Shared electric vehicle headgear replenishment prediction method and system based on data analysis

Abstract

The present invention relates to the field of electric vehicle data processing technology, and in particular to a method and system for predicting the replenishment of shared electric vehicle headgear based on data analysis. The present invention obtains the daily demand for electric vehicle use based on the difference between the number of riding users and the number of shared electric vehicles at different time ranges of the day, as well as the distribution of riding start times; obtains the quarterly usage intensity of each quarter based on the distribution of electric vehicle use demand on different days in each quarter, and the correlation between the electric vehicle use demand on different days and riding power consumption; adjusts the riding power consumption and headgear usage based on the popularity of the launch area of ​​each quarter and the battery health distribution of all shared electric vehicles at different time ranges, obtains the headgear usage correction amount and the riding power correction amount; and predicts the optimal replenishment time for headgear replenishment. The present invention optimizes the operation time for headgear replenishment by analyzing the historically accurate headgear usage and riding power consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle data processing, and in particular to a method and system for predicting the replenishment of headgear for shared electric vehicles based on data analysis. Background Art

[0002] The headgear of shared electric vehicles is an important component of the vehicle, mainly used to improve the hygiene of user use. Due to the frequent services of shared electric vehicles, there may be a shortage of headgear, and some vehicles have excess headgear due to low frequency of use, resulting in waste of resources and affecting user experience; therefore, it is necessary to make appropriate adjustments to the replenishment of headgear.

[0003] In the existing technology, the headgear is replenished based on the background warning given by the quantity threshold sensor in the equipment; however, since battery replacement is one of the core tasks of shared electric vehicle operation and maintenance, only considering the replenishment of headgear leads to inconsistent vehicle operation and maintenance cycles, increases unnecessary duplication of work, and lacks effective data analysis methods, resulting in poor prediction results for headgear replenishment. Summary of the Invention

[0004] In order to solve the technical problems that only considering the replenishment of headgear leads to inconsistent vehicle operation and maintenance cycles, lacks effective data analysis methods, and has poor headgear replenishment prediction results, the purpose of the present invention is to provide a method and system for predicting the replenishment of shared electric vehicle headgear based on data analysis. The technical solutions adopted are as follows:

[0005] The present invention proposes a method for predicting the replenishment of shared electric vehicle head covers based on data analysis, the method comprising:

[0006] Obtain the number of riders in the deployment area at different times of day over the past year, as well as the riding data for each shared electric vehicle. The riding data includes the number of rides, power consumption, headgear usage, ride start time, and battery health.

[0007] Based on the difference between the number of riding users and the number of shared electric vehicles at different times of the day, as well as the distribution of riding start times, the daily demand for electric vehicles is obtained;

[0008] Based on the distribution of electric vehicle usage demand on different days in each quarter and the correlation between electric vehicle usage demand and riding electricity consumption on different days, the quarterly usage intensity of each quarter is obtained. Based on the distribution of the number of rides of all shared electric vehicles at all times in each quarter, the popularity of the deployment area in each quarter is obtained.

[0009] Based on the popularity of the launch area in each quarter, the quarterly usage intensity, and the battery health distribution of all shared electric vehicles at different times, the riding power consumption and headgear usage are adjusted to obtain the headgear usage correction amount and the riding power correction amount;

[0010] Based on the distribution of headgear usage correction and riding electricity correction of all shared electric vehicles at different time ranges, the optimal time for headgear replenishment is predicted.

[0011] Furthermore, the method for obtaining the electric vehicle usage demand includes:

[0012] Based on the difference between the number of riding users and the number of shared electric vehicles within each time range, as well as the distribution of riding start times, the electric vehicle utilization rate within each time range is obtained;

[0013] Arrange the electric vehicle usage rates at all time periods of each day in chronological order to obtain a fitting curve for fitting the electric vehicle usage rates;

[0014] Obtain the average slope value between different adjacent data points on the fitting curve as the overall slope level; for the preset peak time range, obtain the average slope value between different adjacent data points in all peak time ranges on the fitting curve as the peak slope level; obtain the average slope value between different adjacent data points in time ranges other than the peak time range as the non-peak slope level;

[0015] The ratio of the peak slope level to the off-peak slope level is obtained, and the product of the ratio result and the overall slope level is calculated as the daily electric vehicle use demand.

[0016] Furthermore, the method for obtaining the electric vehicle usage rate includes:

[0017] Obtain the difference between the number of users and the number of shared electric vehicles within each time range as the quantity difference;

[0018] Obtain the mean difference between different adjacent riding start times within each time range as the time difference level;

[0019] The product of the quantity difference and the time difference level is obtained and negatively correlated with each other to form the electric vehicle usage rate within each time range.

[0020] Furthermore, the method for obtaining the quarterly usage intensity includes:

[0021] According to the distribution of electric vehicle usage demand on different days in each quarter, the electric vehicle thermal value of each quarter is obtained;

[0022] For each quarter, the sum of the riding electricity consumption of all shared electric vehicles at all times of the day is obtained as the total daily riding electricity consumption. The mean square error of the series composed of the electric vehicle usage demand and the total riding electricity consumption on different days is obtained and negatively correlated with each other, which is used as the first trend coefficient.

[0023] The product of the electric vehicle thermal value and the first trend coefficient of the corresponding quarter is obtained as the quarterly usage intensity of each quarter.

[0024] Furthermore, the method for obtaining the thermal value of the electric vehicle includes:

[0025] Obtain the cumulative sum of electric vehicle usage demands for all days in each quarter as the first cumulative sum of demands; obtain the cumulative sum of electric vehicle usage demands for all days in a historical year as the second cumulative sum of demands;

[0026] The ratio between the first demand cumulative sum and the second demand cumulative sum is obtained as the electric vehicle thermal value for each quarter.

[0027] Furthermore, the method for obtaining the popularity of the delivery area includes:

[0028] Obtain the number of hotspots within the service area of ​​the delivery area, including business districts, subway stations, and schools.

[0029] The sum of the number of rides for each shared electric vehicle at all times of the day is obtained as the first sum; the average of the first sums of all shared electric vehicles on all days of each quarter is obtained as the average number of rides;

[0030] The product of the average number of rides and the number of hot spots in the delivery area is obtained and normalized to obtain the delivery area heat index for each quarter.

[0031] Furthermore, the method for obtaining the headgear usage correction amount and the power consumption correction amount includes:

[0032] Based on the popularity of the launch area in each quarter, the quarterly usage intensity, and the battery health distribution of all shared electric vehicles at different time ranges, the battery replacement forecast correction coefficient for each quarter is obtained;

[0033] For the headgear usage or riding electricity consumption as the data to be corrected, the battery replacement forecast correction coefficient of each quarter and the corresponding data to be corrected in each time range are multiplied together to obtain the headgear usage correction amount and riding usage correction amount in each time range.

[0034] Furthermore, the method for obtaining the battery swap prediction correction coefficient includes:

[0035] For each quarter, the average battery health of each shared electric vehicle at all times is obtained as the local battery health level; the average local battery health level of all shared electric vehicles is obtained as the overall battery health level;

[0036] The quarterly usage intensity of each quarter is normalized and mapped, and the sum of the normalized mapping result and the heat of the deployment area is obtained as the weighted value; the product of the weighted value and the overall health level of the battery is obtained as the battery replacement prediction correction coefficient for each quarter.

[0037] Furthermore, the method for obtaining the optimal replenishment time includes:

[0038] For the correction amount for headgear use or riding electricity use, the target data is obtained, and the sum of the target data of each shared electric vehicle at different times of the day is obtained as the target overall data of each shared electric vehicle on a daily basis; the average of the target overall data of all shared electric vehicles on a daily basis is obtained as the target overall average data of the day;

[0039] Construct a fitting curve for fitting the target overall average data of all days, and select the intersection of the corresponding fitting curves between the headgear usage correction amount and the riding power correction amount;

[0040] Obtain the mean slope value between all adjacent data points on the fitting curve between adjacent intersections. If the difference between the corresponding mean slope values ​​of two fitting curves between adjacent intersections is less than a preset difference threshold, the time range between the corresponding adjacent intersections is used as the optimal supplementary time for the headgear.

[0041] The present invention also proposes a shared electric vehicle head cover replenishment prediction system based on data analysis, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, it implements any one of the steps of the shared electric vehicle head cover replenishment prediction method based on data analysis.

[0042] The present invention has the following beneficial effects:

[0043] The present invention obtains the daily demand for electric vehicle usage based on the difference between the number of riding users and the number of shared electric vehicles at different time ranges of each day, as well as the distribution of riding start times, reflects the supply and demand relationship, and accurately quantifies the daily demand for electric vehicle usage; obtains the quarterly usage intensity of each quarter based on the distribution of electric vehicle usage demand on different days in each quarter, and the correlation between the electric vehicle usage demand and riding electricity consumption on different days, comprehensively evaluates the riding activity of each quarter, and identifies seasonal change trends; obtains the popularity of the deployment area of ​​each quarter based on the distribution of the number of rides of all shared electric vehicles at all time ranges in each quarter, identifies high-heat areas, and optimizes vehicle deployment and headgear replenishment; adjusts riding electricity consumption and headgear usage based on the popularity of the deployment area of ​​each quarter, the quarterly usage intensity, and the battery health distribution of all shared electric vehicles at different time ranges, obtains a headgear usage correction amount and a riding electricity correction amount, and improves the prediction accuracy of riding electricity consumption and headgear usage; predicts the optimal replenishment time for headgear replenishment based on the distribution of the headgear usage correction amount and the riding electricity correction amount of all shared electric vehicles at different time ranges. The present invention optimizes the operation time for replenishing the headgear by analyzing the accurate historical usage of the headgear and the power consumption during riding. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 A flowchart of a method for predicting the replenishment of shared electric vehicle headgear based on data analysis provided by one embodiment of the present invention;

[0046] Figure 2 A flow chart of a method for obtaining electric vehicle usage demand according to one embodiment of the present invention;

[0047] Figure 3 This is a flow chart of a method for obtaining quarterly usage intensity provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0048] To further illustrate the technical means and effects of the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a method and system for predicting the replenishment of shared electric vehicle head covers based on data analysis, including its specific implementation, structure, features, and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable form.

[0049] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0050] The following describes in detail a specific scheme of a shared electric vehicle headgear replenishment prediction method and system based on data analysis provided by the present invention in conjunction with the accompanying drawings.

[0051] See also Figure 1 , which shows a method flow chart of a method for predicting the replenishment of shared electric vehicle head covers based on data analysis provided by one embodiment of the present invention, specifically including:

[0052] Step S1: Obtain the number of riding users in the deployment area at different times of the day in the past year, as well as the riding data of each shared electric vehicle, the riding data including the number of rides, riding power consumption, headgear usage, riding start time and battery health.

[0053] In an embodiment of the present invention, in order to meet the demand for headgear replenishment and improve the operation and maintenance efficiency of electric vehicles, it is necessary to analyze the riding data of shared electric vehicles; first, each shared electric vehicle is usually equipped with a GPS positioning module and Beidou positioning, a 4G / 5G communication module, a sensor, a control chip and a battery management system. These modules upload the riding data of the shared electric vehicle to the cloud server in real time through the Internet of Things technology; obtain the number of riding users in the deployment area at different times of the day in the past year, as well as the riding data of each shared electric vehicle, the riding data including the number of rides, riding power consumption, headgear usage, riding start time and battery health.

[0054] It should be noted that, in one embodiment of the present invention, the time range is one hour; historical riding data for one year is obtained for analysis; in other embodiments of the present invention, the time range can be set by the implementer according to the specific situation, and is not limited or elaborated here.

[0055] It should be noted that in the process of riding a shared electric vehicle, effective riding data provides certain data support for the battery replacement and headgear replenishment needs of the shared electric vehicle; however, there will be some invalid data in the existing stored data, such as missing starting and ending points, too short riding distance, etc.; therefore, it is necessary to eliminate such invalid data to improve the prediction accuracy of the battery replacement and headgear replenishment needs of the shared electric vehicle: in one embodiment of the present invention, first, all riding data in the history of one year, if there are data fields in the stored riding data that are empty or missing values, the riding distance is less than 20 meters, or the recorded content is exactly the same, are regarded as invalid data and are eliminated from the acquired data set.

[0056] It should be noted that in one embodiment of the present invention, to facilitate subsequent data processing, the riding data is normalized to eliminate the dimension of the data, allowing comprehensive analysis and comparison of indicators of different units or orders of magnitude. This normalization can be performed using existing methods such as Z-score normalization and min-max transformation. The specific methods are well known to those skilled in the art and will not be detailed here.

[0057] Step S2: Obtain the daily demand for electric vehicle usage based on the difference between the number of riding users and the number of shared electric vehicles at different time ranges of the day, as well as the distribution of riding start times.

[0058] The number of cycling users at different times will be affected by many factors, such as season, weather, holidays, working hours, etc.; during rush hours, the number of cycling users may increase significantly, leading to an increase in the utilization rate of electric vehicles; the number of shared electric vehicles put into use will also affect the utilization rate. If the number of shared electric vehicles put into use is insufficient, it may lead to a shortage of electric vehicles at certain times and a high utilization rate; conversely, it may lead to a low utilization rate; the distribution of cycling start times reflects the travel habits and needs of users. If a large number of users choose to ride in the morning before work, the utilization rate of electric vehicles in the morning period may be higher; therefore, based on the difference between the number of cycling users and the number of shared electric vehicles at different times of the day, as well as the distribution of cycling start times, the daily demand for electric vehicles is obtained.

[0059] Preferably, in one embodiment of the present invention, the method for obtaining the electric vehicle usage demand degree can be found in Figure 2 , which shows a flow chart of a method for obtaining electric vehicle usage demand, including:

[0060] Step S201: Obtain the electric vehicle usage rate within each time range based on the difference between the number of riding users and the number of shared electric vehicles within each time range, as well as the distribution of riding start times.

[0061] Electric vehicle usage data directly reflects the level of actual electric vehicle use. By monitoring and analyzing electric vehicle usage at different times, we can intuitively understand the demand for electric vehicles in different time periods.

[0062] Preferably, in one embodiment of the present invention, the method for obtaining the electric vehicle usage rate includes:

[0063] Obtain the difference between the number of riding users and the number of shared electric vehicles within each time range as the quantity difference;

[0064] Obtain the mean difference between different adjacent riding start times within each time range as the time difference level;

[0065] The product of the quantity difference and the time difference level is obtained and negatively correlated with each other to form the electric vehicle usage rate within each time range.

[0066] It should be noted that the greater the data difference, the greater the difference between the number of riding users and the number of shared electric vehicles, the smaller the number of riding users relative to the number of shared electric vehicles, and the lower the utilization rate of electric vehicles; the greater the time difference level, the greater the difference between the start times of adjacent rides, which means that the farther the start times of adjacent rides are apart, the lower the utilization rate of electric vehicles, showing a negative correlation; in some embodiments of the present invention, the product of the quantity difference and the time difference level can be negatively correlated by taking the inverse method, wherein, in order to avoid the denominator of the formula being 0, it is necessary to manually add a set threshold, such as 0.01; it can also be The function performs negative correlation mapping on the product of the quantity difference and the time difference level. The specific means are well known to those skilled in the art and will not be described in detail here.

[0067] In one embodiment of the present invention, the formula for the electric vehicle utilization rate is expressed as:

[0068] ;

[0069] in, Indicates the The usage rate of electric vehicles within the time range; Indicates the number of cycling users; Indicates the number of shared electric vehicles; It represents the mean difference between different adjacent riding start times within each time range, that is, the time difference level; Take the absolute value.

[0070] In the formula for electric vehicle utilization rate, It represents the difference between the number of cyclists and the number of shared electric vehicles at each time. As the quantity difference, the larger the quantity difference, the smaller the number of cyclists is than the number of shared electric vehicles, and the lower the utilization rate of electric vehicles; the smaller the quantity difference, the closer the number of cyclists is to the number of shared electric vehicles, the more people have demand for shared electric vehicles, and the higher the utilization rate of electric vehicles; It represents the mean difference between different adjacent riding start times within each time range, that is, the time difference level. The smaller the time difference level, the closer the adjacent riding start times are, the greater the demand for shared electric vehicles, and the higher the utilization rate of electric vehicles.

[0071] Step S202: Arrange the usage rates of electric vehicles within all time ranges of each day in chronological order to obtain a fitting curve for fitting the usage rates of electric vehicles.

[0072] By arranging the usage rate data in chronological order, we can intuitively capture this temporal trend of change; by fitting the curve, we can smooth the data, reduce the impact of noise and outliers, and thus more accurately reveal the overall trend of electric vehicle use.

[0073] It should be noted that, in some embodiments of the present invention, curve fitting can be performed using existing fitting methods such as the least squares method and the polynomial fitting method. The specific means are technical means well known to those skilled in the art and will not be described in detail here.

[0074] Step S203: Obtain the average slope value between different adjacent data points on the fitting curve as the overall slope level; within the preset peak time range, obtain the average slope value between different adjacent data points in all peak time ranges on the fitting curve as the peak slope level; obtain the average slope value between different adjacent data points in time ranges other than the peak time range as the non-peak slope level;

[0075] The slope reflects the rate of change between data points and can be used to evaluate the upward or downward trend of the curve. The overall level of the slope between all adjacent data points on the fitted curve is quantified by taking the average value, reflecting the average rate of change in the electric vehicle usage rate. The greater the overall slope level, the greater the upward trend in the electric vehicle usage rate and the greater the demand for use.

[0076] The peak slope level is used to evaluate changes in cycling demand during peak hours, and the off-peak slope mean is used to evaluate changes in cycling demand during off-peak hours. The larger the peak slope level value, the faster the growth in cycling demand during peak hours, and it may be necessary to increase the number of vehicles or replenish headgear.

[0077] It should be noted that, in one embodiment of the present invention, the slope is calculated as the ratio of the vertical coordinate difference to the horizontal coordinate difference between the data points on the fitting curve.

[0078] It should be noted that, in one embodiment of the present invention, the utilization rate of electric vehicles may vary significantly in different time ranges. For example, the utilization rate is higher during the morning and evening rush hours, while it is lower at night. Therefore, the preset peak time ranges are 7:00-8:00, 8:00-9:00, 17:00-18:00, and 18:00-19:00; in other embodiments of the present invention, specific settings can be made according to specific circumstances, and no limitation or elaboration is made here.

[0079] Step S204: Obtain the daily demand for electric vehicle use based on the deviation between the peak slope level and the off-peak slope level, as well as the overall slope level. Both the deviation and the overall slope level are positively correlated with the demand for electric vehicle use.

[0080] It should be noted that, in some embodiments of the present invention, the deviation between the peak slope level and the off-peak slope level can be calculated by calculating the ratio between the peak slope level and the off-peak slope level. The larger the ratio, the larger the peak slope level is compared to the off-peak slope level, and the higher the frequency of electric vehicle use during peak hours. Alternatively, the difference between the peak slope level and the off-peak slope level can be calculated and normalized to 0-1. The larger the result, the larger the peak slope level is compared to the off-peak slope level, and the greater the usage demand. The specific means are well known to those skilled in the art and will not be elaborated here.

[0081] Step S3: Based on the distribution of electric vehicle usage demand on different days in each quarter, and the correlation between electric vehicle usage demand and riding electricity consumption on different days, the quarterly usage intensity of each quarter is obtained; based on the distribution of the number of rides of all shared electric vehicles within all time ranges in each quarter, the popularity of the deployment area in each quarter is obtained.

[0082] Shared electric vehicle usage varies across quarters. Analyzing the distribution of demand across different days within each quarter provides a more comprehensive understanding of overall usage for that quarter. There's a correlation between demand for electric vehicles and riding electricity consumption; periods of high demand typically also see higher electricity consumption. This correlation analysis allows for a more accurate identification of the seasonal impact of demand for electric vehicles and an assessment of quarterly usage intensity. Based on the distribution of demand for electric vehicles across different days within each quarter and the correlation between demand and riding electricity consumption, we can derive the quarterly usage intensity for each quarter.

[0083] Preferably, in one embodiment of the present invention, the method for obtaining quarterly usage intensity can be found in Figure 3 , which shows a flow chart of a method for obtaining quarterly usage intensity, including:

[0084] Step S301: Obtaining the electric vehicle thermal value for each quarter based on the electric vehicle usage demand distribution on different days in each quarter.

[0085] The electric vehicle thermal value is a quantitative representation of usage demand in that quarter and can reflect the riding activity in the region.

[0086] Preferably, in one embodiment of the present invention, the method for obtaining the thermal value of an electric vehicle includes:

[0087] Obtain the cumulative sum of electric vehicle usage demands for all days in each quarter as the first cumulative sum of demands; obtain the cumulative sum of electric vehicle usage demands for all days in a historical year as the second cumulative sum of demands;

[0088] The ratio between the first demand cumulative sum and the second demand cumulative sum is obtained as the electric vehicle thermal value for each quarter.

[0089] Step S302: For each quarter, the sum of the cycling electricity consumption within all time ranges of each day is obtained as the total cycling electricity consumption per day; the mean square error of the sequence composed of the electric vehicle usage demand and the total cycling electricity consumption on different days is obtained, and negative correlation mapping is performed, which is used as the first trend coefficient.

[0090] The total amount of electricity consumed during riding is a direct indicator of the daily usage intensity of electric vehicles. By summarizing the amount of electricity consumed during riding at all times of the day, it reflects the comprehensive value of the electric vehicle usage activity on that day.

[0091] The mean square error is a statistical method to measure the degree of difference between two sequences. It helps to quantify the correlation between the demand for electric vehicle use and the total electricity consumption. The larger the mean square error, the smaller the correlation and the smaller the first trend coefficient. The smaller the mean square error, the greater the correlation and the larger the first trend coefficient.

[0092] Step S303: Obtain the product of the electric vehicle thermal value and the first trend coefficient for the corresponding quarter as the quarterly usage intensity for each quarter.

[0093] In one embodiment of the present invention, the formula for quarterly usage intensity is expressed as:

[0094] ;

[0095] ;

[0096] in, Indicates the quarterly intensity of usage by quarter; Indicates the Quarterly electric vehicle thermal value; Indicates the A sequence of electric vehicle usage demands on different days of the quarter; Indicates the A sequence consisting of the total amount of electricity used for cycling on different days of the quarter; Indicates the Daily demand for electric vehicles; Indicates the Daily demand for electric vehicles; Indicates the Number of days in the quarter; Indicates the number of days in a year; Represents the mean square error function.

[0097] In the formula for quarterly usage intensity, Express Perform negative correlation mapping, that is, the first trend coefficient, To avoid the denominator of the formula being 0, the formula will be meaningless; Indicates the The mean square error between the series composed of the electric vehicle usage demand and the series composed of the total riding electricity consumption on different days in the quarter. The larger the mean square error, the smaller the correlation between the electric vehicle usage demand and the total riding electricity consumption. The smaller the mean square error, the greater the correlation between the electric vehicle usage demand and the total riding electricity consumption. The greater the total riding electricity consumption, the greater the electric vehicle usage demand, the greater the electric vehicle thermal value, and the greater the quarterly usage intensity.

[0098] The distribution of ride counts reflects the usage of shared electric vehicles in different time periods and regions. By analyzing the distribution of ride counts in each quarter, high- and low-heat areas can be identified. The more rides, the higher the heat. The heat of the deployment area is a comprehensive assessment of the riding activity in the area, which can reflect the intensity of regional demand. The heat of the deployment area in each quarter is obtained based on the distribution of ride counts of all shared electric vehicles at all times in each quarter.

[0099] Preferably, in one embodiment of the present invention, the method for obtaining the popularity of the delivery area includes:

[0100] Get the number of hotspots within the service area of ​​the delivery area. Hotspots include business districts, subway stations, and schools.

[0101] The sum of the number of rides for each shared electric vehicle at all times of the day is obtained as the first sum; the average of the first sums of all shared electric vehicles on all days of each quarter is obtained as the average number of rides;

[0102] The product of the average number of rides and the number of hot spots in the delivery area is obtained and normalized to obtain the delivery area heat index for each quarter.

[0103] It should be noted that the service scope of the deployment area can be obtained in advance by the implementation personnel based on relevant professional information, and the number of hot spots within the service scope can be analyzed through map software. The more hot spots there are, the greater the flow of people, and the greater the demand for shared electric vehicles may be. The greater the average number of rides in each quarter, the more popular the deployment area will be.

[0104] In one embodiment of the present invention, for each quarter, the formula for the popularity of the delivery area is expressed as: ;in, Indicates the popularity of the delivery area in each quarter; Indicates the number of hotspot areas corresponding to the delivery area; represents the average number of rides; Represents the normalization function.

[0105] Step S4: Based on the popularity of the launch area in each quarter, the quarterly usage intensity, and the battery health distribution of all shared electric vehicles at different time ranges, the riding power consumption and the headgear usage are adjusted to obtain the headgear usage correction amount and the riding power consumption correction amount.

[0106] New batteries or batteries in good health can provide longer driving range, so under the same power conditions, these batteries can support longer riding distances; aging batteries or batteries in poor health can significantly shorten the riding distance under the same power conditions, which may increase the frequency of vehicle battery replacement; the popularity of the deployment area and the quarterly usage intensity can reflect the changes in riding demand and frequency, affecting power consumption and the speed of headgear usage. High-heat areas consume power and use headgear faster, while low-heat areas consume power more slowly. By combining regional heat data, the riding power consumption and headgear usage in different regions can be corrected under different quarterly conditions; according to the popularity of the deployment area in each quarter, the quarterly usage intensity and the battery health distribution of all shared electric vehicles at different time ranges, the riding power consumption and headgear usage are adjusted to obtain the headgear usage correction amount and the riding power correction amount.

[0107] Preferably, in one embodiment of the present invention, the method for obtaining the headgear usage correction amount and the power consumption correction amount includes:

[0108] Based on the popularity of the launch area in each quarter, the quarterly usage intensity, and the battery health distribution of all shared electric vehicles at different time ranges, the battery replacement forecast correction coefficient for each quarter is obtained;

[0109] Preferably, in one embodiment of the present invention, the method for obtaining the battery swap prediction correction coefficient includes:

[0110] For each quarter, the average battery health of each shared electric vehicle at all times is obtained as the local battery health level; the average local battery health level of all shared electric vehicles is obtained as the overall battery health level;

[0111] The quarterly usage intensity of each quarter is normalized and mapped, and the sum of the normalized mapping result and the heat of the deployment area is obtained as the weighted value; the product of the weighted value and the overall health level of the battery is obtained as the battery replacement prediction correction coefficient for each quarter.

[0112] It should be noted that the greater the quarterly usage intensity, the more popular the deployment area, the greater the usage demand, the faster the power consumption, and the more positively correlated with the battery replacement prediction correction coefficient; the lower the overall health level of the battery, the significantly shortened riding distance, which may increase the frequency of vehicle battery replacement, and is negatively correlated with the battery replacement prediction correction coefficient.

[0113] In one embodiment of the present invention, the formula for the battery swap prediction correction coefficient is expressed as:

[0114] ;

[0115] in, Indicates the Quarterly battery replacement forecast correction factor; Indicates the The quarterly popularity value of the delivery area; Indicates the quarterly intensity of usage by quarter; Indicates the The overall battery health of shared electric vehicles during the quarter; represents the hyperbolic tangent function.

[0116] For the headgear usage or riding electricity consumption as the data to be corrected, the battery replacement forecast correction coefficient of each quarter and the corresponding data to be corrected in each time range are multiplied together to obtain the headgear usage correction amount and riding usage correction amount in each time range.

[0117] Step S5: predicting the optimal time for replenishing the headgear based on the distribution of the headgear usage correction amount and the riding power correction amount of all shared electric vehicles within different time ranges.

[0118] Combined with the distribution of correction amounts for headgear usage and riding electricity consumption in a historical year, it reflects the usage habits and demand characteristics of electric vehicles in the deployment area, which helps to improve the operation and maintenance efficiency of shared electric vehicles.

[0119] Preferably, in one embodiment of the present invention, the optimal replenishment time includes:

[0120] For the correction amount for headgear use or the correction amount for riding electricity consumption as the target data, the sum of the target data of each shared electric vehicle at different time ranges every day is obtained as the target overall data of each shared electric vehicle on each day; the mean of the target overall data of all shared electric vehicles on each day is obtained as the target overall average data of each day; a fitting curve is constructed to fit the target overall average data of all days, and the intersection of the corresponding fitting curves between the correction amount for headgear use and the correction amount for riding electricity consumption is selected;

[0121] Obtain the mean slope value between all adjacent data points on the fitting curve between adjacent intersections. If the difference between the corresponding mean slope values ​​of two fitting curves between adjacent intersections is less than a preset difference threshold, the time range between the corresponding adjacent intersections is used as the optimal supplementary time for the headgear.

[0122] It should be noted that the smaller the difference in the mean slopes, the more consistent the changing trends between the fitting curves, and the more similar the headgear usage and power consumption performance; in one embodiment of the present invention, the size of the preset difference threshold is set to 0.3; in other embodiments of the present invention, the size of the preset difference threshold can be set according to the specific situation, and is not limited or elaborated here.

[0123] After obtaining the optimal time to replenish the headgear, the number threshold sensor in the headgear mechanism is used to obtain the number of missing headgear in the deployment area of ​​shared electric vehicles at the corresponding time in the next year, and then the demand value for headgear carried while replacing batteries in the area is obtained, so as to replenish the headgear more accurately and provide accurate operation and maintenance plans for operation and maintenance personnel to adapt to the high-frequency use of shared electric vehicles.

[0124] It should be noted that if the electricity consumption curve grows at a faster rate, it means that the demand for cycling in the deployment area is high, but the headgear usage curve grows at a slower rate, which means that users in the area may be more inclined to use their own helmets or a cycling method that does not rely on headgear. At this time, the headgear inventory should be dynamically adjusted and the excess headgear should be allocated to areas with higher cycling demand and headgear usage frequency to optimize resource utilization.

[0125] In summary, the present invention obtains the daily demand for electric vehicle use based on the difference between the number of riding users and the number of shared electric vehicles at different time ranges of the day, as well as the distribution of riding start times; obtains the quarterly usage intensity of each quarter based on the distribution of electric vehicle use demand on different days in each quarter, and the correlation between the electric vehicle use demand and riding power consumption on different days; adjusts riding power consumption and headgear usage based on the popularity of the launch area of ​​each quarter and the battery health distribution of all shared electric vehicles at different time ranges, obtains the headgear usage correction amount and riding power correction amount; and predicts the optimal replenishment time for headgear replenishment. The present invention optimizes the operation time for headgear replenishment by analyzing the historically accurate headgear usage and riding power consumption.

[0126] The present invention also proposes a shared electric vehicle head cover replenishment prediction system based on data analysis, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, it implements any one of the steps of a shared electric vehicle head cover replenishment prediction method based on data analysis.

[0127] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0128] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for predicting the replenishment of shared electric vehicle headgear based on data analysis, characterized in that: The method comprises: Obtain the number of riders in the deployment area at different times of day over the past year, as well as the riding data for each shared electric vehicle. The riding data includes the number of rides, power consumption, headgear usage, ride start time, and battery health. According to the difference between the number of riding users and the number of shared electric vehicles at different time ranges of the day, as well as the distribution of riding start times, the daily demand for electric vehicle use is obtained, including: according to the difference between the number of riding users and the number of shared electric vehicles at each time range, as well as the distribution of riding start times, the electric vehicle usage rate at each time range is obtained; the electric vehicle usage rates at all time ranges of the day are arranged in chronological order to obtain a fitting curve for fitting the electric vehicle usage rate; the mean slope between different adjacent data points on the fitting curve is obtained as the overall slope level; for the preset peak time range, the slope between different adjacent data points at all peak time ranges on the fitting curve is obtained. The average value is used as the peak slope level; the average value of the slopes between different adjacent data points in other time ranges except the peak time range is obtained as the off-peak slope level; the ratio of the peak slope level to the off-peak slope level is obtained, and the product of the ratio result and the overall slope level is calculated as the daily demand for electric vehicle use; wherein, the method for obtaining the electric vehicle utilization rate includes: obtaining the difference between the number of users and the number of shared electric vehicles in each time range as the quantity difference; obtaining the average value of the difference between different adjacent riding start times in each time range as the time difference level; obtaining the product of the quantity difference and the time difference level, and performing negative correlation mapping as the electric vehicle utilization rate in each time range; Based on the distribution of electric vehicle usage demand on different days in each quarter and the correlation between electric vehicle usage demand and riding electricity consumption on different days, the quarterly usage intensity of each quarter is obtained. Based on the distribution of the number of rides of all shared electric vehicles at all times in each quarter, the popularity of the deployment area in each quarter is obtained. According to the popularity of the launch area in each quarter, the quarterly usage intensity and the battery health distribution of all shared electric vehicles in different time ranges, the riding electricity consumption and headgear usage are adjusted to obtain the headgear usage correction amount and the riding electricity consumption correction amount, including: according to the popularity of the launch area in each quarter, the quarterly usage intensity and the battery health distribution of all shared electric vehicles in different time ranges, the battery replacement prediction correction coefficient of each quarter is obtained; for the headgear usage or riding electricity consumption as the data to be corrected, the battery replacement prediction correction coefficient of each quarter is multiplied by the data to be corrected in each time range. As the correction amount for headgear use within each time range, and the correction amount for riding use; wherein, the method for obtaining the battery replacement prediction correction coefficient includes: for each quarter, obtaining the average value of the battery health of each shared electric vehicle within all time ranges as the local health level of the battery; obtaining the average value of the local health levels of the batteries of all shared electric vehicles as the overall health level of the battery; performing normalized mapping on the quarterly usage intensity of each quarter, obtaining the sum of the normalized mapping result and the heat of the deployment area as the weighted value; obtaining the product of the weighted value and the overall health level of the battery as the battery replacement prediction correction coefficient for each quarter; Based on the distribution of headgear usage correction amounts and riding electricity correction amounts within different time ranges for all shared electric vehicles, the optimal replenishment time for headgear replenishment is predicted, including: taking the headgear usage correction amount or the riding electricity correction amount as the target data, obtaining the sum of the target data of each shared electric vehicle within different time ranges every day as the target overall data of each shared electric vehicle every day; obtaining the mean of the target overall data of all shared electric vehicles every day as the target overall average data for every day; constructing a fitting curve for fitting the target overall average data of all days, and selecting the intersection of the corresponding fitting curves between the headgear usage correction amount and the riding electricity correction amount; obtaining the mean of the slopes between all adjacent data points on the fitting curve between adjacent intersections. If the difference in the corresponding slope means of the two fitting curves between adjacent intersections is less than a preset difference threshold, the time range between the corresponding adjacent intersections will be used as the optimal replenishment time for headgear replenishment.

2. A method for predicting the replenishment of shared electric vehicle head covers based on data analysis according to claim 1, characterized in that: The method for obtaining the quarterly usage intensity includes: According to the distribution of electric vehicle usage demand on different days in each quarter, the electric vehicle thermal value of each quarter is obtained; For each quarter, the sum of the riding electricity consumption of all shared electric vehicles at all times of the day is obtained as the total daily riding electricity consumption. The mean square error of the series composed of the electric vehicle usage demand and the total riding electricity consumption on different days is obtained and negatively correlated with each other, which is used as the first trend coefficient. The product of the electric vehicle thermal value and the first trend coefficient of the corresponding quarter is obtained as the quarterly usage intensity of each quarter.

3. A method for predicting the replenishment of shared electric vehicle head covers based on data analysis according to claim 2, characterized in that: The method for obtaining the thermal value of the electric vehicle includes: Obtain the cumulative sum of electric vehicle usage demands for all days in each quarter as the first cumulative sum of demands; obtain the cumulative sum of electric vehicle usage demands for all days in a historical year as the second cumulative sum of demands; The ratio between the first demand cumulative sum and the second demand cumulative sum is obtained as the electric vehicle thermal value for each quarter.

4. The method for predicting the replenishment of shared electric vehicle head covers based on data analysis according to claim 1 is characterized in that: The method for obtaining the heat of the delivery area includes: Obtain the number of hotspots within the service area of ​​the delivery area, including business districts, subway stations, and schools. The sum of the number of rides for each shared electric vehicle at all times of the day is obtained as the first sum; the average of the first sums of all shared electric vehicles on all days of each quarter is obtained as the average number of rides; The product of the average number of rides and the number of hot spots in the delivery area is obtained and normalized to obtain the delivery area heat index for each quarter.

5. A shared electric vehicle headgear replenishment prediction system based on data analysis, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of a shared electric vehicle headgear replenishment prediction method based on data analysis as described in any one of claims 1 to 4 are implemented.

Images