Method and system for determining battery replacement threshold value of electric vehicle

Abstract

The embodiment of the invention provides a method and system for determining a battery replacement threshold value of an electric vehicle, and the method comprises the steps: determining a first electric quantity threshold value corresponding to a target space-time unit based on a first historical order of borrowing the electric vehicle corresponding to the target space-time unit, the first electric quantity threshold value is an electric quantity boundary value which cannot be borrowed by the electric vehicle in the limited target space-time unit; dividing the target space-time unit into multiple classes of sub space-time units; and for each type of sub-space-time unit of the target space-time unit, based on a second historical order of borrowing the electric vehicle corresponding to the sub-space-time unit and the first electric quantity threshold, a second electric quantity threshold corresponding to the sub-space-time unit is determined, and the second electric quantity threshold is an electric quantity boundary value for limiting that the electric vehicle in the sub-space-time unit needs to be replaced.

Description

Technical Field

[0001] This specification relates to the technical field of electric vehicles, and particularly to a method and system for determining the battery replacement threshold of an electric vehicle. Background Art

[0002] Shared electric vehicles, including shared two-wheelers and shared cars, are widely used as green travel transportation tools in today's society. As users borrow and ride electric vehicles, the battery power is continuously consumed, and gradually, the power of some batteries will be exhausted and unable to provide power continuously. At this time, the operators of shared electric vehicles (such as operation and maintenance personnel / battery replacement personnel) need to replace the batteries with low power. Usually, the operator will set a battery replacement threshold for the shared electric vehicle, and regard the vehicle with a battery power lower than the battery replacement threshold as a low-power vehicle.

[0003] However, if the battery replacement threshold is set too low, it will lead to an increase in the proportion of low-power vehicles in the operation area, users cannot borrow and ride the vehicle, resulting in order losses and reducing the user experience; if the battery replacement threshold is set too high, it will lead to a significant increase in the battery replacement cost. Therefore, it is necessary to reasonably balance the two indicators to determine the optimal battery replacement threshold. In addition, different operators will use different models of batteries, and users in different regions and seasons also have different riding habits. If factors such as battery models, regions, and seasons are not considered when setting the battery replacement threshold indicators, the set battery replacement threshold will not be reasonable enough.

[0004] Therefore, a method and system for determining the battery replacement threshold of an electric vehicle are provided, which can improve the utilization efficiency of the battery power and reduce the operation and maintenance cost of battery replacement for electric vehicles on the basis of meeting the borrowing needs of users for electric vehicles, so as to balance the user needs and operation and maintenance management. Summary of the Invention

[0005] One embodiment of this specification provides a method for determining the battery replacement threshold of an electric vehicle, including: determining a first power threshold corresponding to the target spatio-temporal unit based on the first historical order of the borrowed electric vehicle corresponding to the target spatio-temporal unit, where the first power threshold is the power demarcation value that limits the non-borrowable electric vehicle in the target spatio-temporal unit; dividing the target spatio-temporal unit into multiple types of sub-spatio-temporal units; and for each type of sub-spatio-temporal unit of the target spatio-temporal unit, determining a second power threshold corresponding to the sub-spatio-temporal unit based on the second historical order of the borrowed electric vehicle corresponding to the sub-spatio-temporal unit and the first power threshold, where the second power threshold is the power demarcation value that limits the electric vehicle in the sub-spatio-temporal unit to need battery replacement.

[0006] One embodiment of this specification provides a system for determining the battery replacement threshold of an electric vehicle, including: a first determination module, configured to determine a first power threshold corresponding to the target spatio-temporal unit based on the first historical order of the borrowed electric vehicle corresponding to the target spatio-temporal unit, where the first power threshold is the power demarcation value that defines that the electric vehicle in the target spatio-temporal unit cannot be borrowed; a spatio-temporal unit management module, configured to divide the target spatio-temporal unit into multiple types of sub-spatio-temporal units; and a second determination module, configured to, for each type of sub-spatio-temporal unit of the target spatio-temporal unit, determine a second power threshold corresponding to the sub-spatio-temporal unit based on the second historical order of the borrowed electric vehicle corresponding to the sub-spatio-temporal unit and the first power threshold, where the second power threshold is the power demarcation value that defines that the electric vehicle in the sub-spatio-temporal unit needs to replace the battery.

[0007] One embodiment of this specification provides a device for determining the battery replacement threshold of an electric vehicle, including a processor, where the processor is configured to execute the method for determining the battery replacement threshold of the electric vehicle described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] This specification will be further described by way of exemplary embodiments, and these exemplary embodiments will be described in detail through the drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where:

[0009] Figure 1 is a schematic diagram of the application scenario of the system for determining the battery replacement threshold of an electric vehicle shown in some embodiments of this specification;

[0010] Figure 2 is a schematic diagram of the modules of the system for determining the battery replacement threshold of an electric vehicle shown in some embodiments of this specification;

[0011] Figure 3 is an exemplary flowchart of the method for determining the battery replacement threshold of an electric vehicle shown in some embodiments of this specification;

[0012] Figure 4a is an exemplary flowchart of the method for determining the first power threshold shown in some embodiments of this specification;

[0013] Figure 4b is a schematic diagram of the process of determining the first power threshold based on the first machine learning model shown in some embodiments of this specification;

[0014] Figure 5a is an exemplary flowchart of the method for determining the second power threshold shown in some embodiments of this specification;

[0015] Figure 5bIt is a schematic diagram of a process for determining a second power threshold based on a second machine learning model shown in some embodiments of this specification. Detailed implementation manners

[0016] To more clearly illustrate the technical solutions of the embodiments of this specification, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some examples or embodiments of this specification. For those of ordinary skill in the art, without creative efforts, this specification can also be applied to other similar scenarios based on these drawings. Unless obvious from the language context or otherwise stated, the same reference numerals in the drawings represent the same structure or operation.

[0017] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the said words can be replaced by other expressions.

[0018] Unless the context clearly indicates an exceptional situation, words such as "a", "an", "one" and / or "the" are not specifically singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0019] Flowcharts are used in this specification to illustrate the operations performed by the systems according to the embodiments of this specification. It should be understood that the previous or subsequent operations are not necessarily executed precisely in sequence. On the contrary, the steps can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0020] Figure 1 It is a schematic diagram of an application scenario of a system for determining an electric vehicle battery swapping threshold shown in some embodiments of this specification.

[0021] As Figure 1 shown, the application scenario 100 of the system for determining an electric vehicle battery swapping threshold may include an electric vehicle 110, a network 120, a terminal 130, a processing device 140, and a storage device 150.

[0022] The electric vehicle 110 refers to a device that uses a battery as an energy source. For example, the electric vehicle 110 may include a two-wheeled electric vehicle with a detachable battery, an electric vehicle, etc., which are devices that need to be charged / replaced. The electric vehicle 110 may include a built-in vehicle management program / application for monitoring the operation of the vehicle (such as the current location of the vehicle, the remaining battery power, the available endurance time, etc.) and may send it to the processing device 140.

[0023] The electric vehicle 110 may include electric vehicles of multiple different brands (or manufacturers) (such as electric vehicle 110-1... electric vehicle 110-n, etc.). The batteries of electric vehicles 110 of different brands can be charged and / or replaced by users (such as operation and maintenance personnel, battery replacement personnel). For example, when the current remaining battery power of the electric vehicle 110 is lower than the first battery power threshold and / or the second battery power threshold, the user can perform charging and battery replacement operations on it.

[0024] The network 120 may include any suitable network that facilitates the exchange of information and / or data in the application scenario 100. In some embodiments, one or more components in the application scenario 100 (for example, the electric vehicle 110, the terminal 130, the processing device 140, or the storage device 150) may transmit information and / or data to one or more other components in the application scenario 100 via the network 120. For example, the processing device 140 may obtain relevant information of the electric vehicle 110 via the network 120, such as the remaining battery power information and location information of the electric vehicle. In some embodiments, the network 120 may be any one or more of a wired network or a wireless network. In some embodiments, the network may be various topological structures such as point-to-point, shared, centralized, etc., or a combination of multiple topological structures.

[0025] The terminal 130 may include a mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, etc., or any combination thereof. In some embodiments, the terminal 130 may interact with other components in the application scenario 100 via the network 120. For example, the terminal 130 may receive the usage conditions of the electric vehicle 110 (such as the order quantity, the distribution of riding mileage, the power consumption of the battery, the power-off situation during riding, etc.) in the operation area (such as a city, district, county, etc.) sent by the processing device 140. In some embodiments, the terminal 130 may receive information and / or instructions input by a user (such as operation and maintenance personnel or battery replacement personnel, etc.) and send the received information and / or instructions to other components in the application scenario 100 via the network 120. For example, the user can adjust the first battery power threshold and / or the second battery power threshold of the electric vehicle 110 according to the demand situation (such as borrowing frequency, order quantity, etc.) and operation and maintenance situation (such as the number of battery replacement personnel, the degree of busyness, etc.) of the electric vehicle 110 in the operation area. For the relevant content of the first battery power threshold and the second battery power threshold, seeFigure 3 and its description.

[0026] In some embodiments, the terminal 130 may display the battery swapping status of the electric vehicle 110 in the operation area. For example, it may display the number or proportion of electric vehicles 110 with the remaining battery power lower than the first power threshold and / or the second power threshold in the operation area, the assignment information of the battery swapping personnel, etc.

[0027] The processing device 140 may process the data and / or information obtained from the electric vehicle 110, the terminal 130, and / or the storage device 150. For example, the processing device 140 may obtain the power usage of the electric vehicle 110. For another example, the processing device 110 may send a control instruction to the electric vehicle 110 through the network 120 to adjust the borrowable or non-borrowable state of the electric vehicle 110. In some embodiments, the processing device 140 may process the historical order data of the electric vehicle 110. For example, adjust the first power threshold and / or the second power threshold according to the historical order data. In some embodiments, the processing device 140 may obtain the data and / or instructions of the terminal 130 and perform corresponding processing. For example, adjust the borrowable or non-borrowable state of the electric vehicle 110 in real time according to the instructions of the user (such as the operation and maintenance personnel).

[0028] In some embodiments, the processing device 140 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processing device 140 may be local or remote. The processing device 140 may be directly connected to the electric vehicle 110, the terminal 130, and the storage device 150 to access the stored or obtained information and / or data. In some embodiments, the processing device 140 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc. or any combination thereof.

[0029] The storage device 150 may store data and / or instructions. In some embodiments, the storage device 150 may store data obtained from the electric vehicle 110, the terminal 130, and / or the processing device 140. For example, the storage device 150 may store usage information of the electric vehicle 110 (such as order information, mileage information, location distribution information, riding power-off rate, low battery rate, etc.). For another example, the storage device 150 may store the first power threshold and the second power threshold corresponding to different operating areas. In some embodiments, the storage device 150 may store data and / or instructions for the processing device 140 to execute the exemplary methods described in this specification. For example, the storage device 150 may store instructions for the processing device 140 to execute the methods shown in each flowchart. In some embodiments, the storage device 150 may include a mass storage device, a removable storage device, a volatile read-write memory, a read-only memory (ROM), etc., or any combination thereof. In some embodiments, the storage device 150 may be implemented on a cloud platform. In some embodiments, the storage device 150 may be a part of the processing device 140

[0030] The above description is for illustrative purposes only, and actual application scenarios can vary in various ways.

[0031] It should be noted that the application scenario 100 is provided only for illustrative purposes and is not intended to limit the scope of the present application. For those of ordinary skill in the art, various modifications or changes can be made according to the description of this specification. However, these changes and modifications will not depart from the scope of the present application.

[0032] Figure 2 is a schematic diagram of the modules of a system for determining the battery swapping threshold of an electric vehicle according to some embodiments of this specification.

[0033] As Figure 2 shown, the system 200 for determining the battery swapping threshold of an electric vehicle (hereinafter referred to as the battery swapping system) may include a first determination module 210, a spatio-temporal unit management module 220, and a second determination module 230. In some embodiments, the first determination module 210, the spatio-temporal unit management module 220, and the second determination module 230 may be implemented by the processing device 140.

[0034] The first determination module 210 is configured to determine a first power threshold corresponding to a target spatio-temporal unit based on a first historical order of a borrowed electric vehicle corresponding to the target spatio-temporal unit, where the first power threshold is a power demarcation value that limits the non-borrowability of the electric vehicle in the target spatio-temporal unit.

[0035] In some embodiments, the target spatio-temporal unit includes a plurality of target spatio-temporal units belonging to a target category, the order levels of the plurality of target spatio-temporal units are within a first preset range, and the average temperature of the plurality of target spatio-temporal units is within a second preset range.

[0036] In some embodiments, the first determination module 210 is further configured to: divide the first historical order into a plurality of first order sets, each first order set including the first historical orders corresponding to the historical first power threshold; for each first order set, determine a first data sample, the first data sample at least including the riding power-off rate and the optimal mileage of the first historical orders in the first order set, the optimal mileage being determined based on the historical first power threshold corresponding to the first order set; and determine the first power threshold based on the first data sample corresponding to each first order set.

[0037] In some embodiments, the first determination module 210 is further configured to: determine a plurality of target first data samples that meet the preset power-off rate condition from the plurality of first data samples based on the riding power-off rate corresponding to each first data sample; determine the target optimal mileage corresponding to the target spatio-temporal unit based on the optimal mileage of each target first data sample in the plurality of target first data samples; and determine the first power threshold based on the target optimal mileage.

[0038] In some embodiments, the first determination module 210 is further configured to: generate a first machine learning model with the riding power-off rate corresponding to the first data sample as the training input and the optimal mileage corresponding to the first data sample as the training label; determine the target optimal mileage corresponding to the target spatio-temporal unit based on the first machine learning model; and determine the first power threshold based on the target optimal mileage.

[0039] In some embodiments, the first data sample further includes the power consumption value per unit mileage and the preset error value.

[0040] In some embodiments, the preset error value is related to the usage information of the battery of the target model.

[0041] The spatio-temporal unit management module 220 is configured to divide the target spatio-temporal unit into multiple types of sub-spatio-temporal units.

[0042] The second determination module 230 is configured to, for each type of sub-spatio-temporal unit of the target spatio-temporal unit, determine the second power threshold corresponding to the sub-spatio-temporal unit based on the second historical order of the borrowed electric vehicle corresponding to the sub-spatio-temporal unit and the first power threshold, the second power threshold being the power demarcation value for limiting the battery replacement of the electric vehicle in the sub-spatio-temporal unit.

[0043] In some embodiments, the second determination module 230 is further configured to: for each type of sub-spatiotemporal unit, divide the second historical orders of the sub-spatiotemporal unit into a plurality of second order sets, where each second order set includes second historical orders corresponding to a corresponding historical second power threshold; for each second order set, determine a second data sample, where the second data sample at least includes the vehicle low power rate and the optimal order number corresponding to the second historical orders in the second order set, the vehicle low power rate is determined based on the first power threshold, and the optimal order number is determined based on the first power threshold and the historical second power threshold corresponding to the second order set; and based on the second data sample corresponding to each second order set, determine the second power threshold corresponding to the sub-spatiotemporal unit.

[0044] In some embodiments, the second determination module 230 is further configured to: based on the vehicle low power rate corresponding to each second data sample, determine a plurality of target second data samples that meet the preset low power rate condition from the plurality of second data samples; based on the optimal order number of each target second data sample among the plurality of target second data samples, determine the target optimal order number corresponding to the sub-spatiotemporal unit; and based on the target optimal order number, determine the second power threshold.

[0045] In some embodiments, the second determination module 230 is further configured to: use the vehicle low power rate corresponding to the second data sample as the training input and the optimal order number corresponding to the second data sample as the training label to generate a second machine learning model; based on the second machine learning model, determine the target optimal order number corresponding to the target spatiotemporal unit; and based on the target optimal order number, determine the second power threshold.

[0046] It should be noted that the above description of the battery swapping system and its modules is only for convenience of description and does not limit this specification to the scope of the examples given. It can be understood that for those skilled in the art, after understanding the principle of the system, they may, without departing from this principle, make any combination of the modules, or form a subsystem and connect it with other modules. For example, the first determination module 210, the spatiotemporal unit management module 220, and the second determination module 230 may be different modules in the system, or one module may implement the functions of two or more of the above modules. For example, each module may share a storage module, or each module may have its own storage module respectively. Such variations are all within the protection scope of this specification.

[0047] Figure 3 is an exemplary flowchart of a method for determining the battery swapping threshold of an electric vehicle according to some embodiments of this specification.

[0048] In some embodiments, process 300 may be executed by a battery swapping system. As Figure 3 shown, process 300 includes the following steps.

[0049] Step 310: Determine a first power threshold corresponding to the target spatio-temporal unit based on the first historical order of the borrowed electric vehicle corresponding to the target spatio-temporal unit. The first power threshold is the power demarcation value that limits the non-borrowability of the electric vehicle in the target spatio-temporal unit.

[0050] In this specification, a spatio-temporal unit refers to a unit with a certain spatial range and a certain time range. For example, a map can be divided into multiple spatial regions (such as urban, district, county, etc. regions), and the whole year can be divided into multiple time periods (such as the four seasons of spring, summer, autumn, and winter, holidays). Each spatio-temporal unit can refer to a specific spatial region in a specific time period. Only as an example, a spatio-temporal unit can be a region in a high-latitude area during the summer period, a region in a high-latitude area during the winter period, a certain city during the holiday period, etc.

[0051] The target spatio-temporal unit refers to the spatio-temporal unit for which the power replacement threshold analysis needs to be performed. For example, the target spatio-temporal unit can be Beijing in spring.

[0052] In some embodiments, the target spatio-temporal unit may include multiple spatio-temporal units belonging to the target category. The spatio-temporal units can be classified based on various classification metrics. Exemplarily, the classification metrics include order level, average temperature, population density, geographical information (such as the level of latitude), economic development (such as GDP, per capita consumption level), residents' living habits, etc. Only as an example, when the classification metrics are the order level and the average temperature, the order levels of the multiple target spatio-temporal units in the target category are within the first preset range, and the average temperatures of the multiple target spatio-temporal units are within the second preset range. That is to say, the power replacement system can use multiple spatio-temporal units with similar order levels and average temperatures as spatio-temporal units of the same target category (i.e., target spatio-temporal units).

[0053] The order level can reflect the magnitude of the borrowing demand of consumer users for borrowed electric vehicles, and it can be determined based on the number of car borrowing orders (such as the daily average order volume) in each spatio-temporal unit. In some embodiments, the power replacement system can set multiple preset order level intervals according to a preset order quantity step (such as 20,000 orders). For example, (0, 2], (2, 4], (4, 8], (8, 12], (12, 16]... etc. The first preset range can be any one of the multiple preset order level intervals.

[0054] The average temperature can be determined based on the daily average temperature in each spatio-temporal unit. In some embodiments, the power replacement system can set multiple preset average temperature intervals according to a preset temperature step (such as 10°). For example, (-20, -10], (-10, 0], (0, 10], (10, 20], (20, 35]... etc. The second preset range can be any one of the multiple preset average temperature intervals.

[0055] By way of example only, the target spatio-temporal unit may include a plurality of spatio-temporal units with an order level in the range of (2, 4] and an average temperature in the range of (0, 10]. In some embodiments, the spatio-temporal units of each category may be respectively used as the target spatio-temporal unit, and process 300 may be executed to determine the corresponding battery replacement thresholds. It can be understood that the riding habits of users in spatio-temporal units with similar order levels and average temperatures are relatively similar, and the battery replacement threshold analysis can be jointly performed to improve the efficiency of the battery replacement threshold analysis. Other factors (such as population density, per capita consumption level, etc.) may also be considered by the battery replacement system when determining the target spatio-temporal unit.

[0056] In some embodiments of the present specification, by classifying spatio-temporal units and performing battery replacement threshold analysis for different categories of spatio-temporal units, targeted operation and management of multiple categories of spatio-temporal units can be achieved, which is beneficial to the subsequent unified configuration or promotion of the first power threshold and / or the second power threshold for the same type of spatio-temporal units, and improves the efficiency of operation and management.

[0057] The first historical order refers to the historical vehicle borrowing order of the electric vehicle within the target spatio-temporal unit. For example, the first historical order may include a set of vehicle borrowing orders generated after the electric vehicle in the target spatio-temporal unit is borrowed by a consumer user (such as a renter) within a certain time period (such as one month, one week, etc.).

[0058] In some embodiments, the first historical order refers to the vehicle borrowing order of the electric vehicle equipped with the battery of the target model. At this time, the first power threshold and the second power threshold determined based on process 300 are applicable to the electric vehicle equipped with the battery of the target model. Different models of batteries have different endurance capabilities and power consumption characteristics. By performing battery replacement threshold analysis for different models of batteries respectively, the accuracy of the determined battery replacement threshold can be improved.

[0059] The vehicle borrowing order may include order information. The order information may include trip information. The trip information is used to characterize the trip trajectory of the vehicle, which may include the starting position point, the ending position point of the trip, and multiple position points during the trip. In some embodiments, the battery replacement system may obtain the real-time position information of the vehicle (such as geographical longitude and latitude, road name, etc.) through the position acquisition device (such as GPS, etc.) of the electric vehicle to determine the trip information. The order information may also include the order mileage. The battery replacement system may determine the mileage of the vehicle borrowing order according to the trip information, for example, 2 km, 5 km, etc. The order information also includes the vehicle power information. It may include the initial power at the start of the ride, the remaining power at the end of the ride, the power consumed during the ride, etc. The order information may also include the order completion status. For example, if the user rides to the end smoothly and locks the vehicle, the vehicle borrowing order is in the completed state. If the user does not ride to the end and the electric vehicle stops running due to too low power, the vehicle borrowing order is in the state of power-off during the ride.

[0060] It should be noted that the vehicle borrowing order may also include other information, such as time information (such as the start and end times of the order), cost information, battery life information (such as the battery life duration), etc.

[0061] The first power threshold is the power demarcation value that limits the non - borrowability of electric vehicles in the target space - time unit, and it can be used to determine whether an electric vehicle in the target space - time unit can be borrowed by a consumer user. When the current power of the vehicle is lower than the first power threshold, the battery swapping system can determine that the vehicle is in a non - borrowable state (for example, controlling the switch - locking device of the vehicle to remain locked to prevent the consumer user from borrowing the vehicle). The first power threshold can be represented in various forms. For example, it can be in the form of a percentage relative to the total battery power (such as 2%, 4%, etc.), or in other forms such as power values.

[0062] The first power threshold can be any suitable value such as 5%, 10%, 15, etc.

[0063] In some embodiments, the battery swapping system can determine the first power threshold based on the number of the first historical orders corresponding to the target space - time unit. For example, for a target space - time unit with a relatively large number of the first historical orders, the corresponding first power threshold can be set higher; otherwise, it can be set lower.

[0064] In some embodiments, the battery swapping system can also determine the first power threshold in combination with the first operation and maintenance level of the target space - time unit. The first operation and maintenance level can reflect the overall operation and maintenance ability of the target space - time unit (such as the scheduling of battery swapping personnel), and it can be determined based on the number of operation and maintenance personnel (or battery swapping personnel), business capabilities, battery swapping efficiency, etc. For a target space - time unit with a higher first operation and maintenance level, the corresponding first power threshold can be set lower; otherwise, the corresponding first power threshold can be set higher.

[0065] It should be noted that the battery swapping system can dynamically adjust the first power threshold according to the order level and / or the first operation and maintenance level of the target space - time unit to meet the actual situation of user needs and operation and maintenance scheduling management.

[0066] In some embodiments of this specification, by considering the order level of the target space - time unit to determine the first power threshold, the utilization rate of borrowed vehicles can be improved, the loss of vehicle borrowing orders caused by being prematurely lower than the first power threshold can be avoided, and at the same time, the situation of power outage during the ride of consumer users can be reduced; in addition, by considering the first operation and maintenance level, the setting of the first power threshold can be made more in line with the actual situation of operation and maintenance capabilities.

[0067] In some embodiments, the battery swapping system may also determine the first power threshold according to other factors (such as the target category to which the target space-time unit belongs). For example, for a target space-time unit corresponding to a target category with a relatively high order level, the corresponding first power threshold may be set relatively high; otherwise, it may be set relatively low.

[0068] In some embodiments, the battery swapping system may determine the first power threshold based on the riding power-off rate of the target space-time unit. For more content on determining the first power threshold, see Figure 4a 、 Figure 4b and its description.

[0069] Step 320, divide the target space-time unit into multiple types of sub space-time units.

[0070] A sub space-time unit refers to a sub-unit within the target space-time unit that has a certain spatial range and a certain time range. It should be understood that the spatial range corresponding to the sub space-time unit is within the spatial range corresponding to the target space-time unit, and its corresponding time range is within the time range corresponding to the target space-time unit. In some embodiments, the battery swapping system may divide the spatial range corresponding to the target space-time unit into multiple smaller regional units (such as geographical grid units) through a Geographic Information System (GIS), and divide the time range corresponding to the target space-time unit into shorter time segments to generate multiple sub space-time units.

[0071] In some embodiments, the battery swapping system may also classify multiple sub space-time units within the target space-time unit. For example, clustering can be performed according to one or a combination of the geographical features (such as bustling areas, suburbs, etc.) and order features (such as quantity, borrowing frequency, mileage length) of the multiple sub space-time units to obtain multiple different types of sub space-time units. Among them, clustering can be performed based on algorithms such as the K-Means clustering algorithm (K-means algorithm) and the density-based clustering algorithm (DBSCAN algorithm). Different sub space-time units may have the same or different area sizes and shapes.

[0072] In some embodiments, the battery swapping system may classify multiple sub space-time units within the target space-time unit based on various classification metrics. For example, the classification metrics may include order level, population density, etc. Taking the order level as an example, the battery swapping system may set multiple order level intervals according to the order level of the target space-time unit (such as the daily average order volume, in units of ten thousand orders), such as (0, 2], (2, 4], (4, 8], (8, 12], (12, 16]... etc. Multiple sub space-time units with similar order levels can be used as sub space-time units of the same category.

[0073] Step 330: For each type of sub - space - time unit in the target space - time unit, based on the second historical order of the borrowed electric vehicles corresponding to the sub - space - time unit and the first power threshold, determine the second power threshold corresponding to the sub - space - time unit. The second power threshold is the power demarcation value that defines when the electric vehicles in the sub - space - time unit need to swap batteries.

[0074] The second historical order refers to the historical vehicle - borrowing orders of the electric vehicles within the sub - space - time unit, which is a subset of the first historical order. For example, the second historical order can include the set of vehicle - borrowing orders generated after the electric vehicles within the sub - space - time unit are borrowed by consumer users (such as renters) within a certain time period (such as one month, one week, etc.).

[0075] The second power threshold is the power demarcation value that defines when the electric vehicles in the sub - space - time unit need to swap batteries, and it can determine whether the electric vehicles within the sub - space - time unit need to be swapped by the operation and maintenance team (such as battery - swapping personnel). When the current power of the vehicle is lower than the second power threshold, the battery - swapping system can determine that the vehicle needs to swap batteries and instruct the operation and maintenance team to perform battery - swapping operations (such as replacing or charging the battery).

[0076] The representation form of the second power threshold is the same as that of the first power threshold. It can be in the form of a percentage relative to the total battery power (such as 6%, 8%), or in other forms such as power values.

[0077] In some embodiments, the second power threshold is greater than the first power threshold. For example, the battery - swapping system can increase a preset power threshold (such as 5%) based on the first power threshold to generate the second power threshold. In some embodiments, the battery - swapping system can set different second power thresholds for different types of sub - space - time units.

[0078] In some embodiments, the second power threshold can be determined according to the order level corresponding to the sub - space - time unit. The order level corresponding to the sub - space - time unit can be represented by the number of second historical orders corresponding to the sub - space - time unit (such as the average order volume). For sub - space - time units with a larger number of second historical orders, a higher second power threshold can be set.

[0079] In some embodiments, the battery - swapping system can also determine the second power threshold in combination with the second operation and maintenance level of the sub - space - time unit. The second operation and maintenance level can reflect the operation and maintenance capabilities of different sub - space - time units. For sub - space - time units with a higher second operation and maintenance level, the corresponding second power threshold can be set lower.

[0080] In some embodiments of the present specification, by dynamically adjusting the second power threshold according to the borrowing requirements of users in different sub - space - time units (such as the number of orders, borrowing frequency, etc.) and the second operation and maintenance level, the balance between the order level and the operation and maintenance requirements can be better coordinated, so that the borrowing requirements of users in the sub - space - time unit, the operation and maintenance pressure, and the cost reach a balance.

[0081] In some embodiments, the battery - swapping system can determine the second power threshold based on the low - power rate index of the sub - space - time unit. For more content on determining the second power threshold, see Figure 5a 、 Figure 5b and its description.

[0082] In some embodiments, the battery - swapping system can determine multiple target borrowing electric vehicles based on the battery power information of the borrowed electric vehicles in the sub - space - time unit, the first power threshold, and the second power threshold, and determine the battery - swapping strategy based on the position distribution information of the multiple target borrowing electric vehicles.

[0083] The target borrowing electric vehicle refers to one or more low - power vehicles in the sub - space - time unit, which may include a first target vehicle with a battery power lower than the first power threshold and a second target vehicle with a battery power lower than the second power threshold.

[0084] The battery - swapping strategy refers to a plan for guiding the operation and maintenance personnel to perform battery - swapping operations on multiple target borrowing electric vehicles, which may include but are not limited to the battery - swapping priority and the battery - swapping route. Among them, the battery - swapping priority of the first target vehicle can be higher than that of the second target vehicle.

[0085] In some embodiments, the battery - swapping system can determine multiple target battery - swapping vehicles according to the first position distribution information corresponding to the multiple first target vehicles and the second position distribution information corresponding to the multiple second target vehicles. The multiple target battery - swapping vehicles can be within the preset area range (such as 1 km) of the operation and maintenance personnel, and they can be a combination of multiple first target vehicles and second target vehicles.

[0086] The battery - swapping route can be used to indicate the battery - swapping order of each target battery - swapping vehicle among the multiple target battery - swapping vehicles, and it can be in the form of a map route. In some embodiments, the battery - swapping order of the target battery - swapping vehicle can be determined based on the battery - swapping priority of the target battery - swapping vehicle and the number of supportable orders of the target battery - swapping vehicle. The higher the battery - swapping priority, the earlier the battery - swapping order. When the number of supportable orders of the target battery - swapping vehicle decreases to the preset quantity threshold (such as 1), its battery - swapping order can be adjusted to be earlier. For relevant content on the number of supportable orders, see Figure 5a and its description.

[0087] It should be noted that the battery - swapping system can send the battery - swapping strategy to the terminal device of the target operation and maintenance personnel (such as terminal 130), which includes but is not limited to one or a combination of forms such as text, sound, and map route.

[0088] In some embodiments of this specification, by monitoring multiple borrowed electric vehicles within a sub - space - time unit, the target borrowed electric vehicle can be obtained in a timely manner, and a battery - swapping strategy can be generated to perform timely battery - swapping processing on the target borrowed electric vehicle, improving the operation efficiency and meeting the borrowing needs of consumer users.

[0089] In some embodiments of this specification, by setting a first power threshold, the overall power - off rate of riding in the target space - time unit can be reduced. At the same time, different first power thresholds are set for target space - time units of different target types, making the first power threshold more in line with the actual needs of the borrowed vehicles. In addition, by setting a second power threshold for different sub - space - time units within the target space - time unit, the actual needs and operation and maintenance capabilities of the borrowed vehicles in each sub - space - time unit can be distinguished more finely. While reducing the low - power rate of the borrowed electric vehicles, the usage efficiency and operation and maintenance efficiency of the vehicles are improved.

[0090] Figure 4a It is an exemplary flowchart of a method for determining a first power threshold shown in some embodiments of this specification.

[0091] In some embodiments, process 400 can be executed by a battery - swapping system. As Figure 4a shown, process 400 includes the following steps.

[0092] Step 410, divide the first historical order into multiple first order sets, and each first order set includes first historical orders corresponding to the corresponding historical first power threshold.

[0093] The historical first power threshold refers to the first power threshold (i.e., the power demarcation value at which the electric vehicle cannot be borrowed) that has been set in the target space - time unit. During operation, the battery - swapping system or operation and maintenance personnel may set different historical first power thresholds according to the actual situation of the target space - time unit. Only as an example, if the first power threshold of the target space - time unit has been set to 6%, 5%, and 4%, then the historical first power thresholds include 6%, 5%, and 4%.

[0094] When a certain historical first power threshold is set in the target space - time unit, the historical orders generated in this target space - time unit can be regarded as first historical orders corresponding to this historical first power threshold, and these orders form a first order set corresponding to this historical first power threshold. Different historical first power thresholds correspond to different first order sets. For example, if the historical first power threshold set in a certain historical time period T1 (such as January 2023) of the target space - time unit is 6%, and N1 first historical orders are generated in this target space - time unit during this period, then these N1 first historical orders form a first order set corresponding to the historical first power threshold of 6%. According to this principle, different historical time periods T1, T2... Tk They can respectively correspond to different historical first power thresholds U1, U2... U k , and respectively correspond to different first order sets O1, O2... O k .

[0095] Step 420, for each first order set, determine a first data sample, where the first data sample includes at least the riding power-off rate and the optimal mileage of the first historical order in the first order set.

[0096] The first data sample can include characteristic parameters determined based on relevant information of the first historical order in the first order set. One first order set corresponds to one first data sample.

[0097] The riding power-off rate refers to the proportion of riding power-off orders in the first order set. Among them, a riding power-off order refers to an order in which the battery runs out of power during the riding of an electric vehicle by a consumer user, resulting in the interruption of the journey. In some embodiments, the battery swapping system can determine a riding power-off order based on the remaining power information of the electric vehicle (such as the power dropping to 0), or according to the feedback information obtained from the consumer user (such as complaints and reports from the consumer user).

[0098] It can be understood that the greater the riding mileage expected by the consumer user, the greater the battery power consumption, and the greater the probability of riding power-off. For different historical first power thresholds, the riding power-off rates of the corresponding first order sets are different.

[0099] The optimal mileage refers to the minimum mileage that the borrowed electric vehicle can ride estimated based on the historical first power threshold corresponding to the first order set. In some embodiments, the optimal mileage can be equal to where U is the historical first power threshold corresponding to the first order set, and Q Mi is the power consumption value per unit mileage. In some embodiments, the optimal mileage can be equal to where e is a preset error value. For descriptions of the power consumption value per unit mileage and the preset error value, please refer to the following text.

[0100] In some embodiments, the battery swapping system can determine the optimal mileage corresponding to each target first data sample based on the mileage distribution information of the target first order set corresponding to each target first data sample.

[0101] The mileage distribution information can reflect the riding mileage preferences of the consuming users in the target spatio-temporal unit. It may include mileage intervals and their corresponding order quantiles. Among them, the order quantile can be determined based on the proportion (such as 5%) of the number of second historical orders falling within the mileage interval in the second order set. For example, the mileage distribution information may include: the mileage interval is from 500m to 800m, and its corresponding order quantile is 5%; the mileage interval is from 800m to 1.5km, and its corresponding order quantile is 8%; the mileage interval is from 3km to 5km, and its corresponding order quantile is 70%; the remaining mileage intervals are above 5km, above 8km, etc. The above mileage distribution information indicates that in the target spatio-temporal unit, most (such as 70%) of the consuming users' riding demands are within 5km, so the optimal mileage is 5km, and the optimal order quantile is 5% + 8% + 70% = 83% (or 0.83).

[0102] In some embodiments of the present specification, by using the mileage distribution information to determine the optimal mileage, the optimal mileage can be made more in line with the borrowing needs of most consuming users in the target spatio-temporal unit, so that the first power threshold determined subsequently can better meet the balance between the riding needs and operation and maintenance needs of consuming users.

[0103] In some embodiments, the first data sample further includes at least one of the power consumption value per unit mileage, the preset error value, and the optimal order quantile.

[0104] The power consumption value per unit mileage refers to the power consumption value that the battery of an electric vehicle needs to consume per unit mileage (such as per km). In some embodiments, the power consumption value per unit mileage can be a preset value, or a value determined by the user according to experience. In some embodiments, when the first historical order is related to an electric vehicle equipped with a battery of the target model, the power consumption value per unit mileage can be determined according to the target model. For example, based on the total power consumption and the total mileage of all first historical orders, the power consumption value per unit mileage can be calculated. For another example, the power consumption value per unit mileage can be an empirical value determined according to the target model. At this time, the power consumption values per unit mileage in multiple first data samples are equal.

[0105] In some embodiments, the battery swapping system can also determine the power consumption value per unit mileage corresponding to each first order set according to the ratio of the total power consumption corresponding to different first order sets to the total mileage of riding. At this time, the power consumption values per unit mileage in multiple first data samples may be equal or unequal.

[0106] The preset error value is the power error value. Since the actual power of the electric vehicle may be less than the power value reported to the battery swapping system, a preset error value is set to correct the error of power reporting.

[0107] In some embodiments, the preset error value can be a preset value or a value determined by the user according to experience. The preset error value can also be determined according to the target model of the battery. In some embodiments, the preset error value is related to the battery usage information, and the battery usage information includes the average number of charging times and the usage duration. Among them, the usage condition of the battery affects the performance of the battery (such as the decrease in battery life). The more the charging times and the longer the usage duration of the borrowed electric vehicle placed in the target space-time unit, the larger the preset error value can be set. In some embodiments, the preset error values in multiple first data samples are equal.

[0108] In some embodiments, the battery swapping system can dynamically adjust the preset error value. For example, as the operation time increases, the average number of charging times and the usage duration of the batteries of multiple borrowed vehicles in the target space-time unit can be dynamically counted to dynamically determine the preset error value.

[0109] In some embodiments of this specification, considering the battery usage information of the borrowed vehicles in different target space-time units to determine the preset error value can make the subsequently determined first power threshold more accurate.

[0110] The optimal order percentile refers to the proportion of the first historical orders in the first order set whose riding mileage is less than the optimal mileage. Only as an example, when the optimal mileage corresponding to the first historical order set is 5 kilometers, the optimal order percentile is the ratio of the number of the first historical orders with a mileage less than 5 kilometers to the total number of orders in the first order set.

[0111] Step 430, based on the first data sample corresponding to each first order set, determine the first power threshold.

[0112] In some embodiments, the battery swapping system can determine the first power threshold based on the following steps 431 to 433. Or, the battery swapping system can be based on Figure 4b the shown process to determine the first power threshold.

[0113] Step 431, based on the riding power-off rate corresponding to each first data sample, determine multiple target first data samples that meet the preset power-off rate condition from multiple first data samples.

[0114] In some embodiments, the preset power-off rate condition may be that the riding power-off rate is less than a preset power-off rate threshold. For example, when the riding power-off rate corresponding to a certain first data sample is less than the preset power-off rate threshold, then the first data sample is a target first data sample. The preset power-off rate threshold is the maximum value of the acceptable riding power-off rate. For example, 0.03%, 0.05%, 0.1%. When the riding power-off rate is too large, the user experience will be greatly affected, indicating that the set historical first power threshold is not very reasonable at this time. The riding power-off rate of the target first data samples screened based on the preset power-off rate condition meets the requirements, indicating that the corresponding historical first power threshold is relatively reasonable.

[0115] Step 432: Determine the target optimal mileage corresponding to the target spatio-temporal unit based on the optimal mileage of each target first data sample among the multiple target first data samples.

[0116] The target optimal mileage refers to the optimal mileage corresponding to the overall target spatio-temporal unit, which can be determined based on the optimal mileage of multiple target first data samples.

[0117] In some embodiments, the power exchange system may determine the average value of the optimal mileage of multiple target first data samples as the target optimal mileage. In some embodiments, the power exchange system may determine the target optimal mileage based on the average value and standard deviation of the optimal mileage of multiple target first data samples. For example, the target optimal mileage can be calculated based on the following formula (1). S tat = avg(s) + m1 * dev(s) Formula (1) Where, S in formula (1) tat represents the target optimal mileage; avg(s) represents the average value of the optimal mileage of multiple target first data samples; dev(s) represents the standard deviation of the optimal mileage of multiple target first data samples, which can reflect the fluctuation or deviation between multiple optimal mileages; m1 is a preset coefficient, which can be set as a value within the interval [0, 3].

[0118] Step 433: Determine the first power threshold based on the target optimal mileage.

[0119] In some embodiments, the power exchange system may determine the first power threshold based on the following formula (2). U = S tat *Q Mi + e Formula (2)

[0120] Where, U in formula (2) represents the first power threshold of the target spatio-temporal unit; S tat represents the target optimal mileage corresponding to the target spatio-temporal unit; Q Mirepresents the power consumption value per unit mileage; e represents a preset error value. In some embodiments, the preset error value in formula (2) can be omitted.

[0121] Figure 4b is a schematic diagram of the process for determining the first power threshold based on the first machine learning model according to some embodiments of this specification.

[0122] As Figure 4b shown, the battery swapping system can generate the first machine learning model 440 based on the first data sample 441 through a machine learning algorithm. Specifically, the riding power-off rate in the first data sample 441 can be used as the training input, and the optimal mileage in the first data sample 441 can be used as the training label. Optionally, the power consumption value per unit mileage and / or the preset error value in the first data sample 441 can also be used as the training label. Optionally, the optimal order quantile in the first data sample 441 can also be used as the training label.

[0123] Merely as an example, the first data sample 441 can be represented by the first feature vector (Q Mi , L, s, e, w1), where Q Mi represents the power consumption value per unit mileage; L represents the optimal order quantile; s represents the optimal mileage; e represents the preset error value; w1 represents the riding power-off rate corresponding to this group of the first data samples. Among them, Q Mi , e, and w1 are the training inputs, and L and s are the training labels.

[0124] The first machine learning model can refer to a model used to predict the optimal mileage (or the optimal mileage and the optimal order quantile). In some embodiments, the first machine learning model can include a Bayesian network model, a Gaussian Process Regression (GPR) model, a Support Vector Machine (SVM) model, or a model constructed by other machine learning algorithms.

[0125] In some embodiments, the first initial model can be iteratively trained to obtain a trained initial first power threshold prediction model. The training input can be input into the first initial model, and the first initial model can output the predicted optimal mileage (or the predicted optimal mileage and the optimal order quantile) corresponding to each training sample. Based on the difference between the model output and the training label, the value of the loss function can be determined. The parameters of the first initial model can be iteratively updated based on the value of the loss function until the training termination condition is met (such as, the loss function converges, a specific number of iterations are performed, etc.). The updated first initial model can be used as the trained first machine learning model.

[0126] Continue to refer to Figure 4b, after obtaining the first machine learning model 440, the battery swapping system can determine the target optimal mileage 442 based on the first machine learning model 440. For example, the riding power-off rates of all the first historical orders can be input into the first machine learning model 440, and the first machine learning model 440 outputs the target optimal mileage 442. As another example, the riding power-off rates of all the first historical orders, the power consumption value per unit mileage of the target spatio-temporal unit, and a preset error value can be input into the first machine learning model 440, and the first machine learning model 440 outputs the target optimal mileage 442.

[0127] Furthermore, the battery swapping system can determine a first power threshold 443 corresponding to the target spatio-temporal unit based on the target optimal mileage 442. For example, the first power threshold 443 can be calculated based on formula (2).

[0128] In some embodiments of this specification, determining the first power threshold based on the first order set under different historical first power thresholds can combine the actual situation of the historical adjustment process, making the determined first power threshold more accurate. At the same time, considering the riding power-off rate during the determination of the first power threshold can improve the utilization rate of the battery power on the basis of enhancing the borrowing experience of consumer users within the target spatio-temporal unit.

[0129] Figure 5a is an exemplary flowchart of a method for determining a second power threshold shown in some embodiments of this specification.

[0130] In some embodiments, process 500 can be executed by the battery swapping system. As Figure 5a shown, process 500 includes the following steps.

[0131] Step 510, for each type of sub-spatio-temporal unit, divide the second historical orders of the sub-spatio-temporal unit into multiple second order sets, and each second order set includes the second historical orders corresponding to the corresponding historical second power threshold.

[0132] The historical second power threshold refers to the second power threshold (i.e., the power demarcation value at which the electric vehicle needs to swap the battery) that has been set for the sub-spatio-temporal unit. During operation, the battery swapping system or the operation and maintenance personnel may set different historical second power thresholds according to the actual situation of the sub-spatio-temporal unit. Only as an example, if the second power threshold of the sub-spatio-temporal unit has been set to 12%, 10%, and 8%, then the historical second power thresholds include 12%, 10%, and 8%.

[0133] When a certain historical second power threshold is set in a sub - spacetime unit, the historical orders generated in this sub - spacetime unit can be regarded as second historical orders corresponding to this historical second power threshold, and these orders form a second order set corresponding to this historical second power threshold. Different historical second power thresholds correspond to different second order sets. Different historical time periods t1, t2... t k can respectively correspond to different historical second power thresholds C1, C2... C p , and respectively correspond to different second order sets o1, o2... o p .

[0134] Step 520, for each second order set, determine a second data sample. The second data sample includes at least the vehicle low - power rate and the optimal order number corresponding to the second historical orders in the second order set. The vehicle low - power rate is determined based on the first power threshold, and the optimal order number is determined based on the first power threshold and the historical second power threshold corresponding to the second order set.

[0135] The second data sample can include characteristic parameters determined based on the relevant information of the second historical orders in the second order set. One second order set corresponds to one second data sample.

[0136] The vehicle low - power rate refers to the proportion of low - power orders in the second order set. A low - power order refers to an order in which the power of the electric vehicle is lower than a preset low - power threshold when the consuming user ends the order (such as when locking the car). The preset low - power threshold can be the system default value or set by the user. In some embodiments, the preset low - power threshold is the first power threshold.

[0137] The optimal order number can refer to the number of orders that can be supported by the power value between the second power threshold and the first power threshold. The number of orders that can be supported can be used to reflect the number of borrowing orders generated by one or more consuming users borrowing vehicles with the power values of the second power threshold and the first power threshold.

[0138] In some embodiments, the optimal order number can be determined based on the first power threshold and the historical second power threshold corresponding to the second order set. Only as an example, the optimal order number can be equal to where C refers to the historical second power threshold, U refers to the first power threshold, k represents the average mileage of all the second historical orders in the second order set; Q Mi represents the power consumption value per unit mileage corresponding to the sub - spacetime unit.

[0139] In some embodiments, the second data sample may further include the average daily order volume and the average riding mileage of the order. The average daily order volume refers to the average daily order volume of each electric vehicle in the sub - space - time unit (e.g., 50 orders per day); the average riding mileage of the order refers to the average value of the mileage of all orders in the second order set. Step 530: Based on the second data sample corresponding to each second order set, determine the second power threshold corresponding to the sub - space - time unit.

[0140] In some embodiments, the battery swapping system may determine the second power threshold based on the following steps 531 to 533. Alternatively, the battery swapping system may determine the second power threshold based on Figure 5b the process shown.

[0141] Step 531: Based on the vehicle low - power rate corresponding to each second data sample, determine multiple target second data samples that meet the preset low - power rate condition from multiple second data samples.

[0142] In some embodiments, the preset low - power rate condition may be that the vehicle low - power rate is less than the preset low - power rate threshold, or the vehicle low - power rate is within the preset low - power rate threshold range. For example, when the vehicle low - power rate corresponding to a certain second data sample is less than the preset low - power rate threshold, or within the preset low - power rate threshold range, then the second data sample is a target second data sample. When the vehicle low - power rate is too large, the user experience will be greatly affected, indicating that the set historical second power threshold is not very reasonable at this time. The vehicle low - power rate of the target second data samples selected based on the preset vehicle low - power rate meets the requirements, indicating that the corresponding historical second power threshold is more reasonable.

[0143] Step 532: Based on the optimal order number of each target second data sample among multiple target second data samples, determine the target optimal order number corresponding to the sub - space - time unit.

[0144] The target optimal order number refers to the optimal order number corresponding to the whole sub - space - time unit, which can be determined based on the optimal order numbers of multiple target second data samples of the sub - space - time unit.

[0145] In some embodiments, the battery swapping system may determine the average value of the optimal order numbers of multiple target second data samples as the target optimal order number. In some embodiments, the battery swapping system may determine the target optimal order number based on the average value and standard deviation of the optimal order numbers of multiple target second data samples, which can be calculated based on the following formula (3): D tat = avg(d) + m2*dev(d) Formula (3)

[0146] Where D in formula (3) tatrepresents the target optimal order quantity; avg(d) represents the average value of the optimal order quantities of multiple target second data samples; dev(d) represents the standard deviation of the optimal order quantities of multiple target first data samples, which can reflect the fluctuation or deviation among multiple optimal order quantities; m2 is a preset coefficient, which can be set to a value within the interval [0, 3].

[0147] In some embodiments, the battery swapping system can also adjust the second power threshold during operation based on the preset optimal order quantity index, and then correct the target optimal order quantity so that the target optimal order quantity meets the requirements of the preset optimal order quantity index. Among them, the preset optimal order quantity index can be in the form of a preset optimal order quantity threshold range. Only as an example, the preset optimal order quantity index can be [2, 4], which means that the target optimal order quantity should be between 2 orders and 4 orders.

[0148] It can be understood that the larger the second power threshold is set, the greater the power difference between the second power threshold and the first power threshold, and the larger the potential supportable order quantity (target optimal order quantity). This indicates that on the one hand, the second power threshold meets the riding needs of consumer users, but on the other hand, it also indicates that the possibility of battery power waste is higher.

[0149] If the second power threshold is set smaller, the power difference between the second power threshold and the first power threshold is also smaller, and the potential supportable order quantity is smaller. This indicates that on the one hand, the risk of riding power outage is greater, and the riding needs of consumer users cannot be met. At the same time, the risk that the vehicle becomes a low-power vehicle (such as when the power is lower than the first power threshold, the vehicle can no longer be borrowed) is also greater.

[0150] When the target optimal order quantity meets the requirements of the preset optimal order quantity index (such as within the preset optimal order quantity threshold range), it indicates that the second power threshold is better for the riding needs and operation and maintenance effects of users in the sub-time-space unit; when the target optimal order quantity is lower than the lower limit of the preset optimal order quantity threshold range (such as 2), it indicates that the riding needs of users are tense or the operation and maintenance pressure is too high, and the battery swapping system can increase the second power threshold to increase the potential supportable order quantity; when the target optimal order quantity is higher than the upper limit of the preset optimal order quantity threshold range (such as 4), it indicates that the utilization rate of battery power is not ideal, and the battery swapping system can lower the second power threshold.

[0151] In some embodiments of this specification, by setting the preset optimal order quantity index to correct the target optimal order quantity, the finally determined second power threshold makes the riding needs of consumer users in the sub-time-space unit and the utilization rate of battery power more balanced.

[0152] Step 533, determine the second power threshold based on the target optimal order quantity.

[0153] In some embodiments, the battery swapping system may determine the second power threshold based on formula (4) shown below. C = U + D tat * k * Q Mi Formula (4)

[0154] Wherein, C in formula (4) represents the second power threshold corresponding to the sub - spacetime unit; U represents the first power threshold corresponding to the target spacetime unit to which the sub - spacetime unit belongs; D tat represents the target optimal order number; k represents the average mileage of all the second historical orders in the second order set corresponding to this sub - spacetime unit; Q Mi represents the power consumption value per unit mileage.

[0155] Figure 5b is a schematic diagram of the process of determining the second power threshold based on the second machine learning model according to some embodiments of this specification.

[0156] The second machine learning model may refer to a model for predicting the target optimal mileage, and it can be a trained machine learning model. In some embodiments, the second machine learning model may include a Bayesian network model, a Gaussian process regression (GPR) model, and a support vector machine (SVM) model or a model constructed by other machine learning algorithms.

[0157] As Figure 5b shown, the battery swapping system may generate the second machine learning model 540 based on the second data sample 541 through a machine learning algorithm. Specifically, the low - power rate of the vehicle in the second data sample 541 may be used as the training input, and the optimal order number in the second data sample 541 may be used as the training label.

[0158] In some embodiments, the second data sample 541 may be represented based on the second feature vector (Q Mi , K, ord, d, w2), wherein, the element Q Mi in this second feature vector represents the power consumption value per unit mileage; the element K represents the average riding mileage of the orders corresponding to the second data sample 541, which can be determined based on the average mileage of all the second historical orders in the second order set; the element ord represents the daily average order volume of each electric vehicle in the sub - spacetime unit corresponding to the second data sample 541; the element d represents the optimal order number; w2 represents the low - power rate of the vehicle corresponding to this group of second data samples. Among them, Q Mi , K, ord and w2 are training inputs, and d is used as the training label.

[0159] In some embodiments, the initial second machine learning model can be iteratively trained to obtain a trained second machine learning model. The training method is similar to that of the first machine learning model and will not be elaborated here.

[0160] Continuing to refer to Figure 5b , after obtaining the second machine learning model 540, the battery swapping system can determine the target optimal order number 542 based on the second machine learning model 540. For example, the low battery rate of all the second historical orders can be input into the second machine learning model 540, and the second machine learning model 540 outputs the target optimal order number 542. Another example is that the low battery rate of all the second historical orders, the average daily order volume of the sub - space - time unit, the power consumption value per unit mileage, and the average value of the mileage can be input into the second machine learning model 540, and the second machine learning model 540 outputs the target optimal order number 542.

[0161] Furthermore, the battery swapping system can determine the second power threshold 543 corresponding to the sub - space - time unit based on the target optimal order number. For example, the second power threshold 543 can be calculated based on formula (4).

[0162] In some embodiments of this specification, determining the second power threshold based on the second order set under different historical second power thresholds makes the determination result of the second power threshold more accurate; at the same time, considering the evaluation index of the vehicle's low battery rate makes the consumption user requirements and operation and maintenance requirements of the sub - space - time unit more balanced; in addition, adjusting the second power threshold according to the change of the supportable order number improves the utilization rate of the battery power and also makes the second power threshold more accurate.

[0163] It should be noted that the above description of the process is only for illustration and explanation, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the process under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0164] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only an example and does not constitute a limitation to this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are proposed in this specification, so such modifications, improvements, and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0165] In the meantime, this specification uses specific terms to describe the embodiments of this specification. For example, "an embodiment", "one embodiment", and / or "some embodiments" mean a certain feature, structure, or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that the "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more at different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0166] In addition, unless clearly stated in the claims, the order of the processing elements and sequences, the use of numerical letters, or the use of other names in this specification are not used to limit the order of the processes and methods in this specification. Although some currently considered useful embodiments of the invention are discussed through various examples in the above disclosure, it should be understood that such details only serve the purpose of illustration. The appended claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that conform to the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only through software solutions, such as installing the described system on existing servers or mobile devices.

[0167] Similarly, it should be noted that, in order to simplify the expression of the disclosure in this specification and thus help the understanding of one or more embodiments of the invention, in the previous description of the embodiments of this specification, sometimes multiple features are merged into one embodiment, drawing, or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than those mentioned in the claims. In fact, the features of the embodiments are fewer than all the features of the individual embodiments disclosed above.

[0168] In some embodiments, numbers are used to describe the components and the quantity of attributes. It should be understood that such numbers used for the description of embodiments are modified by the modifiers "about", "approximate", or "substantially" in some examples. Unless otherwise stated, "about", "approximate", or "substantially" indicate that the said numbers allow a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, and these approximate values can change according to the characteristics required by individual embodiments. In some embodiments, the numerical parameters should consider the specified significant digits and adopt the method of retaining the general number of digits. Although the numerical ranges and parameters used in some embodiments of this specification to confirm the breadth of their ranges are approximate values, in specific embodiments, such numerical settings are as precise as possible within the feasible range.

[0169] For each patent, patent application, patent application publication, and other materials cited in this specification, such as articles, books, specifications, publications, documents, etc., the entire content thereof is hereby incorporated by reference into this specification. Except for the application history files that are inconsistent with or conflict with the content of this specification, and also except for the files that limit the broadest scope of the claims of this specification (currently or subsequently appended to this specification). It should be noted that if there are any inconsistencies or conflicts between the descriptions, definitions, and / or uses of terms in the supplementary materials of this specification and the content described in this specification, the descriptions, definitions, and / or uses of terms in this specification shall prevail.

[0170] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered to be consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A method for determining the battery swapping threshold of an electric vehicle, characterized in that Including: Based on the first historical order of the borrowed electric vehicle corresponding to the target spatio-temporal unit, determining a first power threshold corresponding to the target spatio-temporal unit, where the first power threshold is a power demarcation value that defines that the electric vehicle in the target spatio-temporal unit cannot be borrowed; Dividing the target spatio-temporal unit into multiple types of sub-spatio-temporal units; And For each type of sub-spatio-temporal unit of the target spatio-temporal unit, based on the second historical order of the borrowed electric vehicle corresponding to the sub-spatio-temporal unit and the first power threshold, determining a second power threshold corresponding to the sub-spatio-temporal unit, where the second power threshold is a power demarcation value that defines that the electric vehicle in the sub-spatio-temporal unit needs to change the battery.

2. The method according to claim 1, characterized in that, The determining the first power threshold corresponding to the target spatio-temporal unit includes: Dividing the first historical order into multiple first order sets, where each first order set includes the first historical orders corresponding to the historical first power threshold; For each first order set, determining a first data sample, where the first data sample at least includes the riding power-off rate and the optimal mileage of the first historical orders in the first order set, and the optimal mileage is determined based on the historical first power threshold corresponding to the first order set; Based on the first data sample corresponding to each first order set, determining the first power threshold.

3. The method according to claim 2, wherein The determining the first power threshold based on the first data sample corresponding to each first order set includes: Based on the riding power-off rate corresponding to each first data sample, determining multiple target first data samples that meet the preset power-off rate condition from the multiple first data samples; Based on the optimal mileage of each target first data sample in the multiple target first data samples, determining the target optimal mileage corresponding to the target spatio-temporal unit; Based on the target optimal mileage, determining the first power threshold.

4. The method according to claim 2, wherein The determining the first power threshold based on the first data sample corresponding to each first order set further includes: Using the riding power-off rate corresponding to the first data sample as the training input and the optimal mileage corresponding to the first data sample as the training label to generate a first machine learning model; Based on the first machine learning model, determining the target optimal mileage corresponding to the target spatio-temporal unit; Based on the target optimal mileage, determining the first power threshold.

5. The method according to claim 2, wherein The first data sample further includes the power consumption value per unit mileage and the preset error value.

6. The method according to claim 1, characterized in that, For each type of sub-spatio-temporal unit of the target spatio-temporal unit, determining the second power threshold corresponding to the sub-spatio-temporal unit includes: For each type of sub-spatio-temporal unit, Dividing the second historical order of the sub-spatio-temporal unit into multiple second order sets, where each second order set includes the second historical orders corresponding to the historical second power threshold; For each second order set, determining a second data sample, where the second data sample at least includes the corresponding vehicle low-power rate and the optimal order number of the second historical orders in the second order set, the vehicle low-power rate is determined based on the first power threshold, and the optimal order number is determined based on the first power threshold and the historical second power threshold corresponding to the second order set; and Determine a second power threshold corresponding to the sub - spatio - temporal unit based on the second data samples corresponding to each of the second order sets.

7. The method according to claim 6, wherein The determining the second power threshold corresponding to the sub - spatio - temporal unit based on the second data samples corresponding to each of the second order sets includes: Based on the vehicle low - power rate corresponding to each of the second data samples, determine a plurality of target second data samples that meet the preset low - power rate condition from the plurality of second data samples; Based on the optimal order number of each of the target second data samples among the plurality of target second data samples, determine a target optimal order number corresponding to the sub - spatio - temporal unit; Based on the target optimal order number, determine the second power threshold.

8. The method according to claim 6, wherein The determining the second power threshold corresponding to the sub - spatio - temporal unit based on the second data samples corresponding to each of the second order sets further includes: Using the vehicle low - power rate corresponding to the second data sample as a training input and the optimal order number corresponding to the second data sample as a training label, generate a second machine - learning model; Based on the second machine - learning model, determine the target optimal order number corresponding to the target spatio - temporal unit; Based on the target optimal order number, determine the second power threshold.

9. A system for determining the battery swapping threshold of an electric vehicle, characterized in that, Including: A first determination module, configured to determine a first power threshold corresponding to the target spatio - temporal unit based on the first historical orders of the borrowed electric vehicles corresponding to the target spatio - temporal unit, where the first power threshold is a power demarcation value that limits the non - borrowability of the electric vehicles in the target spatio - temporal unit; A spatio - temporal unit management module, configured to divide the target spatio - temporal unit into multiple types of sub - spatio - temporal units; And A second determination module, configured to, for each type of the sub - spatio - temporal units of the target spatio - temporal unit, determine a second power threshold corresponding to the sub - spatio - temporal unit based on the second historical orders of the borrowed electric vehicles corresponding to the sub - spatio - temporal unit and the first power threshold, where the second power threshold is a power demarcation value that limits the need for battery replacement of the electric vehicles in the sub - spatio - temporal unit.

10. A device for determining the battery swapping threshold of an electric vehicle, characterized in that, Including a processor, where the processor is configured to execute the method for determining the electric vehicle battery - swapping threshold according to any one of claims 1 to 8.

Images