A method and apparatus for predicting the number of purchases of a pure electric vehicle

Abstract

The application relates to the technical field of electric digital processing, in particular to a pure electric vehicle replacement quantity prediction method and device. The method comprises the following steps: for each pure electric vehicle in a pure electric vehicle set, acquiring user charging behavior data, average temperature of a region where the vehicle is located and current battery parameters; determining an actual attenuation rate of the battery according to the current battery parameters; determining a remaining service life of the battery according to the actual attenuation rate and a current health degree of the battery; determining a charging index according to the user charging behavior data; determining a replacement confidence degree of each pure electric vehicle according to the charging index, the remaining service life of the battery and the average temperature of the region where the vehicle is located; and predicting a replacement quantity of the pure electric vehicles in the pure electric vehicle set according to the replacement confidence degree of each pure electric vehicle. The application scientifically and reasonably predicts the replacement quantity of the pure electric vehicles based on the user charging behavior data, the battery parameters and the temperature.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital processing, in particular to a method and device for predicting the number of pure electric vehicle replacement. BACKGROUND

[0002] Pure electric vehicle replacement refers to that a vehicle owner replaces the currently used pure electric vehicle with a new pure electric vehicle. In the current environment, the most important factor affecting the number of pure electric vehicle replacement is the battery mileage. The present application focuses on the replacement behavior caused by the battery mileage decay.

[0003] At present, professional institutions generally detect the mileage of pure electric vehicles at regular intervals, and suggest replacement when the reduction rate of the mileage is lower than a certain threshold. However, this method requires the detection institution to conduct professional detection on the vehicle, which is time-consuming and laborious, and it is difficult to organize all pure electric vehicle owners for detection, so it is impossible to predict the number of pure electric vehicle replacement in the future.

[0004] Therefore, the present application is proposed. SUMMARY

[0005] The present application aims to provide a method and device for predicting the number of pure electric vehicle replacement, based on user charging behavior data, battery parameters and temperature, to scientifically and reasonably predict the number of pure electric vehicle replacement.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a method for predicting the number of pure electric vehicle replacement, comprising:

[0008] For each pure electric vehicle in the set of pure electric vehicles, obtain user charging behavior data, average temperature of the area where the vehicle is located, and current battery parameters; wherein the current battery parameters at least include: current health degree of the battery, initial capacity, activation energy of battery decay, pre-exponential factor, average battery charging temperature and current full charging capacity;

[0009] According to the current battery parameters, determine the actual decay rate of the battery;

[0010] According to the actual decay rate and the current health degree of the battery, determine the remaining life of the battery;

[0011] According to the user charging behavior data, determine the charging index; wherein the user charging behavior data includes: the number of fast charging of the battery, the total number of charging, the battery charging temperature, the optimal charging temperature range, the number of overcharging of the battery, and the initial and ending power of the battery; the charging index includes: the proportion of the number of fast charging of the battery, the charging temperature adaptation degree, the overcharging frequency and the charging depth.

[0012] determining a replacement confidence of each pure electric vehicle according to the charging index, the remaining life of the battery, and the average temperature of the area where the vehicle is located;

[0013] predicting the replacement quantity of the pure electric vehicles in the set of pure electric vehicles according to the replacement confidence of each pure electric vehicle.

[0014] In a second aspect, the present application provides an electronic device, comprising:

[0015] at least one processor, and a memory connected to the at least one processor in communication;

[0016] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned pure electric vehicle replacement quantity prediction method.

[0017] Compared with the prior art, the present application has the following beneficial effects:

[0018] 1. The prediction model provided by the present application realizes accurate and forward-looking prediction. Traditional methods rely on macro statistical data such as vehicle age and historical sales, and the prediction results are rough and lagging. The present application creatively combines user charging behavior data, the average temperature of the area where the vehicle is located, and the current battery parameters to construct an actual decay rate model consistent with the Arrhenius equation. This enables the prediction model to deeply reflect the scientific nature of battery performance degradation, thereby enabling earlier and more accurate prediction of replacement demand caused by serious reduction in vehicle range, greatly improving prediction accuracy.

[0019] 2. The present application realizes multi-dimensional, dynamic and personalized replacement risk assessment. Instead of considering battery life in isolation, the present application analyzes battery remaining life, charging indicators reflecting user usage habits (such as fast charging ratio and charging depth), and environmental temperature to determine the "replacement confidence". This scheme fully considers the performance differentiation of the same vehicle model under different users and different regions, and converts static vehicle data into dynamic and personalized replacement probability. This makes the prediction results more accurately reflect the replacement willingness in the complex real market. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the specific embodiments or prior art of the present application, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0021] Figure 1 is a flowchart of a method for predicting the number of pure electric vehicles to be replaced according to an embodiment of the present application;

[0022] Figure 2 is a structural schematic diagram of an electronic device provided by the present application. DETAILED DESCRIPTION

[0023] Exemplary embodiments of the present application are described herein below with reference to the accompanying drawings, in which various details are set forth to provide an understanding of the present application. It will be apparent to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Also, for the sake of brevity and clarity, descriptions of well-known functions and constructions are omitted herein.

[0024] The present application is described in further detail below with reference to the embodiments.

[0025] The present application provides a method for predicting the number of pure electric vehicles to be replaced. The method is applicable to predicting the number of pure electric vehicles to be replaced due to battery endurance degradation. The present embodiment does not consider the replacement behavior caused by social activities such as price subsidies and exemptions. Related personnel can consider the impact of social activities on the number of replacements on the basis of the method provided by the present application to make more accurate predictions, but this is not within the scope of protection of the present application.

[0026] Referring to Figure 1 The method provided by the present embodiment includes:

[0027] S110, for each pure electric vehicle in the set of pure electric vehicles, obtaining user charging behavior data, average temperature of the area where the vehicle is located, and current battery parameters.

[0028] The current battery parameters include at least the current health of the battery, the initial capacity, the activation energy of battery degradation, the pre-exponential factor, the average temperature of battery charging, and the current full charging capacity.

[0029] The set of pure electric vehicles should cover all pure electric vehicle brands, models, and ages (in years) on the market to obtain more general data.

[0030] The user charging behavior data is the charging behavior data of the user on the battery of the pure electric vehicle, which is a set of quantitative values. For example, it includes the number of times of fast charging of the battery, the total number of times of charging, the battery charging temperature, the optimal charging temperature range, the number of times of overcharging of the battery, and the initial and final power of the battery. The charging behavior data can be obtained through the battery management system (BMS) of the pure electric vehicle.

[0031] The average temperature of the area where the vehicle is located is the average temperature of the area where the vehicle is located in a year, for example, 20 degrees or 15 degrees.

[0032] The average temperature of the battery charge in the current battery parameter is the average temperature of the battery in the previous charge, which can be given by the BMS. The initial capacity is the total energy that the battery can store in a brand-new state, which is a benchmark for measuring its performance degradation. The current full charge capacity refers to the maximum capacity that the battery can actually be fully charged in the current state after experiencing multiple uses and aging. By comparing the current full charge capacity with the initial capacity, the current health of the battery in percentage can be directly calculated. The activation energy of battery degradation can be understood as the "energy threshold" that needs to be overcome for the battery degradation reaction to occur. The higher the activation energy, the more sensitive the degradation reaction is to temperature. The pre-exponential factor represents the frequency or probability of the occurrence of the degradation reaction under ideal conditions.

[0033] It should be noted that the present embodiment does not limit the number of users to which the pure electric vehicles in the pure electric vehicle set belong. One user can enjoy one vehicle, or multiple vehicles.

[0034] S120, determining the actual degradation rate of the battery according to the current battery parameter.

[0035] First, according to the initial capacity of the battery, the current full charge capacity and the service life of the vehicle, the basic degradation rate of the battery is obtained :

[0036] ;

[0037] Among them, is the initial capacity of the battery (obtainable according to the battery model), is the current full charge capacity of the battery (obtainable according to the vehicle-mounted BMS), is the service life of the vehicle (the number of registered days / 365). The charging temperature of the battery affects the degradation rate of the battery, so the influence of the average temperature of the battery charge on the degradation of the battery is quantified based on the Arrhenius equation (the Arrhenius equation is a core formula describing how the rate of a chemical reaction changes with temperature). Based on this, the temperature acceleration coefficient :

[0038] ;

[0039] ;

[0040] Among them, is the degradation rate when the battery charging temperature is , is the activation energy of battery degradation (which can be obtained according to the battery model of the vehicle model used by the user), is the gas constant (8.314 J / (mol·k)), is the pre-exponential factor (which can be obtained according to the battery model of the vehicle model used by the user), and e is the natural constant. is the average temperature of battery charging, is the reference temperature, which is usually 25℃.

[0041] The temperature acceleration coefficient is used to measure the degradation acceleration multiple of the actual charging temperature relative to the reference temperature, and the temperature effect is considered on the basis of the basic degradation rate of the battery, which is more in line with the real aging law of the battery. Based on this, according to the basic degradation rate of the battery and the temperature acceleration coefficient, the actual degradation rate of the battery is obtained

[0042]

[0043] is the actual degradation rate of the battery.

[0044] S130, according to the actual degradation rate and the current health degree of the battery, determining the remaining life of the battery.

[0045] In combination with the current health degree of the battery, the actual degradation rate of the battery, and the empirical model of battery life prediction, the remaining life of the battery is predicted as:

[0046]

[0047] is the health degree threshold, which is generally 80%, and this threshold is a threshold recognized by professional institutions, which considers that when the battery cannot meet the design requirements of vehicle endurance and safety; is the current health degree of the battery.

[0048] S140, determining the charging index according to the user charging behavior data.

[0049] This embodiment needs to calculate the remaining life of the battery, the charging index and the average temperature of the region for each pure electric vehicle. In this step, the charging index includes: the proportion of the number of times of fast charging of the battery, the charging temperature adaptation degree, the excessive charging frequency and the charging depth.

[0050] ​​​​​When the number of fast charging cycles is too high, the powerful current forces lithium ions to migrate violently between the electrodes, generating additional heat and continuously damaging the microstructure of the electrode materials. This can accelerate the abnormal growth of the solid electrolyte interphase (SEI) film and the formation of lithium dendrites, gradually reducing the maximum usable capacity—a fundamental indicator determining battery range. Charging at excessively low temperatures slows lithium ion movement and causes a surge in internal resistance, resulting in low charging efficiency and potentially leading to lithium metal deposition and permanent capacity damage. At excessively high temperatures, electrolyte decomposition and side reactions are drastically accelerated. According to the Arrhenius equation, the chemical aging rate increases exponentially, rapidly impacting the battery's range potential. Overcharging, especially maintaining a 100% charge for extended periods, means the battery's positive electrode material is continuously subjected to high voltage stress, disrupting the crystal structure's stability and accelerating electrolyte oxidation, thus reducing the battery's capacity retention rate. Frequent "deep cycling" from 0% to 100% subjectes the electrode materials to enormous volume expansion and contraction stress, accelerating physical fatigue and significantly reducing the actual driving range after a full charge.

[0051] Specifically, the percentage of fast charging cycles is obtained based on the number of fast charging cycles and the total number of charging cycles. :

[0052] ;

[0053] Where N is the total number of charging cycles. It refers to the number of times the battery can be fast charged.

[0054] Based on the battery charging temperature and the optimal charging temperature range, the charging temperature fit degree F is obtained, as shown in the following formula:

[0055] ;

[0056] ;

[0057] Where N is the total number of charging cycles. It is the charging temperature adaptation for the i-th charge. For the first Secondary battery charging temperature and These are the maximum and minimum values ​​for the optimal charging temperature range, which are determined according to the different battery models.

[0058] The overcharge frequency O is obtained based on the number of overcharges and the total number of charges of the battery.

[0059] ;

[0060] in, is the number of overcharging times, specifically the number of times that the battery is still connected to the charger for more than 1 hour after being charged to 100% SOC (State of Charge).

[0061] According to the battery power at the beginning and end of charging, the charging depth D is obtained:

[0062] ;

[0063] ;

[0064] wherein, is the battery power at the end of charging, is the battery power at the beginning of charging, is the charging depth of the i-th charging.

[0065] Optionally, after obtaining the charging depth and other charging indicators, in order to comprehensively consider the influence of multiple charging indicators on the endurance, the multiple charging indicators are processed:

[0066] Suppose there are M pure electric vehicles in the set, and there are 4 charging indicators (see the description above). Then there are charging indicators: , is the j-th charging indicator of the k-th pure electric vehicle. After standardizing , the standardized value of the charging indicator of each pure electric vehicle is obtained , is the standardized value of the j-th charging indicator of the k-th pure electric vehicle. Then, according to the standardized value of each charging indicator, the entropy value of each charging indicator is calculated to measure the information disorder degree of each charging indicator:

[0067] ;

[0068] ;

[0069] wherein, is the proportion of the j-th charging indicator of the k-th pure electric vehicle in all indicators, is the entropy value of the j-th charging indicator, is the standardized value of the j-th charging indicator of the k-th pure electric vehicle, is a very small value to avoid when is meaningless, and when , .

[0070] The entropy weight method is adopted to weight and sum each charging indicator to obtain the health degree. First, the weight of each charging indicator is calculated :

[0071] ;

[0072] Calculate the health degree of each pure electric vehicle user charging behavior based on weight:

[0073] ;

[0074] wherein, is the health degree of the kth vehicle, and 10 is a proportional factor, which can also be replaced by other numerical values.

[0075] S150, according to the charging index, the remaining life of the battery and the average temperature of the area where the vehicle is located, determine the replacement confidence of each pure electric vehicle.

[0076] Firstly, the health degree , the remaining life of the battery L and the average temperature of the area where the vehicle is located constitute a feature matrix. That is, the data of each pure electric vehicle is arranged in rows or columns as a matrix format. The feature matrix is input into the pre-trained XGBoost model to obtain the replacement confidence of each pure electric vehicle.

[0077] Before using the XGBoost model for prediction, the model needs to be trained. The training process is described in detail as follows:

[0078] Step 1: Collect training samples, including: feature matrix of historical period and real replacement conclusion (label) detected by professional institutions. Among them, the historical period can be the past year. Collect user charging behavior data of multiple pure electric vehicle samples in the past year, average temperature of the area where the vehicle is located and current battery parameters, and calculate the health degree, battery remaining life and average temperature of the area where the vehicle is located of each pure electric vehicle sample according to the method described in the above steps, and constitute a feature matrix. The real replacement conclusion detected by professional institutions (for example, professional institutions with professional qualifications). Professional institutions will connect the OBD interface of the vehicle through professional equipment to read the original data in the battery management system (BMS). In the laboratory environment, the professional charging and discharging test system is used to perform standardized cycle test on the battery pack to accurately measure its current actual capacity. Compare the vehicle's advertised range, apparent range and detected actual battery capacity, and finally give the conclusion of "yes / no" to meet the replacement requirements through professional evaluation, with 0 representing "no" and 1 representing "yes".

[0079] Step 2: Define the objective function as the sum of the loss function and the model complexity; the loss function is constructed according to the absolute error loss of the replacement conclusion (i.e. the confidence of replacement or not) predicted by the XGBoost model and the real replacement conclusion.

[0080] The XGBoost model uses Q additive models to predict the output. That is, for a given sample, when training the qth tree, the first q-1 trees are known, assuming Q trees are trained, the prediction value of the cth sample is

[0081]

[0082] where, is the set of all possible functions (i.e. decision trees), is the prediction contribution of the qth tree to the cth sample, is the prediction value of the cth sample. The objective function obj is defined as the sum of the loss function and the model complexity, i.e.

[0083]

[0084]

[0085] where, is the prediction value obtained by the q-1th iteration, is the regularization term of is the regularization term of the Q-1 weak learners, is the label of the cth sample, is the prediction value obtained by the qth iteration, and C is the total number of samples. is the absolute error loss of the prediction value output by the XGBoost model based on the cth sample and the true replacement conclusion (label) .

[0086] In view of the problem that the traditional XGBoost model is very sensitive to outliers in the training data when using the least square loss function, which easily leads to large prediction bias, the loss function is improved to an absolute value loss function l to improve the robustness of the model and ensure more stable and accurate replacement prediction, i.e.

[0087]

[0088] At this time, the objective function can be written as:

[0089]

[0090] where, and are the first and second derivatives of the loss function with respect to the prediction value, is the first derivative of the loss function with respect to the current prediction value at the cth sample, ​​​​​​​is the second order derivative of the loss function with respect to the current prediction value at the cth sample.

[0091] Third step: using the training sample, the XGBoost model is trained by minimizing the objective function. XGBoost uses a greedy algorithm to build tree models step by step. Each step iteration learns a new tree model to reduce the objective function, and iterates constantly to output the final confidence by weighting.

[0092] S160, according to the replacement confidence of each pure electric vehicle, the replacement number of pure electric vehicles in the pure electric vehicle set is predicted.

[0093] According to the description of S150, the replacement confidence of the kth pure electric vehicle can be obtained If the replacement confidence of the pure electric vehicle is greater than the replacement threshold , the count is added by 1, and the replacement number NN in the pure electric vehicle set is obtained:

[0094] ;

[0095] Since the brands and battery performance of the pure electric vehicles in the current pure electric vehicle set have limitations, they cannot reflect all current vehicle / battery performance, and cannot reflect future vehicle / battery performance. Based on this, the nonlinear fitting scheme is designed in this embodiment, which can directly obtain the replacement ratio according to the user charging behavior data of a small part of vehicle samples, the average temperature of the area where the vehicle is located, and the current battery parameters, without the foregoing complex operation. Specifically, the replacement number obtained by S160 is divided by the total number of the pure electric vehicle set to obtain the replacement ratio. For example, if 30 vehicles are replaced, the total number is 50, and the replacement ratio is 60%. The user charging behavior data of each pure electric vehicle in the pure electric vehicle set, the average temperature of the area where the vehicle is located, and the current battery parameters are averaged respectively to obtain an average value set, the average value set including the average value of the user charging behavior data, the average value of the average temperature of the area where the vehicle is located, and the average value of the current battery parameters. The significance of the average value is to explore the average level of all vehicles in the set. The average value set is nonlinearly fitted with the replacement ratio to obtain a nonlinear relationship. Specifically, the average value set is taken as the independent variable, and the replacement ratio is taken as the dependent variable. If there is a to-be-predicted pure electric vehicle set at this time, only the average value set of the to-be-predicted pure electric vehicle set needs to be input into the nonlinear relationship to directly obtain the replacement ratio; and based on the total data of the to-be-predicted pure electric vehicle set and the replacement ratio (the total number multiplied by the replacement ratio), the replacement number is obtained, without calculating the charging index, the remaining battery life, and without training the model, which is conducive to speeding up the prediction efficiency.

[0096] Compared with the prior art, the beneficial effects of the present application are:

[0097] 1. The prediction model provided by the application realizes accurate and forward-looking prediction. Traditional methods rely on macro statistical data such as vehicle age and historical sales, and the prediction result is rough and lagging. The application creatively combines user charging behavior data, the average temperature of the area where the vehicle is located, and the current battery parameters to build an actual decay rate model consistent with the Arrhenius equation. This enables the prediction model to deeply reflect the scientific nature of battery performance degradation, thereby enabling earlier and more accurate prediction of vehicle replacement demand due to severe range degradation, greatly improving prediction accuracy.

[0098] 2. The application realizes multi-dimensional, dynamic and personalized replacement risk assessment. Instead of considering battery life in isolation, the application analyzes battery remaining life, charging indicators reflecting user usage habits (such as fast charging ratio, charging depth, etc.), and environmental temperature to determine the "replacement confidence" together. This solution fully considers the performance differentiation of the same vehicle model under different users and different regions, and converts static vehicle data into dynamic and personalized replacement probability. This makes the prediction result more accurately reflect the replacement willingness in the complex real market.

[0099] As shown in Figure 2 The embodiment provides an electronic device, which comprises:

[0100] at least one processor; and

[0101] a memory connected in communication with the at least one processor; wherein

[0102] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above method. The at least one processor in the electronic device can perform the above method, and thus has at least the same advantages as the above method.

[0103] Optionally, the electronic device further comprises an interface for connecting various components, including a high-speed interface and a low-speed interface. Various components are connected to each other by different buses, and can be installed on a common motherboard or in other ways as needed. The processor can process instructions executed within the electronic device, including graphical information stored in the memory or on the memory to display a GUI (Graphical User Interface) on an external input / output device such as a display device coupled to the interface. In other embodiments, multiple processors can be used with multiple memories, and / or multiple buses can be used with multiple memories, if necessary. Similarly, multiple electronic devices can be connected (e.g., as a server array, a group of blade servers, or a multi-processor system), each device providing part of the necessary operations.Figure 2 The processor 301 is taken as an example.

[0104] The memory 302 is a computer readable storage medium, which can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the method for predicting the number of purchases of the pure electric vehicle in the embodiments of the present application. The processor 301 executes various function applications and data processing of the device by running the software programs, instructions and modules stored in the memory 302, that is, implements the method for predicting the number of purchases of the pure electric vehicle.

[0105] The memory 302 can mainly include a program storage area and a data storage area. The program storage area can store an operating system and at least one application required by a function; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory 302 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state memory device. In some examples, the memory 302 can further include a memory remotely arranged with respect to the processor 301, which can be connected to the device through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0106] The electronic device can further include an input device 303 and an output device 304. The processor 301, the memory 302, the input device 303 and the output device 304 can be connected by a bus or other means, Figure 2 The connection by the bus is taken as an example.

[0107] The input device 303 can receive input digital or character information, and the output device 304 can include a display device, an auxiliary lighting device (for example, an LED), a tactile feedback device (for example, a vibration motor), etc. The display device can include but is not limited to a liquid crystal display (LCD), a light emitting diode (LED) display and a plasma display. In some embodiments, the display device can be a touch screen.

[0108] The embodiment provides a computer readable storage medium, and the medium stores computer instructions. The computer instructions are used to make a computer execute the method described above. The computer instructions on the computer readable storage medium are used to make a computer execute the method, and thus at least have the same advantages as the method.

[0109] The medium in the present application can adopt any combination of one or more computer readable mediums. The medium can be a computer readable signal medium or a computer readable storage medium. The medium may, for example, but is not limited to, an electrical, a magnetic, an optical, an electromagnetic, an infrared, or a semiconductor system, device or apparatus, or any suitable combination of the above. More specific examples (a non-exhaustive list) of the medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the medium can be any tangible medium that contains or stores a program used by or in connection with an instruction execution system, apparatus or device.

[0110] The computer readable signal medium can include a computer readable program code in a baseband or propagated as a carrier wave in a propagation medium. Such a propagated signal can take a wide variety of forms, including but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium can be any computer readable medium that can be used to carry or store the program codes for use by or in connection with an instruction execution system, apparatus or device.

[0111] Program code embodied on a computer readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the above.

[0112] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, application specific circuitry, or field programmable gate array (FPGA) circuitry can execute the program code. In some embodiments, multiple processor or multi-core processors can execute the program code.

[0113] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through a wired manner, for example, coaxial cable, optical fiber, digital subscriber line (DSL) or a wireless manner, for example, infrared, wireless, microwave, etc. The computer-readable storage medium can be any available medium accessible by a computer, or a data storage device such as a server, data center, etc. integrated with one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium or a semiconductor medium, etc. It should be noted that the computer-readable storage medium mentioned in the embodiments of the present application can be a non-volatile storage medium, in other words, it can be a non-transitory storage medium.

[0114] It should be understood that the above-mentioned various forms of processes can be reordered, added or deleted. For example, the steps described in the present application can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in the present application can be achieved, which are not limited herein.

[0115] The above detailed description does not constitute a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for predicting the replacement quantity of pure electric vehicles, characterized in that, include: For each pure electric vehicle in the pure electric vehicle set, user charging behavior data, the average temperature of the area where the vehicle is located, and the current battery parameters are obtained; wherein, the current battery parameters include at least: the current battery health, initial capacity, activation energy of battery degradation, pre-exponential factor, average battery charging temperature, and current full charge capacity. Based on the current battery parameters, determine the actual battery degradation rate; The remaining battery life is determined based on the actual degradation rate and the current battery health. Charging metrics are determined based on the user charging behavior data; wherein, the user charging behavior data includes: the number of times the battery is fast charged, the total number of times it is charged, the battery charging temperature, the optimal charging temperature range, the number of times the battery is overcharged, and the battery's charge level at the beginning and end of charging; the charging metrics include: the percentage of times the battery is fast charged, the charging temperature compatibility, the frequency of overcharging, and the depth of charging. The confidence level for replacing each pure electric vehicle is determined based on the charging indicators, remaining battery life, and average temperature of the area where the vehicle is located. Based on the replacement confidence level of each pure electric vehicle, predict the number of pure electric vehicles to be replaced in the pure electric vehicle set; where the replacement confidence level is the confidence level of whether or not to replace, i.e. the replacement probability; if the replacement confidence level of a pure electric vehicle is greater than the replacement threshold, increment the count by 1 to obtain the number of pure electric vehicles to be replaced in the pure electric vehicle set.

2. The method for predicting the replacement quantity of pure electric vehicles according to claim 1, characterized in that, Based on the current battery parameters, the actual battery degradation rate is determined, including: The basic degradation rate of the battery is obtained based on its initial capacity, current full charge capacity, and the vehicle's service life. The temperature acceleration coefficient is determined based on the activation energy, pre-exponential factor, and average charging temperature of the battery. The actual degradation rate of the battery is obtained based on the battery's base degradation rate and temperature acceleration coefficient.

3. The method for predicting the replacement quantity of pure electric vehicles according to claim 1, characterized in that, The charging metrics are determined based on the user charging behavior data, including: The percentage of fast charging cycles is obtained based on the number of fast charging cycles and the total number of charging cycles. The charging temperature fit is obtained based on the battery charging temperature and the optimal charging temperature range. The overcharge frequency is obtained based on the number of overcharges and the total number of charges of the battery. The depth of charge is determined based on the battery's charge level at the beginning and end of the charging process.

4. The method for predicting the replacement quantity of pure electric vehicles according to claim 3, characterized in that, After determining the depth of charge based on the battery's initial and final charge levels, the following steps are also included: Calculate the entropy value of each charging indicator based on its standardized value. The health score is obtained by weighting and summing each charging indicator using the entropy weighting method.

5. The method for predicting the replacement quantity of pure electric vehicles according to claim 4, characterized in that, Based on the aforementioned charging indicators, remaining battery life, and the average temperature of the area where the vehicle is located, the confidence level for trade-in of each pure electric vehicle is determined, including: The health status, remaining battery life, and average temperature of the area where the vehicle is located are used to construct a feature matrix; The feature matrix is ​​input into a pre-trained XGBoost model to obtain the replacement confidence score for each pure electric vehicle.

6. The method for predicting the replacement quantity of pure electric vehicles according to claim 5, characterized in that, Before inputting the feature matrix into the pre-trained XGBoost model to obtain the replacement confidence score for each pure electric vehicle, the process also includes: Collect training samples, which include feature matrices for historical periods and real trade-in conclusions detected by professional institutions; The objective function is defined as the sum of the loss function and the model complexity; the loss function is constructed based on the absolute error loss between the trade-in conclusion predicted by the XGBoost model and the actual trade-in conclusion. The XGBoost model is trained using the training samples by minimizing the objective function.

7. The method for predicting the replacement quantity of pure electric vehicles according to claim 6, characterized in that, If the confidence level for replacing a pure electric vehicle is greater than the replacement threshold, the count is incremented by 1. After obtaining the number of replacements in the pure electric vehicle set, the following is also included: Divide the number of vehicles to be traded in by the total number of vehicles in the pure electric vehicle collection to obtain the trade-in ratio. The user charging behavior data, the average temperature of the area where the vehicle is located, and the current battery parameters of each pure electric vehicle in the pure electric vehicle set are averaged to obtain an average value set, which includes the average value of the user charging behavior data, the average value of the average temperature of the area where the vehicle is located, and the average value of the current battery parameters. The average value set is nonlinearly fitted to the trade-in ratio to obtain a nonlinear relationship. The average value set of the pure electric vehicle set to be predicted is input into a nonlinear relation to obtain the replacement ratio; and the replacement quantity is obtained based on the total data of the pure electric vehicle set to be predicted and the replacement ratio.

8. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to at least one of the processors; The memory stores instructions executable by at least one of the processors, which are executed to enable the at least one processor to perform the method for predicting the replacement quantity of pure electric vehicles according to any one of claims 1-7.

Images