Intelligent management method and system for shared power banks

Abstract

The invention discloses an intelligent management method and system for a shared power bank, relates to the field of intelligent management, and solves the problems that multi-source heterogeneous data are difficult to fuse and associate, a cross influence area is difficult to judge, demand distribution and weight calculation are difficult to carry out, and the system reliability is high. A density factor and a cross influence factor are lacked to correct a basic charging demand coefficient, and a plurality of data are lacked to construct a priority comprehensive score to determine an optimal source area; comprising the following steps: acquiring city big data and power bank real-time data; dividing a charge pal coverage area based on UTM projection and a Voronoi algorithm, and calculating a cross influence area and an RBF weight; correcting the basic charging demand coefficient in combination with the charge pal density factor and the cross influence factor; setting upper and lower limits of a scheduling threshold, and defining a target area and a source area; and a control instruction is automatically generated, the power bank is preferentially called out from the source area with the highest comprehensive score, and resource optimization configuration is realized.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent management, and specifically relates to a method and system for intelligent management of shared power banks. Background Art

[0002] With the booming development of the sharing economy, shared power banks have become an essential emergency charging tool for people when traveling. However, the management of shared power banks still faces many challenges. For example, in densely populated areas such as business districts and transportation hubs, there is often a shortage of power banks or the power banks are full and cannot be returned. In some non-peak areas, power banks may be idle in large numbers.

[0003] Traditional management methods are mostly based on historical experience or simple real-time borrowing and returning data for scheduling. It is difficult to accurately respond to dynamically changing demands, difficult to integrate and correlate multi-source heterogeneous data, difficult to determine cross-impact areas, and to perform demand allocation and weight calculations. There is a lack of setting power bank density factors and cross-impact factors to correct the basic charging demand coefficient, and there is also a lack of comprehensive scoring based on a variety of data to determine the optimal source area.

[0004] Therefore, how to use urban big data and advanced sensor technology to achieve accurate prediction and intelligent scheduling of shared power bank demand has become an urgent problem to be solved. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a method and system for intelligent management of shared power banks, which are used to solve the following technical problems:

[0006] Traditional management methods are mostly based on historical experience or simple real-time borrowing and returning data for scheduling. It is difficult to accurately respond to dynamically changing demands, difficult to integrate and correlate multi-source heterogeneous data, difficult to determine cross-impact areas, and to perform demand allocation and weight calculations. There is a lack of setting power bank density factors and cross-impact factors to correct the basic charging demand coefficient, and there is also a lack of comprehensive scoring based on a variety of data to determine the optimal source area.

[0007] To solve the above problems, the first aspect of the present invention provides a method for intelligent management of shared power banks, comprising the following steps:

[0008] S1: Collaborate with various departments and institutions to obtain urban big data. At the same time, built-in GPS and Bluetooth sensors in shared power banks transmit location and status data to the cloud in real time and link it with large city data.

[0009] S2: Convert the longitude and latitude coordinates of the shared power bank cabinets into plane coordinates using UTM projection, generate Voronoi polygon coverage areas, and calculate the distance between cabinets to determine whether there is a cross-influence area. If so, calculate the weight of each cabinet using RBF, normalize it, and distribute it proportionally.

[0010] S3: Based on the Voronoi polygons and RBF weights, the basic power bank demand coefficient of each shared power bank cabinet is calculated, the power bank density factor and cross-influence factor are defined, the basic power bank demand coefficient is corrected to obtain the final power bank demand index, and normalization is performed;

[0011] S4: Draw a charging demand heat map based on the normalized final power bank demand index, set upper and lower limits for scheduling, formulate a power bank scheduling strategy, and define the target area and source area;

[0012] S5: Based on the calculated number of incoming and outgoing transfers, management instructions are automatically generated and sent to the intelligent control system. A priority comprehensive score is calculated for each source area to determine the optimal source area, and transfers are prioritized from the optimal source area.

[0013] Preferably, the step S1 comprises the following steps:

[0014] Collaborate with city management departments, commercial organizations, and transportation departments to obtain urban big data, including business district foot traffic data, event schedule data, and peak hour data at transportation hubs;

[0015] Each shared power bank is equipped with built-in GPS and Bluetooth sensors and a data transmission module;

[0016] The GPS sensor is used to obtain the geographical location information of the power bank in real time, and the Bluetooth sensor is used to detect the connection status between the power bank and the user's mobile phone;

[0017] At the same time, the power bank status data is collected, including power level, quantity and health status, and transmitted to the cloud server once a minute via the mobile network;

[0018] Associate business district traffic data, event schedule data, and peak-hour data at transportation hubs with the latitude, longitude, and status data of power banks according to timestamps and geographic coordinates to establish a unified data format and data warehouse.

[0019] Preferably, the step S2 comprises the following steps:

[0020] Get the longitude and latitude coordinates of all shared power bank cabinets in the city. Assuming there are N shared power bank cabinets, use UTM projection to convert the longitude and latitude coordinates into a plane coordinate system and get (x1, y1), (x2, y2), ... (xN ,y N );

[0021] Use the Voronoi algorithm to generate Voronoi polygons for each shared power bank cabinet. Each polygon represents the coverage area of a shared power bank cabinet, and all points in the area are closest to the cabinet.

[0022] At the same time, the distance between each pair of shared power bank cabinets is calculated. If the distance is less than 800 meters, it is considered that the two cabinets have a cross-influence area;

[0023] The cross-influence area is the intersection of the Voronoi polygons of the two cabinets;

[0024] In the cross-influence area, the weight of each cabinet is calculated by RBF. The specific RBF calculation is:

[0025]

[0026] in, is any point in the cross-influence area, is the coordinate of cabinet i, is the bandwidth parameter of RBF, which controls the weight decay speed;

[0027] For each pair of shared power bank cabinets, calculate the weight of each point in the cross-influence area and , and normalize it to make sure the sum is 1, the normalized weight and Respectively represent the allocation ratio of each cabinet in the cross-influence area;

[0028] In the cross-influence area, the demand is distributed to the main areas of the two cabinets according to the normalized weight.

[0029] Preferably, the step S3 includes the following steps:

[0030] By collecting pedestrian flow data, activity schedule data, and peak-hour data at transportation hubs in each covered area, we calculated pedestrian flow, activity density, and traffic flow. We assigned weights based on actual demand and added them together to obtain the basic power bank demand coefficient, specifically:

[0031]

[0032] in, is the basic power bank demand coefficient, 、 and are the flow of people, activity density and traffic flow, 、 and are the corresponding weight coefficients respectively;

[0033] Define and calculate the power bank density factor and cross-influence factor;

[0034] The final power bank demand coefficient is obtained by multiplying the basic power bank demand coefficient with the power bank density factor and the cross-influence factor, and then correcting it. Specifically:

[0035]

[0036] in, is the final power bank demand coefficient, is the power bank density factor, is the cross-influence factor;

[0037] Normalize the final power bank demand coefficient of all shared power bank cabinets to obtain .

[0038] Preferably, the definition and calculation of the power bank density factor includes the following steps:

[0039] The ideal density of power banks is set based on historical data, the actual density of power banks in the current shared power bank cabinets is calculated, and the average density of power banks in the neighborhood is introduced;

[0040] The average density of power banks in the neighborhood is obtained by calculating the average density of power banks in the surrounding cabinets;

[0041] The power bank density factor is obtained by calculation, specifically:

[0042]

[0043] in, is the power bank density factor, The ideal density for power banks. is the average density of power banks in the neighborhood, is the actual density of the power bank, To adjust the influence weight of the neighborhood average density, Is a positive number.

[0044] Preferably, the cross-influence factor comprises the following steps:

[0045] Assume that The weight of the influence of adjacent cabinets on the current cabinet is ,in, To control the decay rate, for distance;

[0046] At the same time, the weight satisfy:

[0047]

[0048] The total impact is obtained by adding up the impact weights of all adjacent cabinets. The cross-influence factor is obtained by adjusting the speed of the cross-influence, specifically:

[0049]

[0050] in, This is the basic impact of the current cabinet itself. is the cross-influence factor, To adjust the intensity of cross-influence, is the speed of distance attenuation, For the The weight of the influence of adjacent cabinets on the current cabinet, For the current cabinet to the The distance between adjacent cabinets.

[0051] Preferably, the step S4 comprises the following steps:

[0052] According to the normalized final power bank demand coefficient , using GIS technology to draw a heat map of charging demand;

[0053] On the charging demand heat map, different colors represent different charging demand levels. The darker the color, the higher the charging demand.

[0054] Set power bank dispatch thresholds and formulate power bank dispatch strategies based on the charging demand heat map and actual operation conditions;

[0055] The power bank scheduling strategy: if a certain area If the charging demand in a certain area is higher than the set upper threshold, it means that the charging demand in that area is too high, and it is necessary to dispatch power banks from other areas to supplement the area. If the number is lower than or equal to the set lower threshold, it means that there are too many power banks in the area, and some power banks in the area will be dispatched to other areas where the demand for power banks is too high;

[0056] The system automatically filters out The areas above the upper threshold are regarded as target areas where power banks need to be transferred in. At the same time, The areas with scores lower than or equal to the lower threshold and higher than the priority comprehensive score are used as the source areas for calling out power banks.

[0057] Preferably, the step S5 comprises the following steps:

[0058] For each target area, calculate the number of power banks that need to be transferred in, specifically:

[0059]

[0060] in, The number of power banks that need to be transferred in. is the normalized final power bank demand coefficient, is the upper threshold, For the capacity of shared power bank cabinet, is the first scheduling coefficient;

[0061] For each source area, calculate the number of power banks that can be called out, specifically:

[0062]

[0063] in, The number of power banks that can be called out. is the lower threshold, is the normalized final power bank demand coefficient, The number of power banks currently in the cabinet. is the second scheduling coefficient;

[0064] According to the calculated and , automatically generates management instructions and sends the control instructions to the intelligent control system, where the management instructions include input instructions and output instructions.

[0065] Preferably, the determining of the optimal source region comprises the following steps:

[0066] Obtain the plane coordinates of the source area, the status data of the power bank, and the transportation cost from the source area to the target area. At the same time, obtain the plane coordinates of the target area and the number of power banks that need to be transferred in.

[0067] A priority composite score is calculated for each source region, specifically:

[0068]

[0069] in, For priority comprehensive scoring, To calculate the distance from the center of the source area to the center of the target area through geographic coordinates, is the average power of all power banks in the source area, is the average value of the health status of the power bank. If the normal state is assigned a value of 1, the fault state is assigned a value of 0. For transportation costs, The number of power banks that can be called out. 、 、 、 and is the corresponding weight coefficient;

[0070] The priority comprehensive scores calculated for each source area are sorted, and the area with the highest priority comprehensive score is selected as the optimal source area, which is used as the transfer-out area in the management instruction, and the power bank is transferred out first.

[0071] A second aspect of the present invention provides an intelligent management system for shared power banks, comprising the following modules: a data acquisition and fusion module: which collaborates with various departments and institutions to acquire city big data. At the same time, GPS and Bluetooth sensors are built into the shared power banks to transmit location and status data to the cloud in real time and associate it with large city data;

[0072] Spatial division and cross-influence zone calculation module: The longitude and latitude coordinates of shared power bank cabinets are converted into plane coordinates through UTM projection, Voronoi polygon coverage is generated, and the distance between cabinets is calculated to determine whether there is a cross-influence zone. If within the cross-influence zone, the weight of each cabinet is calculated using RBF, and the weight is distributed proportionally after normalization.

[0073] Charging demand coefficient calculation module: Based on Voronoi polygons and RBF weights, the basic charging treasure demand coefficient of each shared charging treasure cabinet is calculated, the charging treasure density factor and cross-influence factor are defined, the basic charging treasure demand coefficient is corrected to obtain the final charging treasure demand index, and normalization is performed;

[0074] Charging demand heat map and scheduling decision-making module: Draws a charging demand heat map based on the normalized final power bank demand index, sets upper and lower scheduling thresholds, formulates power bank scheduling strategies, and defines target and source areas;

[0075] Intelligent control and optimal source area selection module: Based on the calculated input and output quantities, management instructions are automatically generated and control instructions are sent to the intelligent control system. The priority comprehensive score is calculated for each source area to determine the optimal source area, and priority is given to outputting from the optimal source area.

[0076] Beneficial effects of the present invention:

[0077] The present invention achieves deep fusion of multi-source data by acquiring urban big data and combining it with the location and status data transmitted in real time by the built-in GPS and Bluetooth sensors of shared power banks. It also converts the latitude and longitude coordinates of shared power bank cabinets into plane coordinates, generates Voronoi polygon coverage areas, and accurately divides the service range of each cabinet, avoiding the ambiguity of traditional geographical divisions. By calculating the distance between cabinets and determining the cross-impact area, the problem of overlapping services of multiple cabinets is solved, thereby improving the scientific nature of resource allocation.

[0078] The present invention calculates the weight of each cabinet by radial basis function in the cross-influence area, distributes power banks in proportion after normalization, adapts to real-time demand changes in different areas, and improves scheduling flexibility. On the basis of the basic power bank demand coefficient, the power bank density factor and cross-influence factor are introduced to correct the demand coefficient to obtain the final power bank demand index. The normalization process eliminates the dimensionality effect, makes the demand index of different areas comparable, and provides a unified standard for the formulation of scheduling strategies.

[0079] The present invention draws a charging demand heat map through the normalized final power bank demand index, intuitively showing the charging demand distribution in different areas, and automatically generates management instructions by calculating the number of incoming and outgoing charges and sending them to the intelligent control system, thereby realizing the automation and intelligence of the scheduling process, reducing manual intervention, improving scheduling efficiency, and reducing the risk of operational errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 Schematic diagram of the method flow of the present invention;

[0081] Figure 2 It is a schematic diagram of the module flow of the present invention. DETAILED DESCRIPTION

[0082] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0083] See also Figure 1 As shown, the present invention is a method for intelligent management of shared power banks, comprising the following steps:

[0084] S1: Collaborate with various departments and institutions to obtain urban big data. At the same time, built-in GPS and Bluetooth sensors in shared power banks transmit location and status data to the cloud in real time and link it with large city data.

[0085] S2: Convert the longitude and latitude coordinates of the shared power bank cabinets into plane coordinates using UTM projection, generate Voronoi polygon coverage areas, and calculate the distance between cabinets to determine whether there is a cross-influence area. If so, calculate the weight of each cabinet using RBF, normalize it, and distribute it proportionally.

[0086] S3: Based on the Voronoi polygons and RBF weights, the basic power bank demand coefficient of each shared power bank cabinet is calculated, the power bank density factor and cross-influence factor are defined, the basic power bank demand coefficient is corrected to obtain the final power bank demand index, and normalization is performed;

[0087] S4: Draw a charging demand heat map based on the normalized final power bank demand index, set upper and lower limits for scheduling, formulate a power bank scheduling strategy, and define the target area and source area;

[0088] S5: Based on the calculated number of incoming and outgoing transfers, management instructions are automatically generated and sent to the intelligent control system. A priority comprehensive score is calculated for each source area to determine the optimal source area, and transfers are prioritized from the optimal source area.

[0089] In one embodiment of the present invention, step S1 includes the following steps:

[0090] Collaborate with city management departments, commercial organizations, and transportation departments to obtain urban big data, including business district foot traffic data, event schedule data, and peak hour data at transportation hubs;

[0091] Each shared power bank is equipped with built-in GPS and Bluetooth sensors and a data transmission module;

[0092] The GPS sensor is used to obtain the geographical location information of the power bank in real time, and the Bluetooth sensor is used to detect the connection status between the power bank and the user's mobile phone;

[0093] At the same time, the power bank status data is collected, including power level, quantity and health status, and transmitted to the cloud server once a minute via the mobile network;

[0094] Associate business district traffic data, event schedule data, and peak-hour data at transportation hubs with the latitude, longitude, and status data of power banks according to timestamps and geographic coordinates to establish a unified data format and data warehouse.

[0095] Specifically, the business district pedestrian flow data obtains information on the number of people in different time periods and different areas through passenger flow detection systems in various places; the event schedule data is obtained by collecting information on the time, location and expected number of participants of various activities; the transportation hub peak period data obtains information on peak passenger flow periods from the operating departments of various transportation hubs; the power bank data (latitude and longitude, status data) transmitted to the cloud is associated with the city big data (business district pedestrian flow, event schedule, transportation hub peak period data) according to timestamps and geographic coordinates. For example, a power bank is located in a business district at 10:00 Beijing time, and the pedestrian flow data of the business district at this time is 1,000 people / hour; a unified data format is defined to ensure that data from different sources can be compatible and matched, and a data warehouse is established to store the associated data in the same database for subsequent analysis and application.

[0096] In one embodiment of the present invention, step S2 includes the following steps:

[0097] Get the longitude and latitude coordinates of all shared power bank cabinets in the city. Assuming there are N shared power bank cabinets, use UTM projection to convert the longitude and latitude coordinates into a plane coordinate system and get (x1, y1), (x2, y2), ... (x N ,y N );

[0098] Use the Voronoi algorithm to generate Voronoi polygons for each shared power bank cabinet. Each polygon represents the coverage area of a shared power bank cabinet, and all points in the area are closest to the cabinet.

[0099] At the same time, the distance between each pair of shared power bank cabinets is calculated. If the distance is less than 800 meters, it is considered that the two cabinets have a cross-influence area;

[0100] The cross-influence area is the intersection of the Voronoi polygons of the two cabinets;

[0101] In the cross-influence area, the weight of each cabinet is calculated by RBF. The specific RBF calculation is:

[0102]

[0103] in, is any point in the cross-influence area, is the coordinate of cabinet i, is the bandwidth parameter of RBF, which controls the weight decay speed;

[0104] For each pair of shared power bank cabinets, calculate the weight of each point in the cross-influence area and , and normalize it to make sure the sum is 1, the normalized weight and Respectively represent the allocation ratio of each cabinet in the cross-influence area;

[0105] In the cross-influence area, the demand is distributed to the main areas of the two cabinets according to the normalized weight.

[0106] Specifically, the latitude and longitude coordinates of all shared power bank cabinets in the city are obtained through GPS positioning, and the latitude and longitude coordinates are converted into the UTM plane coordinate system using the UTM projection tool to obtain (x1, y1), (x2, y2), ... (x N ,y N ); Put all the plane coordinates of the cabinet ((x1, y1), (x2, y2), ... (x N ,y N ) is input into the Voronoi algorithm, and the Voronoi polygons of each cabinet are generated using the Voronoi algorithm. Each polygon represents the coverage area of a cabinet, and all points in the area are closest to the cabinet. The plane distance between each pair of cabinets is calculated. If the distance between a pair of cabinets is less than 800 meters, it is considered that there is a cross-influence area between the two cabinets. Among them, 800 meters is estimated based on the walking time of user needs. The cross-influence area is the intersection of the Voronoi polygons of the two cabinets. In the cross-influence area, the RBF weights of cabinet i and cabinet j are calculated for each point (x, y) to obtain and , normalize the weight of each point to ensure and The sum is 1, and the demand in the cross-impact area is distributed to the main areas of the two cabinets according to the normalized weight.

[0107] In one embodiment of the present invention, step S3 includes the following steps:

[0108] By collecting pedestrian flow data, activity schedule data, and peak-hour data at transportation hubs in each covered area, we calculated pedestrian flow, activity density, and traffic flow. We assigned weights based on actual demand and added them together to obtain the basic power bank demand coefficient, specifically:

[0109]

[0110] in, is the basic power bank demand coefficient, 、 and are the flow of people, activity density and traffic flow, 、 and are the corresponding weight coefficients respectively;

[0111] Define and calculate the power bank density factor and cross-influence factor;

[0112] The final power bank demand coefficient is obtained by multiplying the basic power bank demand coefficient with the power bank density factor and the cross-influence factor, and then correcting it. Specifically:

[0113]

[0114] in, is the final power bank demand coefficient, is the power bank density factor, is the cross-influence factor;

[0115] Normalize the final power bank demand coefficient of all shared power bank cabinets to obtain .

[0116] Specifically, we divide a day into multiple time periods, count the number of people in each time period, divide the corresponding coverage area into 50m*50m grids, and count the number of people in each grid, specifically: ,in, For time period Inner Grid Traffic flow, is the total number of time periods, is the total number of grids; the activity density is specifically: ,in, is the number of activities in the corresponding area, For activities The number of participants, is the geographical area of the coverage area; the traffic volume is specifically: ,in, is the number of transportation hubs, For site The number of people entering and leaving the station; assigning weights based on actual needs includes determining the importance of the three factors based on business goals, and at the same time, analyzing historical data to find the factors most relevant to charging needs, and obtaining 、 and The weights are 0.5, 0.3 and 0.2 respectively.

[0117] In one embodiment of the present invention, the step of defining and calculating the power bank density factor includes the following steps:

[0118] The ideal density of power banks is set based on historical data, the actual density of power banks in the current shared power bank cabinets is calculated, and the average density of power banks in the neighborhood is introduced;

[0119] The average density of power banks in the neighborhood is obtained by calculating the average density of power banks in the surrounding cabinets;

[0120] The power bank density factor is obtained by calculation, specifically:

[0121]

[0122] in, is the power bank density factor, The ideal density for power banks. is the average density of power banks in the neighborhood, is the actual density of the power bank, To adjust the influence weight of the neighborhood average density, Specifically, the purpose of defining the power bank density factor is to quantify the impact of power bank density on demand. At the same time, the neighborhood average density is introduced to reflect the distribution of power banks in the surrounding cabinets. By adjusting the influence weight of the neighborhood average density through parameters, we can more comprehensively evaluate whether the current cabinet density is reasonable. A very small positive value is added to the denominator, set to 0.0001, to avoid division by zero errors. Among them, the neighborhood average density is specifically: ,in, is the number of cabinets in the neighborhood, For the The density of power banks in the neighborhood cabinets; parameters The parameter used to adjust the influence weight of the neighborhood average density is It is usually a non-negative number. The specific value depends on the actual scenario: if the neighborhood density has little impact on the current cabinet, then ≥1, otherwise, 0≤ <1; when the actual density of the power bank is close to the ideal density of the power bank, Close to 1; when the actual density of the power bank deviates from the ideal density of the power bank, It will increase or decrease, reflecting the irrationality of density.

[0123] In one embodiment of the present invention, the cross-influence factor comprises the following steps:

[0124] Assume that The weight of the influence of adjacent cabinets on the current cabinet is ,in, To control the decay rate, for distance;

[0125] At the same time, the weight satisfy:

[0126]

[0127] The total impact is obtained by adding up the impact weights of all adjacent cabinets. The cross-influence factor is obtained by adjusting the speed of the cross-influence, specifically:

[0128]

[0129] in, This is the basic impact of the current cabinet itself. is the cross-influence factor, To adjust the intensity of cross-influence, is the speed of distance attenuation, For the The weight of the influence of adjacent cabinets on the current cabinet, For the current cabinet to the The distance between adjacent cabinets.

[0130] Specifically, in actual scenarios, the coverage areas of multiple cabinets may overlap. The cross-impact factor is defined to quantify this cross-impact and incorporate it into demand calculations, thereby allocating resources more reasonably. When two cabinets are close to each other, their scheduling decisions will affect each other. For example, the dispatch of a power bank at one cabinet may significantly affect the available resources of another cabinet. Therefore, the closer the two cabinets are, the stronger the cross-impact; the farther the distance, the weaker the cross-impact. Parameter Used to adjust the intensity of cross-influence, The larger the value, the more significant the cross-effect. Used to control the speed of distance attenuation. The larger the value, the faster the distance decays; introducing the cross-influence factor into the basic power bank demand coefficient can more accurately reflect the competitive relationship between cabinets and the rationality of resource allocation; using the exponential function Simulate the attenuation of the influence weight due to distance. If =1, it means that the influence of the cabinet itself is the greatest; if it approaches 0, it means that the influence of the remote cabinet approaches 0;

[0131] is the distance between cabinets, calculated using the Euclidean distance, specifically:

[0132]

[0133] in, is the distance between the two cabinets, ( , )and( , ) are the coordinates of the two cabinets respectively;

[0134] Among them, the parameters The value of is adjusted according to the sensitivity of the cross-impact in the actual scenario. If the cross-impact has a greater impact on demand allocation, =1; if the cross-effect is small, take =0.5; parameter The value of is adjusted according to the cabinet distribution density and the impact range. If the cabinet distance is large (> 1000 meters), a smaller value is used. , for example 0.1; if the cabinet distance is small (less than 500 meters), take the larger value , for example 0.5.

[0135] In one embodiment of the present invention, step S4 includes the following steps:

[0136] According to the normalized final power bank demand coefficient , using GIS technology to draw a heat map of charging demand;

[0137] On the charging demand heat map, different colors represent different charging demand levels. The darker the color, the higher the charging demand.

[0138] Set power bank dispatch thresholds and formulate power bank dispatch strategies based on the charging demand heat map and actual operation conditions;

[0139] The power bank scheduling strategy: if a certain area If the charging demand in a certain area is higher than the set upper threshold, it means that the charging demand in that area is too high, and it is necessary to dispatch power banks from other areas to supplement the area. If the number is lower than or equal to the set lower threshold, it means that there are too many power banks in the area, and some power banks in the area will be dispatched to other areas where the demand for power banks is too high;

[0140] The system automatically filters out The areas above the upper threshold are regarded as target areas where power banks need to be transferred in. At the same time, The areas with scores lower than or equal to the lower threshold and higher than the priority comprehensive score are used as the source areas for calling out power banks.

[0141] Specifically, according to The values are divided into multiple levels and the corresponding colors are defined: low demand: light green; medium demand: yellow; high demand: orange; extremely high demand: red; use ArcGIS, QGIS or Python geographic visualization library to generate heat maps to show the cabinet location and Import into GIS software and use kernel density estimation function to Convert it into heat distribution and fill different areas with colors according to the demand level and color mapping rules;

[0142] Among them, the upper and lower thresholds are found by analyzing historical data. The critical point of the imbalance between supply and demand of power banks is based on the mean and standard deviation, dynamically adjusted in combination with business objectives, and the threshold is flexibly set according to different regions, time periods and business needs.

[0143] In one embodiment of the present invention, step S5 includes the following steps:

[0144] For each target area, calculate the number of power banks that need to be transferred in, specifically:

[0145]

[0146] in, The number of power banks that need to be transferred in. is the normalized final power bank demand coefficient, is the upper threshold, For the capacity of shared power bank cabinet, is the first scheduling coefficient;

[0147] For each source area, calculate the number of power banks that can be called out, specifically:

[0148]

[0149] in, The number of power banks that can be called out. is the lower threshold, is the normalized final power bank demand coefficient, The number of power banks currently in the cabinet. is the second scheduling coefficient;

[0150] According to the calculated and , automatically generates management instructions and sends the control instructions to the intelligent control system, where the management instructions include input instructions and output instructions.

[0151] Specifically, The first scheduling coefficient is used to control the sensitivity of the number of transfers, and the value is usually between 0.5 and 1; It is the second scheduling coefficient, which is used to control the sensitivity of the number of power banks transferred out, and its value is usually 0.5 to 1. According to the transfer-in demand and transfer-out capacity, the optimal source area is selected first, and the generated instruction content includes transfer-in instructions and transfer-out instructions. The transfer-in instructions include: target area, transfer-in quantity, transfer-out time and scheduling time; the transfer-out instructions include: source area, transfer-out quantity, transfer-in area and scheduling time. For example, the transfer-in instruction is to transfer 10 power banks from source area A to target area B, and the scheduling time is 15:00. The transfer-out instruction is to transfer 10 power banks from source area A to target area B, and the scheduling time is 15:00. The instructions are sent to the intelligent control system through the API interface or message queue, and the system automatically executes the scheduling operation and updates the cabinet status.

[0152] In one embodiment of the present invention, determining the optimal source region comprises the following steps:

[0153] Obtain the plane coordinates of the source area, the status data of the power bank, and the transportation cost from the source area to the target area. At the same time, obtain the plane coordinates of the target area and the number of power banks that need to be transferred in.

[0154] A priority composite score is calculated for each source region, specifically:

[0155]

[0156] in, For priority comprehensive scoring, To calculate the distance from the center of the source area to the center of the target area through geographic coordinates, is the average power of all power banks in the source area, is the average value of the health status of the power bank. If the normal state is assigned a value of 1, the fault state is assigned a value of 0. For transportation costs, The number of power banks that can be called out. 、 、 、 and is the corresponding weight coefficient;

[0157] The priority comprehensive scores calculated for each source area are sorted, and the area with the highest priority comprehensive score is selected as the optimal source area, which is used as the transfer-out area in the management instruction, and the power bank is transferred out first.

[0158] Specifically, the total transportation cost is calculated based on the distance from the source area to the target area, the mode of transportation including labor, vehicle, etc., and the unit transportation cost. The transportation cost can be a fixed value or a dynamic value (charged by distance); the number of power banks that can be dispatched is calculated based on the status of the power banks in the source area and the dispatching rules; the distance between the source area and the target area is calculated using the Euclidean distance; the power levels of all normal power banks in the source area are averaged to obtain the average power level of all power banks in the source area; the health status of all power banks in the source area is averaged to obtain the average health status; wherein, the corresponding weight coefficient 、 、 、 and The values of are 0.3, 0.2, 0.2, 0.1 and 0.2 respectively.

[0159] See also Figure 2 As shown, the present invention is a method for intelligent management of shared power banks, including the following modules:

[0160] Data acquisition and fusion module: Collaborate with various departments and institutions to obtain urban big data. At the same time, built-in GPS and Bluetooth sensors in shared power banks transmit location and status data to the cloud in real time and associate it with large city data;

[0161] Spatial division and cross-influence zone calculation module: The longitude and latitude coordinates of shared power bank cabinets are converted into plane coordinates through UTM projection, Voronoi polygon coverage is generated, and the distance between cabinets is calculated to determine whether there is a cross-influence zone. If within the cross-influence zone, the weight of each cabinet is calculated using RBF, and the weight is distributed proportionally after normalization.

[0162] Charging demand coefficient calculation module: Based on Voronoi polygons and RBF weights, the basic charging treasure demand coefficient of each shared charging treasure cabinet is calculated, the charging treasure density factor and cross-influence factor are defined, the basic charging treasure demand coefficient is corrected to obtain the final charging treasure demand index, and normalization is performed;

[0163] Charging demand heat map and scheduling decision-making module: Draws a charging demand heat map based on the normalized final power bank demand index, sets upper and lower scheduling thresholds, formulates power bank scheduling strategies, and defines target and source areas;

[0164] Intelligent control and optimal source area selection module: Based on the calculated input and output quantities, management instructions are automatically generated and control instructions are sent to the intelligent control system. The priority comprehensive score is calculated for each source area to determine the optimal source area, and priority is given to outputting from the optimal source area.

[0165] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for intelligent management of shared power banks, characterized in that: The following steps are involved: S1: Collaborate with various departments and institutions to obtain urban big data. At the same time, built-in GPS and Bluetooth sensors in shared power banks transmit location and status data to the cloud in real time and link it with large city data. S2: Convert the longitude and latitude coordinates of the shared power bank cabinets into plane coordinates using UTM projection, generate Voronoi polygon coverage areas, and calculate the distance between cabinets to determine whether there is a cross-influence area. If so, calculate the weight of each cabinet using RBF, normalize it, and distribute it proportionally. S3: Based on the Voronoi polygons and RBF weights, the basic power bank demand coefficient of each shared power bank cabinet is calculated, the power bank density factor and cross-influence factor are defined, the basic power bank demand coefficient is corrected to obtain the final power bank demand index, and normalization is performed; S4: Draw a charging demand heat map based on the normalized final power bank demand index, set upper and lower limits for scheduling, formulate a power bank scheduling strategy, and define the target area and source area; S5: Based on the calculated number of incoming and outgoing transfers, management instructions are automatically generated and sent to the intelligent control system. A priority comprehensive score is calculated for each source area to determine the optimal source area, and transfers are prioritized from the optimal source area.

2. A method for intelligent management of shared power banks according to claim 1, characterized in that: The step S1 comprises the following steps: Collaborate with city management departments, commercial organizations, and transportation departments to obtain urban big data, including business district foot traffic data, event schedule data, and peak hour data at transportation hubs; Each shared power bank is equipped with built-in GPS and Bluetooth sensors and a data transmission module; The GPS sensor is used to obtain the geographical location information of the power bank in real time, and the Bluetooth sensor is used to detect the connection status between the power bank and the user's mobile phone; At the same time, the power bank status data is collected, including power level, quantity and health status, and transmitted to the cloud server once a minute via the mobile network; Associate business district traffic data, event schedule data, and peak-hour data at transportation hubs with the latitude, longitude, and status data of power banks according to timestamps and geographic coordinates to establish a unified data format and data warehouse.

3. The intelligent management method for shared power banks according to claim 1, characterized in that: The step S2 comprises the following steps: Get the longitude and latitude coordinates of all shared power bank cabinets in the city. Assuming there are N shared power bank cabinets, use UTM projection to convert the longitude and latitude coordinates into a plane coordinate system and get (x1, y1), (x2, y2), ... (x N ,y N ); Use the Voronoi algorithm to generate Voronoi polygons for each shared power bank cabinet. Each polygon represents the coverage area of a shared power bank cabinet, and all points in the area are closest to the cabinet. At the same time, the distance between each pair of shared power bank cabinets is calculated. If the distance is less than 800 meters, it is considered that the two cabinets have a cross-influence area; The cross-influence area is the intersection of the Voronoi polygons of the two cabinets; In the cross-influence area, the weight of each cabinet is calculated by RBF. The specific RBF calculation is: in, is any point in the cross-influence area, is the coordinate of cabinet i, is the bandwidth parameter of RBF, which controls the weight decay speed; For each pair of shared power bank cabinets, calculate the weight of each point in the cross-influence area and , and normalize it to make sure the sum is 1, the normalized weight and Respectively represent the allocation ratio of each cabinet in the cross-influence area; In the cross-influence area, the demand is distributed to the main areas of the two cabinets according to the normalized weight.

4. The intelligent management method for shared power banks according to claim 1, characterized in that: The step S3 comprises the following steps: By collecting pedestrian flow data, activity schedule data, and peak-hour data at transportation hubs in each covered area, we calculated pedestrian flow, activity density, and traffic flow. We assigned weights based on actual demand and added them together to obtain the basic power bank demand coefficient, specifically: in, is the basic power bank demand coefficient, 、 and are the flow of people, activity density and traffic flow, 、 and are the corresponding weight coefficients respectively; Define and calculate the power bank density factor and cross-influence factor; The final power bank demand coefficient is obtained by multiplying the basic power bank demand coefficient with the power bank density factor and the cross-influence factor, and then correcting it. Specifically: in, is the final power bank demand coefficient, is the power bank density factor, is the cross-influence factor; Normalize the final power bank demand coefficient of all shared power bank cabinets to obtain .

5. The intelligent management method for shared power banks according to claim 4, characterized in that: The definition and calculation of the power bank density factor includes the following steps: The ideal density of power banks is set based on historical data, the actual density of power banks in the current shared power bank cabinets is calculated, and the average density of power banks in the neighborhood is introduced; The average density of power banks in the neighborhood is obtained by calculating the average density of power banks in the surrounding cabinets; The power bank density factor is obtained by calculation, specifically: in, is the power bank density factor, The ideal density for power banks. is the average density of power banks in the neighborhood, is the actual density of the power bank, To adjust the influence weight of the neighborhood average density, Is a positive number.

6. A method for intelligent management of shared power banks according to claim 4, characterized in that: The cross-influence factor comprises the following steps: Assume that The weight of the influence of adjacent cabinets on the current cabinet is ,in, To control the decay rate, for distance; At the same time, the weight satisfy: The total impact is obtained by adding up the impact weights of all adjacent cabinets. The cross-influence factor is obtained by adjusting the speed of the cross-influence, specifically: in, This is the basic impact of the current cabinet itself. is the cross-influence factor, To adjust the intensity of cross-influence, is the speed of distance attenuation, For the The weight of the influence of adjacent cabinets on the current cabinet, For the current cabinet to the The distance between adjacent cabinets.

7. The intelligent management method for shared power banks according to claim 1, characterized in that: The step S4 comprises the following steps: According to the normalized final power bank demand coefficient , using GIS technology to draw a heat map of charging demand; On the charging demand heat map, different colors represent different charging demand levels. The darker the color, the higher the charging demand. Set power bank dispatch thresholds and formulate power bank dispatch strategies based on the charging demand heat map and actual operation conditions; The power bank scheduling strategy: if a certain area If the charging demand in a certain area is higher than the set upper threshold, it means that the charging demand in that area is too high, and it is necessary to dispatch power banks from other areas to supplement the area. If the number is lower than or equal to the set lower threshold, it means that there are too many power banks in the area, and some power banks in the area will be dispatched to other areas where the demand for power banks is too high; The system automatically filters out The areas above the upper threshold are regarded as target areas where power banks need to be transferred in. At the same time, The areas with scores lower than or equal to the lower threshold and higher than the priority comprehensive score are used as the source areas for calling out power banks.

8. The intelligent management method for shared power banks according to claim 1, characterized in that: The step S5 comprises the following steps: For each target area, calculate the number of power banks that need to be transferred in, specifically: in, The number of power banks that need to be transferred in. is the normalized final power bank demand coefficient, is the upper threshold, For the capacity of shared power bank cabinet, is the first scheduling coefficient; For each source area, calculate the number of power banks that can be called out, specifically: in, The number of power banks that can be called out. is the lower threshold, is the normalized final power bank demand coefficient, The number of power banks currently in the cabinet. is the second scheduling coefficient; According to the calculated and , automatically generates management instructions and sends the control instructions to the intelligent control system, where the management instructions include input instructions and output instructions.

9. The intelligent management method for shared power banks according to claim 1, characterized in that: Determining the optimal source region comprises the following steps: Obtain the plane coordinates of the source area, the status data of the power bank, and the transportation cost from the source area to the target area. At the same time, obtain the plane coordinates of the target area and the number of power banks that need to be transferred in. A priority composite score is calculated for each source region, specifically: in, For priority comprehensive scoring, To calculate the distance from the center of the source area to the center of the target area through geographic coordinates, is the average power of all power banks in the source area, is the average value of the health status of the power bank. If the normal state is assigned a value of 1, the fault state is assigned a value of 0. For transportation costs, The number of power banks that can be called out. 、 、 、 and is the corresponding weight coefficient; The priority comprehensive scores calculated for each source area are sorted, and the area with the highest priority comprehensive score is selected as the optimal source area, which is used as the transfer-out area in the management instruction, and the power bank is transferred out first.

10. An intelligent management system for shared power banks, characterized in that: Includes the following modules: Data acquisition and fusion module: Collaborate with various departments and institutions to obtain urban big data. At the same time, built-in GPS and Bluetooth sensors in shared power banks transmit location and status data to the cloud in real time and associate it with large city data; Spatial division and cross-influence zone calculation module: The longitude and latitude coordinates of shared power bank cabinets are converted into plane coordinates through UTM projection, Voronoi polygon coverage is generated, and the distance between cabinets is calculated to determine whether there is a cross-influence zone. If within the cross-influence zone, the weight of each cabinet is calculated using RBF, and the weight is distributed proportionally after normalization. Charging demand coefficient calculation module: Based on Voronoi polygons and RBF weights, the basic charging treasure demand coefficient of each shared charging treasure cabinet is calculated, the charging treasure density factor and cross-influence factor are defined, the basic charging treasure demand coefficient is corrected to obtain the final charging treasure demand index, and normalization is performed; Charging demand heat map and scheduling decision-making module: Draws a charging demand heat map based on the normalized final power bank demand index, sets upper and lower scheduling thresholds, formulates power bank scheduling strategies, and defines target and source areas; Intelligent control and optimal source area selection module: Based on the calculated input and output quantities, management instructions are automatically generated and control instructions are sent to the intelligent control system. The priority comprehensive score is calculated for each source area to determine the optimal source area, and priority is given to outputting from the optimal source area.

Images