk and loan quantity of the samples is performed to establish a model for the optimal expected loan quantity of power banks to calculate the optimal number of power banks loaned in the region. Based on the optimal number, the loan quantity of cabinets is remotely controlled through the server to prevent property losses caused by theft due to excessive loan of power banks. The discharge time, idle time, and location information of the loaned power banks are analyzed to evaluate the risk of abnormal occupation of the power banks. The user's rental price is dynamically adjusted according to the risk of abnormal occupation to compensate for the losses caused by abnormal occupation. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1This is a flow chart of a rental management method based on a shared power bank with an anti-theft mechanism according to the present invention;

[0052] Figure 2 A flow chart of obtaining and analyzing the location information of a shared power bank cabinet in operation and evaluating the risk of theft of a shared power bank in the cabinet at the location of the present invention;

[0053] Figure 3 The present invention obtains the number of power banks loaned out in each period of the historical sample day, analyzes the expected theft risk index curve of the historical sample day, and fits to establish a flow chart of the optimal expected loan quantity model of power banks;

[0054] Figure 4 The flow chart of the present invention is that according to the real-time data of the power bank cabinet, the current optimal expected lending quantity is calculated and adjusted, and the service end controls the cabinet in real time to limit the number of power banks that can be lent;

[0055] Figure 5 A flow chart for analyzing the discharge time, idle time and location change information of a borrowed power bank and dynamically evaluating the risk of the power bank being occupied abnormally;

[0056] Figure 6 The present invention provides a flow chart for dynamically and accurately adjusting the rental price of each lent out power bank in real time according to the abnormal encroachment risk of all lent out power banks. DETAILED DESCRIPTION

[0057] 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.

[0058] Reference Figure 1 As shown, a rental management method based on a shared power bank with an anti-theft mechanism includes:

[0059] Obtain and analyze the location information of shared power bank cabinets in operation, and assess the risk of theft of shared power banks in the cabinets at that location;

[0060] Obtain the number of power banks loaned out in each period of the historical sample day, analyze it in combination with the expected theft risk index curve of the historical sample day, and fit and establish a model for the optimal expected number of power banks loaned out;

[0061] According to the real-time data of the power bank cabinet, the server calculates and adjusts the current optimal expected lending quantity, and controls the cabinet in real time to limit the number of power banks that can be lent out;

[0062] Analyze the discharge time, idle time and location change information of the borrowed power bank, and dynamically assess the risk of the power bank being occupied abnormally;

[0063] According to the abnormal occupation risk of all borrowed power banks, the rental price of each borrowed power bank is adjusted dynamically and accurately in real time.

[0064] Reference Figure 2 As shown in the figure, the location information of the shared power bank cabinets in operation is obtained and analyzed, and the risk of the shared power bank theft in the cabinets at that location is assessed, including:

[0065] Obtain all areas that meet the standards for the operation and deployment of shared power banks, classify the areas according to their usage, and set the risk weight for each area;

[0066] Obtain the locations of shared power bank cabinets in operation, and collect data on the regional usage characteristics and monitoring facility coverage area of ​​each cabinet location;

[0067] Set a fixed number of days N, select N days before the current date and mark them as historical sample days, set a time scale, divide a single day into time points with equal intervals according to the set time scale, mark them as monitoring time points, and collect the flow of people at each monitoring time point on the historical sample day;

[0068] According to the regional risk weight, the coverage area of ​​monitoring facilities and the historical traffic data collected at the monitoring time points on the sample day, the expected theft risk index of the power bank is calculated comprehensively. The calculation formula for the expected theft risk index is:

[0069]

[0070] Where R k is the expected theft risk index at the kth monitoring time point, N is the set number of days, M ki is the flow of people collected at the kth monitoring time point on the i-th sample day, W A is to set the risk weight of the area, S m is the monitoring coverage area of ​​the region, and S is the total area of ​​the region;

[0071] Obtain the expected theft risk index at each monitoring time point and establish an expected theft risk index curve;

[0072] The risk of shared power banks being stolen is highly correlated with the location of the cabinet. In places with large traffic and wide surveillance areas, such as shopping malls, stations, supermarkets, etc., the risk of shared power banks being stolen is low, while in places with small traffic and narrow surveillance areas, such as parking lots, unmanned convenience stores, etc., the risk of shared power banks being stolen is high. In addition, the risk of power banks being stolen increases during special periods, such as at night, and in special areas, such as scenic spots, where the passenger flow is large, the loss of shared power banks due to tourists getting lost and unable to return them or forgetting to return them after use also needs to be taken into consideration. The accuracy of risk assessment is increased by considering multiple factors.

[0073] Reference Figure 3 As shown in the figure, the number of power banks loaned out in each period of the historical sample day is obtained, and the expected theft risk index curve of the historical sample day is analyzed to fit and establish the optimal expected loan quantity model of power banks. Specifically, the following are included:

[0074] Select the first N historical sample days, calculate the expected theft risk index for each monitoring time point corresponding to all historical sample days according to the expected theft risk index calculation formula, and establish the expected theft risk index curve set A for the historical sample days;

[0075] Obtain the number of power banks loaned out at each monitoring time point on the historical sample day, and establish a curve set B of the number of loaned out on the historical sample day;

[0076] The curve sets A and B are vertically divided according to the monitoring time points to obtain the expected theft risk index set A corresponding to the monitoring time points of the historical sample days. kt and the loan quantity set B Kt ;

[0077] Expected theft risk index set A kt =[a t1 …a td …a tN ], where A kt is the expected theft risk index set at the t-th monitoring time point on the historical sample day, a td is the expected theft risk index at the t-th monitoring time point corresponding to the d-th historical sample day;

[0078] Loan quantity set B kt =[b t1 …b td …b tN ], where B kt is the set of loan quantities at the tth monitoring time point on the historical sample day, b td is the lending quantity at the tth monitoring time point corresponding to the dth historical sample day;

[0079] For the expected theft risk index set A kt and the loan quantity set B kt Perform comprehensive analysis to define the kernel function:

[0080]

[0081] In the formula, k(a,a i ) is the kernel function, a is the expected theft risk index at any set time point predicted on the day, and a i is the expected theft risk index of the same monitoring time point as a on the i-th sample day, σ 2 is the variance of the monitoring time point data of all sample days;

[0082] According to the kernel function, the kernel matrix of all sample day monitoring time points is established:

[0083] where k ij = k(a i ,a j ), m and n are the number of monitoring time points on all sample days;

[0084] Establish the covariance vector K of the expected theft risk index at the monitoring time point set on the day and the expected theft risk index at the monitoring time point set on the sample day C , K C =[k(a,a1)…k(a,a i )…k(a,a n )] T ;

[0085] According to the expected theft risk index at the sample monitoring time point of the day, combined with the kernel function fitting, the optimal expected loan quantity calculation formula is established. The optimal expected loan quantity calculation formula is:

[0086] B pi =K C T K -1 B ki ;

[0087] In the formula, B ki is the sample set vector of the number of loans at the i-th monitoring time point, B pi is the prediction result of the optimal lending quantity at the i-th monitoring time point on the day;

[0088] The variance of the prediction results is calculated to evaluate the uncertainty of the prediction results. The formula for calculating the variance of the prediction results is:

[0089] σ C 2 =-K C TK -1 K C +1;

[0090] In the formula, σ C 2 is the variance of the prediction result, when σ C 2 The larger the value, the greater the uncertainty of the prediction result; otherwise, the smaller the uncertainty of the prediction result.

[0091] The kernel function is used to measure the similarity between the longitudinal data points at each time point in the curve set. The model is processed by monitoring all sample data at the time point, and the optimal expected lending quantity at each time point is predicted according to the expected theft risk index of the day. The accuracy of the prediction result is positively correlated with the size of the sample data, and the reliability of the prediction result can be evaluated by the variance of the prediction result. The purpose of predicting the optimal lending quantity is to obtain the expected optimal lending saturation quantity considering the flow of people in each place under the expected theft risk. Exceeding this number will increase the uncontrollability of the risk of theft in lending management.

[0092] Reference Figure 4 As shown, according to the real-time data of the power bank cabinet, the current optimal expected lending quantity is calculated and adjusted. The server controls the cabinet in real time to limit the number of power banks that can be lent out. Specifically, the following are included:

[0093] Obtain the expected theft risk index curve of the day and the expected theft risk index corresponding to each monitoring time point on the curve, calculate the expected loan quantity data corresponding to each monitoring time point using the optimal expected loan quantity calculation formula, and send it to the server;

[0094] After receiving the information, the server takes the current day as a separate sample and calculates the real-time theft risk index at the current monitoring time point using the expected theft risk index calculation formula. According to the deviation between the expected theft risk index at the current monitoring time point and the real-time theft risk index, the optimal expected loan quantity of the current cabinet at the monitoring time point is adjusted in the same proportion;

[0095] Automatically plan and adjust the number of power banks that the cabinet allows to lend out based on the adjusted optimal expected lending quantity. When the cabinet detects that the current lending quantity has reached the maximum allowed lending quantity, it limits the subsequent rental of power banks.

[0096] The optimal expected loan quantity prediction model will be affected by the amount of sample data and other conditional factors. Therefore, when the variance of the prediction results is too large, it is necessary to correct the prediction results based on the current actual situation, calculate the current actual theft risk level and the predicted risk deviation, and adjust the year-on-year deviation of the prediction results to make the corrected prediction results more in line with the actual usage scenario.

[0097] Reference Figure 5 As shown in the figure, the discharge time, idle time and location change information of the borrowed power bank are analyzed, and the risk of abnormal occupation of the power bank is dynamically assessed, including:

[0098] Obtain the information of the power banks that have been borrowed from the cabinet, analyze each borrowed power bank information separately, obtain the discharge time, idle time and location change information of the power bank, and calculate the abnormal occupation risk index of the power bank by the ratio of the discharge time to the idle time of the power bank and the distance between the current borrowed power bank and the borrowing cabinet. The calculation formula of the abnormal occupation risk index is:

[0099]

[0100] In the formula, R e is the risk index of abnormal occupation of power bank, T E is the discharge time of the power bank, T L is the idle time of the power bank, and D is the distance between the power bank and the lending cabinet;

[0101] The discharge and idle time of the power bank is an intuitive reference data for measuring the normal use of users. When the discharge time is long, it means that the power bank is more likely to be used normally, while a long idle time means that the power bank has been used up or has not been used normally and has not been returned for a long time. In addition, the farther the power bank is from the lending cabinet, the higher the expected possibility that the user will not return the power bank. By comprehensively analyzing the discharge and idle time of the power bank and the distance from the lending cabinet, the risk of the power bank being abnormally occupied can be assessed.

[0102] Reference Figure 6 As shown in the figure, according to the abnormal occupation risk of all lent power banks, the rental price of each lent power bank is adjusted dynamically and accurately in real time, including:

[0103] Set a threshold value for the abnormal occupation risk index of the power bank in each cabinet, filter out the power banks whose abnormal occupation risk index is greater than the threshold value, and calculate the deviation ratio between the normal occupation risk index of these power banks and the threshold value of the abnormal occupation risk index;

[0104] Obtain the rental price of each lent power bank, and adjust the rental price of the power bank dynamically in real time according to the deviation ratio of the abnormal risk index;

[0105] The rental price is adjusted in real time according to the current risk of abnormal encroachment. Raising the rental price will help increase the cost of malicious encroachment for users and reduce the risk of loss of power banks. The threshold of the abnormal encroachment risk index is set by operating technicians based on the size of the abnormal encroachment risk index and other actual conditions such as the cost, loss and actual theft of different power banks. It will not be repeated here.

[0106] Furthermore, the present solution also proposes a storage medium for a rental management method of a shared power bank with an anti-theft mechanism, on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned rental management method of a shared power bank with an anti-theft mechanism is executed.

[0107] 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).

[0108] 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 rental management method based on a shared power bank with an anti-theft mechanism, characterized in that: include: Obtain and analyze the location information of shared power bank cabinets in operation, and assess the risk of theft of shared power banks in the cabinets at that location; Obtain the number of power banks loaned out in each period of the historical sample day, analyze it in combination with the expected theft risk index curve of the historical sample day, and fit and establish a model for the optimal expected number of power banks loaned out; According to the real-time data of the power bank cabinet, the server calculates and adjusts the current optimal expected lending quantity, and controls the cabinet in real time to limit the number of power banks that can be lent out; Analyze the discharge time, idle time and location change information of the borrowed power bank, and dynamically assess the risk of the power bank being occupied abnormally; According to the abnormal occupation risk of all borrowed power banks, the rental price of each borrowed power bank is adjusted dynamically and accurately in real time.

2. A rental management method based on a shared power bank with an anti-theft mechanism according to claim 1, characterized in that: The obtaining and analyzing the location information of the shared power bank cabinet in operation and evaluating the theft risk of the shared power bank in the cabinet at the location specifically includes: Obtain all areas that meet the standards for the operation and deployment of shared power banks, classify the areas according to their usage, and set the risk weight for each area; Obtain the locations of shared power bank cabinets in operation, and collect data on the regional usage characteristics and monitoring facility coverage area of ​​each cabinet location; Set a fixed number of days N, select N days before the current date and mark them as historical sample days, set a time scale, divide a single day into time points with equal intervals according to the set time scale, mark them as monitoring time points, and collect the flow of people at each monitoring time point on the historical sample day; According to the regional risk weight, the coverage area of ​​monitoring facilities and the historical traffic data collected at the monitoring time points on the sample day, the expected theft risk index of the power bank is calculated comprehensively. The calculation formula for the expected theft risk index is: Where R k is the expected theft risk index at the kth monitoring time point, N is the set number of days, M ki is the flow of people collected at the kth monitoring time point on the i-th sample day, W A is to set the risk weight of the area, S m is the monitoring coverage area of ​​the region, and S is the total area of ​​the region; Obtain the expected theft risk index at each monitoring time point and establish an expected theft risk index curve.

3. A rental management method based on a shared power bank with an anti-theft mechanism according to claim 2, characterized in that: The obtaining of the number of power banks loaned out in each period of the historical sample day, analyzing the expected theft risk index curve of the historical sample day, and fitting and establishing a model for the optimal expected number of power banks loaned out specifically includes: Select the first N historical sample days, calculate the expected theft risk index for each monitoring time point corresponding to all historical sample days according to the expected theft risk index calculation formula, and establish the expected theft risk index curve set A for the historical sample days; Obtain the number of power banks loaned out at each monitoring time point on the historical sample day, and establish a curve set B of the number of loaned out on the historical sample day; The curve sets A and B are vertically divided according to the monitoring time points to obtain the expected theft risk index set A corresponding to the monitoring time points of the historical sample days. kt and the loan quantity set B Kt ; Expected theft risk index set A kt =[a t1 …a td …a tN ], where A kt is the expected theft risk index set at the t-th monitoring time point on the historical sample day, a td is the expected theft risk index at the t-th monitoring time point corresponding to the d-th historical sample day; Loan quantity set B kt =[b t1 …b td …b tN ], where B kt is the set of loan quantities at the tth monitoring time point on the historical sample day, b td is the lending quantity at the tth monitoring time point corresponding to the dth historical sample day; For the expected theft risk index set A kt and the loan quantity set B kt Perform comprehensive analysis to define the kernel function: In the formula, k(a,a i ) is the kernel function, a is the expected theft risk index at any set time point predicted on the day, and a i is the expected theft risk index of the same monitoring time point as a on the i-th sample day, σ 2 is the variance of the monitoring time point data of all sample days; According to the kernel function, the kernel matrix of all sample day monitoring time points is established: where k ij = k(a i ,a j ), m and n are the number of monitoring time points on all sample days; Establish the covariance vector K of the expected theft risk index at the monitoring time point set on the day and the expected theft risk index at the monitoring time point set on the sample day C , K C =[k(a,a1)…k(a,a i )…k(a,a n )] T ; According to the expected theft risk index at the sample monitoring time point of the day, combined with the kernel function fitting, the optimal expected loan quantity calculation formula is established. The optimal expected loan quantity calculation formula is: B pi =K C T K -1 B ki ; In the formula, B ki is the sample set vector of the number of loans at the i-th monitoring time point, B pi is the prediction result of the optimal lending quantity at the i-th monitoring time point on the day; The variance of the prediction results is calculated to evaluate the uncertainty of the prediction results. The formula for calculating the variance of the prediction results is: s C 2 =-K C T K -1 K C +1; In the formula, σ C 2 is the variance of the prediction result, when σ C 2 The larger it is, the greater the uncertainty of the prediction result is, and vice versa.

4. A rental management method based on a shared power bank with an anti-theft mechanism according to claim 3, characterized in that: The calculation and adjustment of the current optimal expected lending quantity according to the real-time data of the power bank cabinet, and the real-time control of the cabinet by the server to limit the number of power banks that can be lent specifically include: Obtain the expected theft risk index curve of the day and the expected theft risk index corresponding to each monitoring time point on the curve, calculate the expected loan quantity data corresponding to each monitoring time point using the optimal expected loan quantity calculation formula, and send it to the server; After receiving the information, the server takes the current day as a separate sample and calculates the real-time theft risk index at the current monitoring time point using the expected theft risk index calculation formula. According to the deviation between the expected theft risk index at the current monitoring time point and the real-time theft risk index, the optimal expected loan quantity of the current cabinet at the monitoring time point is adjusted in the same proportion; The number of power banks that the cabinet is allowed to lend is automatically planned and adjusted based on the adjusted optimal expected lending quantity. When the cabinet detects that the current lending quantity has reached the maximum allowed lending quantity, the subsequent rental of power banks is restricted.

5. A rental management method based on a shared power bank with an anti-theft mechanism according to claim 4, characterized in that: The analysis of the discharge time, idle time and location change information of the lent power bank and the dynamic assessment of the risk of the power bank being abnormally occupied specifically includes: Obtain the information of the power banks that have been borrowed from the cabinet, analyze each borrowed power bank information separately, obtain the discharge time, idle time and location change information of the power bank, and calculate the abnormal occupation risk index of the power bank by the ratio of the discharge time to the idle time of the power bank and the distance between the current borrowed power bank and the borrowing cabinet. The calculation formula of the abnormal occupation risk index is: In the formula, R e is the risk index of abnormal occupation of power bank, T E is the discharge time of the power bank, T L is the idle time of the power bank, and D is the distance between the power bank and the lending cabinet.

6. A rental management method based on a shared power bank with an anti-theft mechanism according to claim 5, characterized in that: According to the abnormal occupation risk of all lent power banks, the rental price of each lent power bank is adjusted dynamically and accurately in real time, specifically including: Set a threshold value for the abnormal occupation risk index of the power bank in each cabinet, filter out the power banks whose abnormal occupation risk index is greater than the threshold value, and calculate the deviation ratio between the normal occupation risk index of these power banks and the threshold value of the abnormal occupation risk index; The rental price of each lent power bank is obtained, and the server dynamically adjusts the rental price of the power bank in real time according to the deviation ratio of the abnormal risk index.

Citation Information

Cited By

  • Communication base station monitoring alarm method and system

    CN120220353A

  • A method and system for monitoring and alarming communication base stations

    CN120220353B