Method and device for predicting service life of battery of rented vehicle, medium, equipment and product

Abstract

The invention is applied to the technical field of data processing, and provides a rental vehicle battery life prediction method and device, a medium, equipment and a product, and the method comprises the steps: constructing a prediction model for predicting battery parameters in advance; performing multiple rounds of prediction processing on battery parameters of a target battery according to the prediction model until the predicted battery parameters of the target battery meet prediction stopping conditions; and determining the predicted life of the target battery according to the total number of charge and discharge turns determined by multiple turns of battery parameter prediction processing. According to the technical scheme provided by the embodiment of the invention, simulation of the variable charging and discharging mode of the target battery is realized through multiple rounds of prediction processing and setting of the corresponding charging and discharging characteristics for each round, the service life of the battery can be predicted based on multiple charging and discharging characteristics, and the method is suitable for the condition that multiple charging and discharging characteristics exist in a car rental scene.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a method, device, medium, equipment and product for predicting the battery life of rental vehicles. Background Art

[0002] With the development of new energy vehicles, the rental demand for new energy vehicles has been increasing year by year. Due to the high usage frequency of rental vehicles, it is particularly important to monitor the vehicle status, especially to monitor the health status of the power batteries of new energy vehicles, which not only facilitates tenants to select suitable vehicles, but also facilitates merchants to price based on the battery status.

[0003] Different from the usage conditions of traditional household new energy vehicles, the power batteries of rental new energy vehicles are easily affected by the driving habits of drivers and the charging methods, resulting in changes in battery capacity attenuation, making it difficult to predict the health status of power batteries. Summary of the Invention

[0004] To solve the above problems, an object of the embodiments of the present invention is to provide a method, device, medium, equipment and product for predicting the battery life of rental vehicles.

[0005] In a first aspect, an embodiment of the present invention provides a method for predicting the battery life of a rental vehicle, including: Pre-construct a prediction model for predicting battery parameters; Perform multiple rounds of prediction processing on the battery parameters of the target battery according to the prediction model until the battery parameters after prediction of the target battery meet the stop prediction condition; Determine the predicted life of the target battery according to the total number of charge and discharge cycles determined by multiple rounds of the battery parameter prediction processing; Wherein, the prediction processing of the current round includes: Determine the current charge and discharge characteristics of the target battery in the current round; Determine the current number of charge and discharge cycles corresponding to the current round, and add the current number of charge and discharge cycles to the total number of charge and discharge cycles; Take the battery parameters predicted by the previous round of prediction processing as the initial battery parameters of the current round of prediction processing, and use the prediction model to predict the battery parameters of the target battery according to the current charge and discharge characteristics and the current number of charge and discharge cycles, and determine the battery parameters of the target battery after charging and discharging the current number of charge and discharge cycles.

[0006] In some alternative embodiments, the current charge and discharge characteristics are determined based on the following method: Determine the charge-discharge characteristics transferred from the previous charge-discharge characteristics in the previous round according to the target characteristic transfer matrix of the target battery, and use the transferred charge-discharge characteristics as the current charge-discharge characteristics in the current round; Wherein, the target characteristic transfer matrix is used to represent the probability that the target battery transfers from one charge-discharge characteristic to another charge-discharge characteristic.

[0007] In some alternative embodiments, the determining the charge-discharge characteristics transferred from the previous charge-discharge characteristics in the previous round according to the target characteristic transfer matrix of the target battery includes: Statistically analyze the transfer of charge-discharge characteristics of the reference battery during the rental process to generate an initial characteristic transfer matrix Q; Determine the target characteristic transfer matrix P of the target battery according to the initial characteristic transfer matrix Q, and: ; Wherein, represents the probability that the target battery is in the i-th charge-discharge characteristic, represents the probability that the target battery is in the j-th charge-discharge characteristic, i≠j; represents the probability of transferring from the i-th charge-discharge characteristic to the j-th charge-discharge characteristic in the initial characteristic transfer matrix Q, represents the probability of transferring from the j-th charge-discharge characteristic to the i-th charge-discharge characteristic in the initial characteristic transfer matrix Q, represents the probability of transferring from the i-th charge-discharge characteristic to the j-th charge-discharge characteristic in the target characteristic transfer matrix P; According to the target characteristic transfer matrix P, determine the previous charge-discharge characteristic in the previous round The probability distribution of various charge-discharge characteristics transferred to, and select the current charge-discharge characteristic in the current round according to the probability distribution ; T represents the number of rounds of the prediction process.

[0008] In some alternative embodiments, the selecting the current charge-discharge characteristic in the current round according to the probability distribution , includes: According to the probability values corresponding to various charge-discharge characteristics in the probability distribution, divide the value range from 0 to 1 into multiple value ranges, and the size of each value range matches the probability value corresponding to the corresponding charge-discharge characteristic; Randomly generate a random number between 0 and 1, and use the charge-discharge characteristic corresponding to the value range into which the random number falls as the current charge-discharge characteristic in the current round .

[0009] In some alternative embodiments, the determining the current number of charge-discharge cycles corresponding to the current round includes: After determining the current charge-discharge characteristic, repeat the feature transfer operation until the next charge-discharge characteristic for the next round of prediction processing is determined. The feature transfer operation includes: Determine the pending charge-discharge characteristic transferred from the current charge-discharge characteristic according to the target feature transfer matrix. If the pending charge-discharge characteristic is the same as the current charge-discharge characteristic, increment the current charge-discharge cycle count by one, and then continue to execute the feature transfer operation; the initial value of the current charge-discharge cycle count is 1. If the pending charge-discharge characteristic is different from the current charge-discharge characteristic, use the pending charge-discharge characteristic as the next charge-discharge characteristic for the next round of prediction processing.

[0010] In some alternative embodiments, the prediction model includes prediction sub-models corresponding to various charge-discharge characteristics; the prediction sub-model is used to predict the change of the battery parameters of the target battery with the charge-discharge cycle count when the target battery is charged and discharged according to the corresponding charge-discharge characteristic. The step of using the prediction model to predict the battery parameters of the target battery according to the current charge-discharge characteristic and the current charge-discharge cycle count, and determining the battery parameters of the target battery after charging and discharging the current charge-discharge cycle count includes: Determine the target prediction sub-model corresponding to the current charge-discharge characteristic. Input the initial battery parameters of the current round of prediction processing and the current charge-discharge cycle count into the target prediction sub-model, and determine the battery parameters of the target battery after charging and discharging the current charge-discharge cycle count according to the prediction result of the target prediction sub-model.

[0011] In a second aspect, an embodiment of the present invention further provides a prediction device for the battery life of a rental vehicle, including: A model construction module, configured to pre-construct a prediction model for predicting battery parameters. A prediction module, configured to perform multiple rounds of prediction processing on the battery parameters of the target battery according to the prediction model until the battery parameters of the target battery after prediction meet the stop prediction condition. A life determination module, configured to determine the predicted life of the target battery according to the total charge-discharge cycle count determined by multiple rounds of the battery parameter prediction processing. Wherein, the current round of prediction processing executed by the prediction module includes: Determine the current charge-discharge characteristic of the target battery in the current round. Determine the current charge-discharge cycle count corresponding to the current round, and increment the total charge-discharge cycle count by the current charge-discharge cycle count. Take the battery parameters predicted in the previous round of prediction processing as the initial battery parameters for the current round of prediction processing. Using the prediction model, predict the battery parameters of the target battery based on the current charge-discharge characteristics and the current number of charge-discharge cycles, and determine the battery parameters of the target battery after charging and discharging the current number of charge-discharge cycles.

[0012] In a third aspect, an embodiment of the present invention further provides a computer storage medium storing computer-executable instructions for the prediction method of the battery life of a rental vehicle described in any one of the above.

[0013] In a fourth aspect, an embodiment of the present invention further provides an electronic device, including: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, 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 execute the prediction method of the battery life of a rental vehicle described in any one of the above.

[0014] In a fifth aspect, the present invention provides a computer program product including computer instructions for causing a computer to execute the prediction method of the battery life of a rental vehicle described in any one of the above.

[0015] In the solution provided in the first aspect of the above embodiments of the present invention, through multiple rounds of prediction processing, and setting the charge-discharge characteristics of the target battery in each round of prediction processing. By setting corresponding charge-discharge characteristics for each round, the simulation of the variable charge-discharge mode of the target battery is realized, and the prediction of the battery life can be based on multiple charge-discharge characteristics, which is applicable to the situation where there are multiple charge-discharge characteristics in the car rental scenario. When using the rental vehicle battery based on multiple driving habits and charging methods, the service life of the rental vehicle battery can also be predicted more accurately, which can provide a reference basis for the pricing of rental vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 Shows a flowchart of a prediction method for the battery life of a rental vehicle provided by an embodiment of the present invention; Figure 2The flowchart of another method for predicting the battery life of rental vehicles provided by the embodiments of the present invention is shown; Figure 3 A schematic diagram of feature transfer provided by the embodiments of the present invention is shown; Figure 4 The structural schematic diagram of a device for predicting the battery life of rental vehicles provided by the embodiments of the present invention is shown; Figure 5 The structural schematic diagram of an electronic device for executing the method for predicting the battery life of rental vehicles provided by the embodiments of the present invention is shown. Detailed implementation manners

[0018] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0019] In the car rental scenario, the power battery of new energy vehicles is easily affected by the driving habits of drivers and the charging methods. Specifically, the driving habits of each tenant are uncontrollable; for example, since the vehicles rented by tenants are not their own vehicles, some tenants may adopt driving methods of rapid acceleration and rapid braking, which easily cause the battery to discharge / charge with large current frequently, increasing the internal resistance of the battery and causing battery aging.

[0020] Moreover, different from the home scenario where slow charging is mainly used, in the car rental scenario, most tenants often use fast charging to charge the vehicle in order to quickly replenish energy, and fast charging will accelerate the battery polarization reaction, resulting in faster capacity attenuation after long-term use; in addition, some tenants may also frequently discharge deeply, that is, deplete the battery power to an extremely low level (such as below 10%) and then charge, or frequently charge to full and discharge to near zero (such as charge to 100% and then discharge to near 0), which will also exacerbate the battery chemical side reactions, accelerate capacity attenuation, and reduce the battery cycle life.

[0021] Generally, the cycle life of a power battery can reach 1500 times, that is, it can be charged and discharged 1500 cycles. However, poor driving habits and charging methods are extremely likely to affect the battery life, resulting in the actual cycle life of the rental vehicle battery often only reaching about 1200 times, or even lower.

[0022] Currently, there are solutions for predicting the battery cycle life. For example, predicting battery degradation based on the electrochemical model of the battery (e.g., the pseudo-two-dimensional model), or predicting based on deep neural network models such as LSTM (Long Short-Term Memory Network). However, these prediction solutions are mainly applicable to scenarios where the charging changes little, such as household vehicles.

[0023] A method for predicting the battery life of rental vehicles provided by an embodiment of the present invention realizes the prediction of battery life based on multiple charge-discharge characteristics through multiple rounds of prediction processing, and setting the charge-discharge characteristics of the battery in each round of prediction processing, which is applicable to the situation where there are multiple charge-discharge characteristics in the car rental scenario.

[0024] A method for predicting the battery life of rental vehicles provided by an embodiment of the present invention, see Figure 1 as shown, includes: Step 101: Pre-build a prediction model for predicting battery parameters.

[0025] In this embodiment, in order to be able to predict the battery life, predict the battery parameters related to the battery life. For example, the battery parameters may include battery capacity, battery internal resistance, etc. Among them, the prediction technology for battery capacity is relatively mature, and battery capacity can be preferentially selected as the battery parameter to be predicted.

[0026] Pre-build a corresponding prediction model for predicting the battery parameters. For example, the prediction model can be built based on LSTM to be able to predict the change of battery parameters. Among them, the battery parameters are related to the charge-discharge characteristics of the battery, so the prediction model needs to be able to predict the battery parameters under different charge-discharge characteristics, which will be described later.

[0027] Step 102: Perform multiple rounds of prediction processing on the battery parameters of the target battery according to the prediction model until the battery parameters of the target battery after prediction meet the stop prediction condition.

[0028] In this embodiment, for the battery of the rental vehicle whose life needs to be predicted, it is called the target battery. Perform multiple rounds of prediction processing on the target battery based on the prediction model. Each round of prediction processing is used to update the battery parameters of the target battery; after multiple rounds of prediction processing, the predicted battery parameters gradually approach the parameters corresponding to the final life of the target battery. At this time, it can be determined that the battery parameters after prediction meet the stop prediction condition; in other words, the stop prediction condition is that the battery parameters of the target battery after prediction indicate the end of life (EOL) of the target battery.

[0029] For example, if the battery parameter is battery capacity, a capacity decay threshold, such as 80%, can be set. If the battery capacity of the target battery is lower than 80% after multiple rounds of prediction processing, it can be determined that it meets the stop prediction condition.

[0030] Among them, taking the prediction process of the current round as an example, the prediction process of the current round includes steps 1021 to 1023.

[0031] Step 1021: Determine the current charge-discharge characteristics of the target battery in the current round.

[0032] In this embodiment, the target battery of the rental vehicle may have various charge-discharge characteristics, which can be specifically determined based on the driving habits (discharge methods), charging methods, etc. corresponding to the target battery. Different driving habits and charging methods correspond to different charge-discharge characteristics.

[0033] For example, based on whether the driver accelerates and brakes suddenly, two driving habits can be determined; based on whether fast charging is used, two charging methods can also be determined. Based on two driving habits and two charging methods, four charge-discharge characteristics can be determined. It can be understood that the driving habits and charging methods can be further divided to define more types of charge-discharge characteristics. This embodiment does not limit the specific number of types of charge-discharge characteristics.

[0034] For each round of prediction process, a unique charge-discharge characteristic needs to be determined; for the convenience of description, the charge-discharge characteristic corresponding to the prediction process of the current round is called the current charge-discharge characteristic.

[0035] Step 1022: Determine the current charge-discharge cycle number corresponding to the current round, and add the current charge-discharge cycle number to the total charge-discharge cycle number.

[0036] In this embodiment, the charge-discharge condition of the target battery is represented by the total charge-discharge cycle number. In each round of prediction process, the target battery is charged and discharged for a certain number of cycles; for the prediction process of the current round, the number of cycles of charging and discharging the target battery is called the current charge-discharge cycle number. By adding the current charge-discharge cycle number to the total charge-discharge cycle number, the update of the total charge-discharge cycle number is realized.

[0037] It can be understood that when predicting the target battery, the target battery has an initial total charge-discharge cycle number. On this basis, the charge-discharge cycle numbers corresponding to each round of prediction process are successively added to realize the update of the total charge-discharge cycle number.

[0038] Among them, the charge-discharge cycle number of each round (such as the current charge-discharge cycle number) can be a fixed value not less than 1, for example, 1, that is, each round of prediction process only performs one round of charge-discharge process on the target battery. Or, the charge-discharge cycle number of each round can also be determined based on other methods.

[0039] Step 1023: Use the battery parameters predicted in the previous round of prediction processing as the initial battery parameters for the current round of prediction processing. Utilize the prediction model to predict the battery parameters of the target battery based on the current charge-discharge characteristics and the current number of charge-discharge cycles, and determine the battery parameters of the target battery after the current number of charge-discharge cycles.

[0040] In this embodiment, in each round of prediction processing, the battery parameters of the target battery (such as battery capacity, internal resistance, etc.) will be updated. When performing prediction processing in the current round, use the battery parameters predicted in the previous round of prediction processing as the initial battery parameters at the beginning of the current round of prediction processing, that is, the initial battery parameters. It can be understood that the initial battery parameters corresponding to the first round of prediction processing are the actual battery parameters of the target battery.

[0041] After determining the current charge-discharge characteristics and the current number of charge-discharge cycles, the battery parameters of the target battery can be predicted based on the prediction model. The change in the battery parameters of the target battery after charging and discharging the target battery for the current number of charge-discharge cycles can be predicted, that is, how the target battery will change from the initial battery parameters, so as to determine the battery parameters of the target battery after the current number of charge-discharge cycles. And if the stop prediction condition is not met at this time, these battery parameters will also be used as the initial battery parameters for the next round of prediction processing and continue the prediction processing.

[0042] For example, the battery parameter is the battery capacity, and the initial battery capacity (i.e., the initial battery parameter) for the current round of prediction processing is 90%; if the battery capacity of the target battery decreases by 0.5% after charging and discharging for the current number of charge-discharge cycles, it can be determined that the battery capacity of the target battery after the current number of charge-discharge cycles is 89.5%. In the next round of prediction processing, continue to perform prediction on the basis of the initial condition where the battery capacity is 89.5%.

[0043] Step 103: Determine the predicted life of the target battery according to the total number of charge-discharge cycles determined by multiple rounds of battery parameter prediction processing.

[0044] In this embodiment, when the battery parameters of the target battery meet the stop prediction condition, stop the prediction processing, that is, no longer increase the total number of charge-discharge cycles of the target battery. At this time, the total number of charge-discharge cycles is the maximum number of charge-discharge cycles that the target battery can achieve when it reaches the end of its life. Therefore, based on the total number of charge-discharge cycles at this time, the predicted life of the target battery can be determined. The predicted life is specifically the number of cycles of the target battery, that is, the total number of charge-discharge cycles of the target battery.

[0045] For example, the current actual number of charge and discharge cycles of the target battery is 300 times, that is, the initial value of the total number of charge and discharge cycles is 300; if the number of charge and discharge cycles is set to 1 in each round of prediction processing, and the prediction stop condition is reached after 1000 rounds of prediction processing, the total number of charge and discharge times of the target battery at this time is 1300 times, that is, the predicted cycle life (i.e., the predicted life) of the target battery is 1300 times. In other words, the target battery can approximately continue to be charged and discharged 1000 times.

[0046] The prediction method for the battery life of a rental vehicle provided by the embodiment of the present invention, through multiple rounds of prediction processing, and setting the charge and discharge characteristics of the target battery in each round of prediction processing. By setting corresponding charge and discharge characteristics for each round, the simulation of the variable charge and discharge mode of the target battery is realized, and the prediction of the battery life can be based on various charge and discharge characteristics, which is applicable to the situation where there are various charge and discharge characteristics in the car rental scenario. When using the rental vehicle battery based on various driving habits and charging methods, the service life of the rental vehicle battery can also be predicted more accurately, which can provide a reference basis for the pricing of rental vehicles.

[0047] The embodiment of the present invention also provides another prediction method for the battery life of a rental vehicle. See Figure 2 as shown, which includes: Step 201: Pre-construct a prediction model for predicting battery parameters.

[0048] Specifically, refer to the relevant description in step 101 above, which will not be elaborated here.

[0049] Step 202: According to the target feature transition matrix of the target battery, determine the charge and discharge feature transferred from the previous charge and discharge feature of the previous round, and use the transferred charge and discharge feature as the current charge and discharge feature in the current round. The target feature transition matrix is used to represent the probability of the target battery transferring from one charge and discharge feature to another.

[0050] In this embodiment, in the process of "determining the current charge and discharge feature of the target battery in the current round" in the above step 1021, the corresponding charge and discharge feature can be randomly selected as the current charge and discharge feature based on the probability distribution of various charge and discharge features. However, the current charge and discharge feature may have continuity; for example, when a driver rents a vehicle for a long time, the charge and discharge features (driving habits, charging methods) will not change within a certain period of time, and the traditional random selection method of charge and discharge features cannot reflect this continuity.

[0051] Based on this, a matrix for representing the transition probabilities between various charge-discharge characteristics of the target battery is constructed, that is, the target characteristic transition matrix, and hereinafter, the target characteristic transition matrix is denoted by P. In this embodiment, the target characteristic transition matrix P is used to represent the transition probabilities between various charge-discharge characteristics; if there are n types of charge-discharge characteristics in total, then the target characteristic transition matrix P is an n×n matrix, where the element p at the i-th row and j-th column ij represents the probability of transitioning from the i-th charge-discharge characteristic to the j-th charge-discharge characteristic. When i = j, it means that the charge-discharge characteristic does not change.

[0052] If the charge-discharge characteristic in the previous round is charge-discharge characteristic A, then based on the target characteristic transition matrix P, the probabilities of transitioning from this charge-discharge characteristic A to each charge-discharge characteristic can be determined, and then based on each probability, the charge-discharge characteristic to be used next, that is, the current charge-discharge characteristic, can be determined.

[0053] Taking three charge-discharge characteristics as an example, the transition situations of the three charge-discharge characteristics can be seen in Figure 3 as shown. In this embodiment, taking to represent the element p ij , Figure 3 the corresponding target characteristic transition matrix P can be expressed as: .

[0054] Taking the first row of the target characteristic transition matrix P as an example, the three elements respectively represent the probabilities of transitioning from the 1st charge-discharge characteristic to various charge-discharge characteristics.

[0055] If the previous charge-discharge characteristic is the 1st charge-discharge characteristic, then based on the three elements in the first row of the target characteristic transition matrix P, the probabilities of transitioning to various charge-discharge characteristics can be determined, that is , , , and based on these three probabilities, it is determined which specific charge-discharge characteristic the current charge-discharge characteristic is.

[0056] In some alternative embodiments, the above step 202 "determine the charge-discharge characteristic transitioned from the previous charge-discharge characteristic in the previous round according to the target characteristic transition matrix of the target battery" may include steps A1 to A3.

[0057] Step A1: Statistically analyze the transition situations of the charge-discharge characteristics of the reference battery during the leasing process to generate an initial characteristic transition matrix Q.

[0058] Step A2: Determine the target characteristic transition matrix P of the target battery according to the initial characteristic transition matrix Q, and: ; wherein, The probability that the target battery is in the \(i\)-th charge-discharge characteristic The probability that the target battery is in the \(j\)-th charge-discharge characteristic, where \(i\neq j\); The probability of transitioning from the \(i\)-th charge-discharge characteristic to the \(j\)-th charge-discharge characteristic in the initial characteristic transition matrix \(Q\), The probability of transitioning from the \(j\)-th charge-discharge characteristic to the \(i\)-th charge-discharge characteristic in the initial characteristic transition matrix \(Q\), The probability of transitioning from the \(i\)-th charge-discharge characteristic to the \(j\)-th charge-discharge characteristic in the target characteristic transition matrix \(P\).

[0059] Step A3: According to the target characteristic transition matrix \(P\), determine the probability distribution of the various charge-discharge characteristics transferred from the previous charge-discharge characteristic in the previous round and select the current charge-discharge characteristic in the current round according to the probability distribution. ; \(T\) represents the number of rounds of prediction processing.

[0060] In this embodiment, based on the target characteristic transition matrix \(P\), the charge-discharge characteristics in each round of prediction processing can be determined in sequence to form a Markov chain corresponding to the charge-discharge characteristics.

[0061] During the prediction processing of each round, the corresponding charge-discharge characteristics need to conform to the required probability distribution \(S\). This probability distribution \(S\) can be specifically determined by statistically analyzing the various charge-discharge characteristics of the reference battery throughout its life cycle to determine the proportion of each charge-discharge characteristic, thereby determining the corresponding probability distribution \(S\).

[0062] Moreover, the charge-discharge characteristics in the next round of prediction processing determined based on the target characteristic transition matrix \(P\) also need to conform to this probability distribution \(S\) to ensure the stability requirements of the Markov chain, that is, it is necessary to satisfy: ; The probability distribution \(S\) is a \(1\times n\) matrix (i.e., a row vector).

[0063] .

[0064] In the above formula, represents the probability that the target battery is in the \(i\)-th charge-discharge characteristic, represents the probability of transitioning from the \(i\)-th charge-discharge characteristic to the \(j\)-th charge-discharge characteristic in the target characteristic transition matrix \(P\).

[0065] Moreover, from the above formula, it can be obtained that . For example, when \(j = 1\), the first element of the probability distribution \(S\) .

[0066] Since each element in the \(j\)-th row of the target characteristic transition matrix \(P\) represents the probability of transitioning from the \(j\)-th charge-discharge characteristic to another charge-discharge characteristic, the sum of these probabilities should be 1, that is , add it to , we can get: .

[0067] Suppose that for any i, , then the above equation can be guaranteed to hold, that is . In other words, as long as is satisfied, then can be guaranteed. The former is a sufficient condition for the latter.

[0068] In order to accurately determine the target feature transfer matrix P, in this embodiment, the transfer situation of the charge and discharge features of the reference battery during the rental process is obtained, and these transfer situations are statistically analyzed, and the corresponding feature transfer matrix can be initially determined. If the feature transfer matrix is directly used as the target feature transfer matrix P at this time, it does not necessarily meet the requirements, so this feature matrix is called the initial feature transfer matrix Q. It can be understood that the initial feature transfer matrix Q is also an n×n matrix.

[0069] Among them, the reference battery can be a battery of the same model as the target battery; or, the reference battery can also be the target battery, that is, the transfer situation of the charge and discharge features of the target battery during the rental process is statistically analyzed.

[0070] Moreover, the transfer situation specifically includes the proportion of the transfer quantity of the reference battery from one charge and discharge feature to another charge and discharge feature in the total transfer quantity of this charge and discharge feature. For example, if n = 3 (that is, there are three charge and discharge features in total), by statistically analyzing the transfer situation of the charge and discharge features, it can be determined that the transfer quantities from the first charge and discharge feature to the second and third charge and discharge features are 400 times and 500 times respectively, and the number of times the first charge and discharge feature remains unchanged is 100 times. Then the total transfer quantity of the first charge and discharge feature is 1000 times, and its transfer probabilities to the first, second, and third charge and discharge features are 0.1, 0.4, and 0.5 respectively, that is, the three elements in the first row of the initial feature transfer matrix Q are 0.1, 0.4, and 0.5 respectively.

[0071] Although it is difficult to directly determine the accurate target feature transfer matrix P based on the probability distribution S, as shown above, if , then . Using each element in the initial feature transfer matrix Q to represent the corresponding element in the target feature transfer matrix P, it can be expressed as: , is a coefficient to be determined.

[0072] At this time, .

[0073] If , then , that is, the former is also a sufficient condition for the latter.

[0074] Among them, represents the probability that the target battery currently has the j-th charge-discharge characteristic, while represents the probability that the target battery transfers from the j-th charge-discharge characteristic to the i-th charge-discharge characteristic, that is is a probability between 0 and 1; similarly, is also a probability between 0 and 1.

[0075] Therefore, , whose physical meaning is: Although is not necessarily , it has a certain probability of being . In other words, when sampling based on the initial characteristic transition matrix Q, if each probability is used to determine whether to retain the sampled sample, the transfer situation of the retained samples can ultimately conform to the target characteristic transition matrix P.

[0076] However, due to the relatively small probability , most samples will be discarded, that is, only a small number of samples will be retained (for example, , then 90% of the samples will be discarded, and only 10% of the samples can be retained), resulting in low efficiency.

[0077] For and the smaller value of, it satisfies: .

[0078] If , correspondingly, , at this time .

[0079] Therefore, in this embodiment, let the probability , then it can satisfy , that is: .

[0080] And, the probability at this time is greater than , so that sampling can be achieved with a higher probability.

[0081] It should be noted that when i≠j, ; if i = j, then based on the characteristic that the sum of probabilities corresponding to the same charge-discharge characteristic is 1, it can be calculated that: .

[0082] After determining each element in the target feature transition matrix P the probability distribution of various charge and discharge characteristics transferred from the previous charge and discharge characteristic in the previous round can be determined which corresponds to the elements in the corresponding row of the target feature transition matrix P. Furthermore, the current charge and discharge characteristic of the current round is selected according to this probability distribution .

[0083] Optionally, the above step A3 "select the current charge and discharge characteristic of the current round according to the probability distribution " includes steps A31 to A32

[0084] Step A31: According to the probability values corresponding to various charge and discharge characteristics in the probability distribution, divide the value range from 0 to 1 into multiple value ranges, and the size of each value range matches the probability value corresponding to the corresponding charge and discharge characteristic

[0085] Step A32: Randomly generate a random number between 0 and 1, and use the charge and discharge characteristic corresponding to the value range into which the random number falls as the current charge and discharge characteristic of the current round .

[0086] For example, taking three charge and discharge characteristics as an example, if the previous charge and discharge characteristic is the first type of charge and discharge characteristic, then according to the elements in the first row of the target feature transition matrix P, the corresponding probability distribution can be determined; assuming that the probability distribution is (0.5, 0.3, 0.2), then corresponding value ranges can be set for the three probability values: (0, 0.5], (0.5, 0.8], (0.8, 1)

[0087] Randomly generate a random number r between 0 and 1. Based on the size of the random number r, it can be determined which value range it falls into, and then the corresponding charge and discharge characteristic can be determined. For example, if the random number r = 0.68 and it falls into the value range (0.5, 0.8] of the probability 0.3 (the probability corresponding to the second type of charge and discharge characteristic), the second type of charge and discharge characteristic can be used as the current charge and discharge characteristic .

[0088] In this embodiment, as shown in the above step 1022, it is necessary to determine the current charge and discharge cycle number corresponding to the current round. Among them, determining the current charge and discharge cycle number corresponding to the current round includes: after determining the current charge and discharge characteristic, repeatedly performing the feature transfer operation until the next charge and discharge characteristic for the next round of prediction processing is determined

[0089] The above-mentioned feature transfer operation includes: determining the to-be-determined charge-discharge feature transferred from the current charge-discharge feature according to the target feature transfer matrix; if the to-be-determined charge-discharge feature is the same as the current charge-discharge feature, increment the current charge-discharge cycle count by one, and then continue to execute the feature transfer operation; the initial value of the current charge-discharge cycle count is 1; if the to-be-determined charge-discharge feature is different from the current charge-discharge feature, use the to-be-determined charge-discharge feature as the next charge-discharge feature for the next round of prediction processing.

[0090] Specifically, as Figure 2 shown, the process of determining the current charge-discharge cycle count specifically includes steps 203 to 207.

[0091] Step 203: Set the current charge-discharge cycle count of the target battery to 1. That is, the initial value of the current charge-discharge cycle count is 1.

[0092] Step 204: Determine the to-be-determined charge-discharge feature transferred from the current charge-discharge feature according to the target feature transfer matrix. This process can also determine the to-be-determined charge-discharge feature based on the method of steps A31 to A32, which will not be elaborated here.

[0093] Step 205: Whether the to-be-determined charge-discharge feature is the same as the current charge-discharge feature. If the two are the same, continue to step 206; otherwise, continue to step 207.

[0094] Step 206: Increment the current charge-discharge cycle count of the target battery by one. Then execute step 204 again, that is, repeat the above-mentioned feature transfer operation.

[0095] Step 207: Use the to-be-determined charge-discharge feature as the next charge-discharge feature for the next round of prediction processing.

[0096] In this embodiment, when determining the current charge-discharge cycle count, continue to make a transfer judgment based on the target feature transfer matrix P to determine whether there is a situation where the charge-discharge features are the same for multiple consecutive times. If so, regard these situations of the same charge-discharge features as one prediction processing, thereby reducing the number of prediction processings and improving the processing efficiency.

[0097] For example, after determining the current charge-discharge feature, if the determined to-be-determined charge-discharge feature is the same as the current charge-discharge feature, then add 1 to the current charge-discharge cycle count, that is, the current charge-discharge cycle count becomes 2; if the to-be-determined charge-discharge feature determined again by executing step 204 is the same as the current charge-discharge feature later, then the current charge-discharge cycle count becomes 3, and so on.

[0098] If the determined to-be-determined charge-discharge feature is different from the current charge-discharge feature, do not increment the current charge-discharge cycle count. At this time, the determined to-be-determined charge-discharge feature can be used as the charge-discharge feature corresponding to the next round of prediction processing, that is, the next charge-discharge feature.

[0099] It should be noted that during the first-round prediction process, when determining the current charge-discharge characteristics based on step 202, calculation processing is required. For example, the current charge-discharge characteristics are determined based on the random number r. In subsequent rounds of prediction processing, since the next charge-discharge characteristics have been determined in the previous round of prediction processing, no additional calculation is required when executing step 202, and the charge-discharge characteristics determined in the previous round can be directly used.

[0100] Step 208: Increase the total number of charge-discharge cycles by the current number of charge-discharge cycles.

[0101] Step 209: Use the battery parameters predicted in the previous round of prediction processing as the initial battery parameters for the current round of prediction processing. Utilize the prediction model to predict the battery parameters of the target battery based on the current charge-discharge characteristics and the current number of charge-discharge cycles, and determine the battery parameters of the target battery after the current number of charge-discharge cycles.

[0102] In some alternative embodiments, the prediction model includes prediction sub-models corresponding to various charge-discharge characteristics; the prediction sub-models are used to predict the variation of the battery parameters of the target battery with the number of charge-discharge cycles when the target battery is charged and discharged according to the corresponding charge-discharge characteristics.

[0103] The above step 209 "Utilize the prediction model to predict the battery parameters of the target battery based on the current charge-discharge characteristics and the current number of charge-discharge cycles, and determine the battery parameters of the target battery after the current number of charge-discharge cycles" may include steps B1 to B2.

[0104] Step B1: Determine the target prediction sub-model corresponding to the current charge-discharge characteristics.

[0105] Step B2: Input the initial battery parameters of the current round of prediction processing and the current number of charge-discharge cycles into the target prediction sub-model, and determine the battery parameters of the target battery after the current number of charge-discharge cycles according to the prediction result of the target prediction sub-model.

[0106] In this embodiment, a model for predicting the variation of battery parameters, that is, the prediction sub-model, is set for each charge-discharge characteristic. For example, if four charge-discharge characteristics are involved, four prediction sub-models need to be set. The prediction sub-model can predict the variation of the battery parameters of the target battery with the number of charge-discharge cycles under a certain charge-discharge characteristic.

[0107] Specifically, for the current charge-discharge characteristics of the current round, the corresponding prediction sub-model, that is, the target prediction sub-model, can be determined. The target prediction sub-model can predict the variation of the battery parameters of the target battery when the target battery is charged and discharged with the current charge-discharge characteristics. For example, the attenuation rate of the battery capacity of the target battery, etc.

[0108] Input the initial battery parameters and the current number of charge-discharge cycles processed in the current round of prediction into the target prediction sub-model. Use the initial battery parameters as the initial values for charge and discharge, and the current number of charge-discharge cycles as the corresponding number of cycles for charge and discharge. Then, it can be determined what the battery parameters of the target battery will change from the initial battery parameters after charging and discharging the target battery for the current number of charge-discharge cycles according to the current charge-discharge characteristics, so that the battery parameters of the target battery after charging and discharging for the current number of charge-discharge cycles can be determined.

[0109] In each round of prediction processing, in the manner of steps B1 to B2, select a suitable target prediction sub-model, which can realize the prediction of battery parameters under different charge-discharge characteristics, and finally obtain the total number of charge-discharge cycles of the target battery.

[0110] Step 210: Determine whether the battery parameters meet the stop prediction condition. If they meet, continue to step 211; otherwise, re-execute step 202, that is, perform the next round of prediction processing.

[0111] Step 211: Determine the predicted life of the target battery according to the total number of charge-discharge cycles determined by multiple rounds of battery parameter prediction processing.

[0112] The prediction method for the battery life of a rental vehicle provided by the embodiments of the present invention can predict the battery parameters of the battery after the vehicle battery has been charged and discharged a certain number of cycles. Even if the subsequent charge-discharge method of the battery (such as whether it is fast charging) changes, the vehicle life prediction can continue based on the changed charge-discharge behavior. After multiple predictions, the number of cycles that the battery can ultimately be charged and discharged can be predicted, realizing the prediction of the battery life. Based on the feature transfer matrix, the characteristics of continuous change of charge-discharge characteristics can be captured, which can handle the long-term rental scenario; based on the initial feature transfer matrix and probability distribution, the target feature transfer matrix can be accurately determined, which can ensure the stability of the probability distribution of charge-discharge characteristics.

[0113] The above details the prediction method process for the battery life of a rental vehicle. This method can also be implemented through a corresponding device. The following details the structure and functions of the device.

[0114] Based on the same inventive concept, the embodiments of the present invention also provide a prediction device for the battery life of a rental vehicle. As shown in Figure 4 The device includes: A model construction module 401, configured to pre-construct a prediction model for predicting battery parameters; A prediction module 402, configured to perform multiple rounds of prediction processing on the battery parameters of the target battery according to the prediction model until the battery parameters of the target battery after prediction meet the stop prediction condition; A lifespan determination module 403, configured to determine the predicted lifespan of the target battery according to the total number of charge-discharge cycles determined by the multi-round prediction processing of the battery parameters. Wherein, the prediction processing in the current round executed by the prediction module 402 includes: Determine the current charge-discharge characteristics of the target battery in the current round; Determine the current number of charge-discharge cycles corresponding to the current round, and add the current number of charge-discharge cycles to the total number of charge-discharge cycles; Use the battery parameters predicted by the previous round of prediction processing as the initial battery parameters for the current round of prediction processing. Using the prediction model, predict the battery parameters of the target battery according to the current charge-discharge characteristics and the current number of charge-discharge cycles, and determine the battery parameters of the target battery after charging and discharging the current number of charge-discharge cycles.

[0115] In some alternative embodiments, the current charge-discharge characteristics are determined based on the following method: According to the target feature transition matrix of the target battery, determine the charge-discharge characteristics transferred from the previous charge-discharge characteristics of the previous round, and use the transferred charge-discharge characteristics as the current charge-discharge characteristics in the current round; Wherein, the target feature transition matrix is used to represent the probability that the target battery transfers from one charge-discharge characteristic to another charge-discharge characteristic.

[0116] In some alternative embodiments, the step of determining the charge-discharge characteristics transferred from the previous charge-discharge characteristics of the previous round according to the target feature transition matrix of the target battery includes: Statistically analyze the transfer situation of the charge-discharge characteristics of the reference battery during the rental process to generate an initial feature transition matrix Q; Determine the target feature transition matrix P of the target battery according to the initial feature transition matrix Q, and: ; Wherein, represents the probability that the target battery is in the i-th charge-discharge characteristic, represents the probability that the target battery is in the j-th charge-discharge characteristic, i≠j; represents the probability of transferring from the i-th charge-discharge characteristic to the j-th charge-discharge characteristic in the initial feature transition matrix Q, represents the probability of transferring from the j-th charge-discharge characteristic to the i-th charge-discharge characteristic in the initial feature transition matrix Q, represents the probability of transferring from the i-th charge-discharge characteristic to the j-th charge-discharge characteristic in the target feature transition matrix P; According to the target feature transition matrix P, determine the previous charge-discharge characteristics of the previous round The probability distribution of various charge and discharge characteristics to be transferred, and select the current charge and discharge characteristic of the current round according to the probability distribution ; T represents the round of the prediction process.

[0117] In some alternative embodiments, the selecting the current charge and discharge characteristic of the current round according to the probability distribution , includes: According to the probability values corresponding to various charge and discharge characteristics in the probability distribution, divide the value range from 0 to 1 into multiple value ranges, and the size of each value range matches the probability value corresponding to the corresponding charge and discharge characteristic; Randomly generate a random number between 0 and 1, and use the charge and discharge characteristic corresponding to the value range into which the random number falls as the current charge and discharge characteristic of the current round .

[0118] In some alternative embodiments, the determining the current number of charge and discharge cycles corresponding to the current round includes: After determining the current charge and discharge characteristic, repeatedly perform the feature transfer operation until the next charge and discharge characteristic of the next round of prediction process is determined; The feature transfer operation includes: Determine the pending charge and discharge characteristic transferred from the current charge and discharge characteristic according to the target feature transfer matrix; If the pending charge and discharge characteristic is the same as the current charge and discharge characteristic, increment the current number of charge and discharge cycles by one, and then continue to perform the feature transfer operation; the initial value of the current number of charge and discharge cycles is 1; If the pending charge and discharge characteristic is different from the current charge and discharge characteristic, use the pending charge and discharge characteristic as the next charge and discharge characteristic of the next round of prediction process.

[0119] In some alternative embodiments, the prediction model includes prediction sub-models corresponding to various charge and discharge characteristics; the prediction sub-model is used to predict the change of the battery parameters of the target battery with the number of charge and discharge cycles when the target battery is charged and discharged according to the corresponding charge and discharge characteristics; The using the prediction model to predict the battery parameters of the target battery according to the current charge and discharge characteristic and the current number of charge and discharge cycles, and determining the battery parameters of the target battery after charging and discharging the current number of charge and discharge cycles includes: Determine the target prediction sub-model corresponding to the current charge and discharge characteristic; Input the initial battery parameters of the current round of prediction process and the current number of charge and discharge cycles into the target prediction sub-model, and determine the battery parameters of the target battery after charging and discharging the current number of charge and discharge cycles according to the prediction result of the target prediction sub-model.

[0120] An embodiment of the present invention further provides a computer storage medium, which stores computer-executable instructions including a program for executing the above-mentioned method for predicting the battery life of a rented vehicle. The computer-executable instructions can execute the method in any of the above-mentioned method embodiments.

[0121] Among them, the computer storage medium can be any available medium or data storage device accessible by a computer, including but not limited to magnetic memories (such as floppy disks, hard disks, magnetic tapes, magneto-optical discs (MO), etc.), optical memories (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor memories (such as ROM, EPROM, EEPROM, non-volatile memories (NANDFLASH), solid-state drives (SSD)), etc.

[0122] A part of the present invention can be applied as a computer program product, such as computer program instructions. When executed by a computer, through the operation of the computer, the methods and / or technical solutions according to the present invention can be invoked or provided. Those skilled in the art should understand that the forms of existence of computer program instructions in a computer-readable medium include but are not limited to source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include but are not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible by a computer.

[0123] Figure 5 The structural block diagram of an electronic device showing another embodiment of the present invention is presented. The electronic device 1100 can be a host server with computing capabilities, a personal computer PC, or a portable computer or terminal that can be carried, etc. The specific implementation of the electronic device is not limited in the specific embodiments of the present invention.

[0124] The electronic device 1100 includes at least one processor 1110, a communications interface 1120, a memory array 1130, and a bus 1140. Among them, the processor 1110, the communications interface 1120, and the memory 1130 communicate with each other through the bus 1140.

[0125] The communications interface 1120 is used to communicate with network elements, where the network elements include, for example, a virtual machine management center, a shared storage, etc.

[0126] The processor 1110 is used to execute programs. The processor 1110 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0127] The memory 1130 is used for executable instructions. The memory 1130 may include high-speed RAM memory and may also include non-volatile memory, such as at least one disk memory. The memory 1130 may also be a memory array. The memory 1130 may also be partitioned, and the partitions may be combined into virtual volumes according to certain rules. The instructions stored in the memory 1130 can be executed by the processor 1110, enabling the processor 1110 to execute the method for predicting the battery life of a rented vehicle in any of the above method embodiments.

[0128] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for predicting the battery life of a leased vehicle, characterized in that, Including: Pre - constructing a prediction model for predicting battery parameters; Performing multiple rounds of prediction processing on the battery parameters of the target battery according to the prediction model until the battery parameters after prediction of the target battery meet the stop - prediction condition; Determining the predicted life of the target battery according to the total number of charge - discharge cycles determined by multiple rounds of the battery - parameter prediction processing; Wherein, the prediction processing of the current round includes: Determining the current charge - discharge characteristics of the target battery in the current round; Determining the current charge - discharge cycles corresponding to the current round and adding the current charge - discharge cycles to the total number of charge - discharge cycles; Taking the battery parameters predicted by the previous - round prediction processing as the initial battery parameters of the current - round prediction processing, and using the prediction model to predict the battery parameters of the target battery according to the current charge - discharge characteristics and the current charge - discharge cycles, so as to determine the battery parameters of the target battery after charging and discharging the current charge - discharge cycles.

2. The method according to claim 1, wherein The current charge - discharge characteristics are determined based on the following method: According to the target characteristic transition matrix of the target battery, determining the charge - discharge characteristic transferred from the previous - charge - discharge characteristic of the previous round, and taking the transferred charge - discharge characteristic as the current charge - discharge characteristic in the current round; Wherein, the target characteristic transition matrix is used to represent the probability that the target battery transfers from one charge - discharge characteristic to another charge - discharge characteristic.

3. The method according to claim 2, wherein The determining the charge - discharge characteristic transferred from the previous - charge - discharge characteristic of the previous round according to the target characteristic transition matrix of the target battery includes: Statistically analyzing the transfer situation of the charge - discharge characteristics of the reference battery during the rental process to generate an initial characteristic transition matrix Q; Determining the target characteristic transition matrix P of the target battery according to the initial characteristic transition matrix Q, and: ； Among them, represents the probability that the target battery is in the i-th charge-discharge characteristic, represents the probability that the target battery is in the j-th charge-discharge characteristic, where i ≠ j; represents the probability of transitioning from the i-th charge-discharge characteristic to the j-th charge-discharge characteristic in the initial characteristic transition matrix Q, represents the probability of transitioning from the j-th charge-discharge characteristic to the i-th charge-discharge characteristic in the initial characteristic transition matrix Q, represents the probability of transitioning from the i-th charge-discharge characteristic to the j-th charge-discharge characteristic in the target characteristic transition matrix P; Determine, according to the target feature transfer matrix P, the probability distribution of various charge-discharge features transferred from the charge-discharge feature of the previous round of the previous charge-discharge cycle, and select the current charge-discharge feature of the current round according to the probability distribution ; T represents the round of the prediction process ; T represents the round of the prediction process 4. The method according to claim 3, wherein selecting the current charge and discharge feature of the current round according to the probability distribution , including: According to the probability values corresponding to various charge - discharge characteristics in the probability distribution, dividing the value range from 0 to 1 into multiple value ranges, and the size of each value range matches the probability value corresponding to the corresponding charge - discharge characteristic; Randomly generate a random number between 0 and 1, and use the charge-discharge characteristics corresponding to the value range into which the random number falls as the current charge-discharge characteristics for the current round .

5. The method according to claim 2, wherein The determining the current charge - discharge cycles corresponding to the current round includes: After determining the current charge - discharge characteristics, repeatedly performing the characteristic - transfer operation until determining the next charge - discharge characteristic of the next - round prediction processing; The characteristic - transfer operation includes: According to the target characteristic transition matrix, determining the to - be - determined charge - discharge characteristic transferred from the current charge - discharge characteristic; If the to - be - determined charge - discharge characteristic is the same as the current charge - discharge characteristic, incrementing the current charge - discharge cycles by 1, and then continuing to perform the characteristic - transfer operation; the initial value of the current charge - discharge cycles is 1; If the to - be - determined charge - discharge characteristic is different from the current charge - discharge characteristic, taking the to - be - determined charge - discharge characteristic as the next charge - discharge characteristic of the next - round prediction processing.

6. The method according to any one of claims 1 to 5, characterized in that The prediction model includes prediction sub - models corresponding to various charge - discharge characteristics; the prediction sub - models are used to predict the change of the battery parameters of the target battery with the number of charge - discharge cycles when the target battery is charged and discharged according to the corresponding charge - discharge characteristics. Using the prediction model, predicting the battery parameters of the target battery according to the current charge and discharge characteristics and the current number of charge and discharge cycles, and determining the battery parameters of the target battery after charging and discharging the current number of charge and discharge cycles, includes: Determining a target prediction sub-model corresponding to the current charge and discharge characteristics; Inputting the initial battery parameters of the current round of prediction processing and the current number of charge and discharge cycles into the target prediction sub-model, and determining the battery parameters of the target battery after charging and discharging the current number of charge and discharge cycles according to the prediction result of the target prediction sub-model.

7. A prediction device for the battery life of a leased vehicle, characterized in that, Includes: A model construction module for pre-constructing a prediction model for predicting battery parameters; A prediction module for performing multiple rounds of prediction processing on the battery parameters of the target battery according to the prediction model until the battery parameters after prediction of the target battery meet the stop prediction condition; A life determination module for determining the predicted life of the target battery according to the total number of charge and discharge cycles determined by multiple rounds of the battery parameter prediction processing; Wherein, the current round of prediction processing executed by the prediction module includes: Determining the current charge and discharge characteristics of the target battery in the current round; Determining the current number of charge and discharge cycles corresponding to the current round, and adding the current number of charge and discharge cycles to the total number of charge and discharge cycles; Using the battery parameters predicted by the previous round of prediction processing as the initial battery parameters of the current round of prediction processing, using the prediction model, predicting the battery parameters of the target battery according to the current charge and discharge characteristics and the current number of charge and discharge cycles, and determining the battery parameters of the target battery after charging and discharging the current number of charge and discharge cycles.

8. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions for executing the prediction method for the battery life of a rental vehicle according to any one of claims 1 to 6.

9. An electronic device, characterized in that, Includes: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, 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 execute the prediction method for the battery life of a rental vehicle according to any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions for causing a computer to execute the prediction method for the battery life of a rental vehicle according to any one of claims 1 to 6.

Images