1. A battery capacity prediction method based on model migration is characterized by comprising the following steps:

acquiring an industrial database of the battery;

determining first sensitive information according to the industrial database and the aging sensitive item;

determining a first intermediate capacity of the battery according to the trained base model and the first sensitive information;

and determining the predicted capacity of the battery according to the trained migration model and the first intermediate capacity.

2. The model migration based battery capacity prediction method of claim 1, further comprising:

when the first sensitive information is missing, determining a missing cycle serial number of the missing first sensitive information according to the industrial database;

and determining the completion capacity according to the trained completion model and the missing cycle number.

3. The method for predicting battery capacity based on model migration according to claim 1, wherein the training step of the base model specifically comprises:

acquiring aging experiment information of the battery;

determining second sensitive information according to the aging experiment information and the aging sensitive item;

determining the experimental capacity of the battery according to the aging experimental information;

and finishing the training of the base model according to the mapping relation between the second sensitive information and the experimental capacity.

4. The method for predicting battery capacity based on model migration according to claim 1, wherein the training step of the migration model specifically comprises:

acquiring battery maintenance information;

determining third sensitive information according to the battery maintenance information and the aging sensitive item;

determining the calibration capacity of the battery according to the battery maintenance information;

determining a second intermediate capacity of the battery according to the trained base model and the third sensitive information;

and finishing the training of the migration model according to the mapping relation between the calibration capacity and the second intermediate capacity.

5. The method for predicting battery capacity based on model migration according to claim 2, wherein the training step of the completion model specifically comprises:

determining a normal cycle number according to the industrial database;

determining the predicted capacity according to the normal cycle number;

and finishing the training of the completion model according to the mapping relation between the normal sequence number and the predicted capacity.

6. The model migration-based battery capacity prediction method according to claim 2, wherein the aging sensitivity term comprises internal resistance of the battery, a capacity increment track of the battery, and a charge-discharge curve of the battery.

7. The method for predicting battery capacity based on model migration according to claim 2, wherein the step of obtaining aging test information of the battery specifically includes:

carrying out accelerated aging treatment on the battery;

acquiring aging experiment information of the battery;

wherein the aging process includes cycling the battery using a high current.

8. A model migration based battery capacity prediction system, comprising:

the acquisition module is used for acquiring an industrial database of the battery;

the sensitive information extraction module is used for determining first sensitive information according to the industrial database and the aging sensitive item of the battery;

the middle capacity determining module is used for determining a first middle capacity of the battery according to the trained base model and the first sensitive information;

and the battery capacity estimation module is used for determining the predicted capacity of the battery according to the trained migration model and the first intermediate capacity.

9. An apparatus, comprising:

at least one processor;

at least one memory for storing at least one program;

when executed by the at least one processor, cause the at least one processor to implement the model migration-based battery capacity prediction method of any one of claims 1-7.

10. A computer storage medium having stored therein a processor-executable program, wherein the processor-executable program, when executed by the processor, is configured to implement the model migration based battery capacity prediction method of any one of claims 1-7.

Background

With the development of battery technology, the capacity and the service life of the battery are greatly improved. As a power source with wide application, the performance of the battery directly influences the performance of the whole battery driven by the battery, and the battery is not prevented from aging in the long-time use process, and the situations of capacity reduction, internal resistance increase and the like occur, so that the health management of the battery is very important. The related technology has a prediction method based on data driving, which mainly utilizes various parameters in the battery aging process to evaluate the health condition of the battery, and an important parameter of the battery health management is the capacity of the battery.

However, the battery capacity prediction based on data driving needs to rely on large-scale battery aging data, and because the battery life can often bear ten thousand charge-discharge cycles, it usually takes time and labor to obtain the battery aging data in a laboratory; and limited by laboratory resources, researchers can generally only obtain aging data of the battery under limited conditions (e.g., constant current conditions). In the practical use of the battery, a part of test data can be generated and stored in the industrial database, but under the influence of practical working conditions, the battery test data in the industrial database is always incomplete and is difficult to be directly used for the health management of the battery. The shortage of battery aging data directly affects the accuracy, reliability and generalization capability of the battery capacity estimation of the data-driven battery health management system.

Disclosure of Invention

The present application is directed to solving, at least to some extent, one of the technical problems in the related art. Therefore, the application provides a battery capacity prediction method, a system, a device and a medium based on model migration.

In a first aspect, an embodiment of the present application provides a method for predicting battery capacity based on model migration, including: acquiring an industrial database of the battery; determining first sensitive information according to the industrial database and the aging sensitive item; determining a first intermediate capacity of the battery according to the trained base model and the first sensitive information; and determining the predicted capacity of the battery according to the trained migration model and the first intermediate capacity.

Optionally, the method further comprises: and when the first sensitive information is lost, determining a lost cycle sequence number of the lost first sensitive information according to the industrial database, and determining a completion capacity according to a trained completion model and the lost cycle sequence number.

Optionally, the training step of the base model specifically includes: acquiring aging experiment information of the battery; determining second sensitive information according to the aging experiment information and the aging sensitive item; determining the experimental capacity of the battery according to the aging experimental information; and finishing the training of the base model according to the mapping relation between the second sensitive information and the experimental capacity.

Optionally, the training step of the migration model specifically includes: acquiring battery maintenance information; determining third sensitive information according to the battery maintenance information and the aging sensitive item; determining the calibration capacity of the battery according to the battery maintenance information; determining a second intermediate capacity of the battery according to the trained base model and the third sensitive information; and finishing the training of the migration model according to the mapping relation between the calibration capacity and the second intermediate capacity.

Optionally, the training step of the completion model specifically includes: determining a normal cycle number according to the industrial database; determining the predicted capacity according to the normal cycle number; and finishing the training of the completion model according to the mapping relation between the normal sequence number and the predicted capacity.

Optionally, the aging sensitive item includes internal resistance of the battery, a capacity increment track of the battery, and a charge-discharge curve of the battery.

Optionally, the step of acquiring aging test information of the battery specifically includes: carrying out accelerated aging treatment on the battery to obtain aging experiment information of the battery; wherein the aging process includes cycling the battery using a high current.

In a second aspect, an embodiment of the present application provides a battery capacity prediction system based on model migration, including: the acquisition module is used for acquiring an industrial database of the battery; the sensitive information extraction module is used for determining first sensitive information according to the industrial database and the aging sensitive item of the battery; the middle capacity determining module is used for determining a first middle capacity of the battery according to the trained base model and the first sensitive information; and the battery capacity estimation module is used for determining the predicted capacity of the battery according to the trained migration model and the first intermediate capacity.

In a third aspect, an embodiment of the present application provides an apparatus, including: at least one processor; at least one memory for storing at least one program; when the at least one program is executed by the at least one processor, the at least one program causes the at least one processor to implement the model migration-based battery capacity prediction method according to the first aspect.

In a fourth aspect, the present application provides a computer storage medium, in which a processor-executable program is stored, and the processor-executable program is used to implement the battery capacity prediction method based on model migration according to the first aspect when executed by the processor.

The beneficial effects of the embodiment of the application are as follows: firstly, acquiring an industrial database of a battery, and determining first sensitive information from the industrial database according to an aging sensitive item of battery aging characteristics in battery parameters; inputting the first sensitive information into the trained base model to complete the mapping from the first sensitive information to the first intermediate capacity of the battery; and inputting the first intermediate capacity into the trained migration model to complete the mapping from the first intermediate capacity to the predicted capacity. According to the method and the device, incomplete data in the industrial database are utilized through the base model and the migration model, a large amount of predicted capacity is generated, and the time for carrying out a large amount of battery aging experiments in a laboratory is saved. The generated predicted capacity data of the battery is beneficial to improving the accuracy, reliability and generalization capability of the battery health management system based on data driving to the estimation of the battery capacity.

Drawings

The accompanying drawings are included to provide a further understanding of the claimed subject matter and are incorporated in and constitute a part of this specification, illustrate embodiments of the subject matter and together with the description serve to explain the principles of the subject matter and not to limit the subject matter.

FIG. 1 is a flow chart illustrating steps of a method for predicting battery capacity based on model migration according to an embodiment of the present disclosure;

FIG. 2 is a flowchart illustrating the steps of training a base model according to an embodiment of the present disclosure;

FIG. 3 is a graph of a capacity increment trajectory provided by an embodiment of the present application;

FIG. 4 is a flowchart illustrating steps of migration model training provided by an embodiment of the present application;

FIG. 5 is a flowchart illustrating the steps of completion model training provided in an embodiment of the present application;

FIG. 6 is a graph of a battery capacity trace provided by an embodiment of the present application;

FIG. 7 is a schematic diagram of a model migration-based battery capacity prediction system according to an embodiment of the present disclosure;

fig. 8 is a device according to an embodiment of the present disclosure.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

It should be noted that although functional block divisions are provided in the system drawings and logical orders are shown in the flowcharts, in some cases, the steps shown and described may be performed in different orders than the block divisions in the systems or in the flowcharts. The terms first, second and the like in the description and in the claims, and the drawings described above, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

In actual use, batteries also generate a lot of data, and the data of the batteries are stored in an industrial database of the batteries, but the battery data in the industrial database is usually incomplete, and most of the cases do not contain the capacity information of the batteries. Therefore, the battery data in the industrial database cannot be directly used in the data-driven battery capacity prediction system, which results in a large amount of data waste in the industrial database, and much effort and time are required to obtain the battery aging data by performing the battery aging test under laboratory conditions.

Based on the above, the embodiment of the present application provides a battery capacity prediction method based on model migration, which can generate a predicted capacity of a battery by using a large amount of incomplete data in an industrial database through a base model and a migration model, and improve the accuracy and the generalization of a battery capacity prediction system based on data driving through generated battery capacity information.

The embodiments of the present application will be further explained with reference to the drawings.

Referring to fig. 1, fig. 1 is a flowchart illustrating steps of a method for predicting battery capacity based on model migration according to an embodiment of the present application, where the method includes, but is not limited to, steps S100-150.

S100, acquiring an industrial database of the battery;

specifically, an industrial database of the battery is obtained, the industrial database comprises test data of the battery in actual use, and most of the test data has battery capacity deficiency.

S110, determining first sensitive information according to an industrial database and an aging sensitive item of the battery;

specifically, some items sensitive to the aging process of the battery are contained in the industrial database, which are called aging sensitive items, and these aging sensitive items include, but are not limited to, the internal resistance of the battery, a capacity increment track, the curvature, the slope and the like of a charge-discharge curve, and according to these aging sensitive items, the first sensitive information can be determined from the industrial database, taking the capacity increment track of the battery as the aging sensitive item as an example, the first sensitive information can be a peak amplitude, a peak voltage and the like.

It should be noted that the other aging-sensitive items listed above (such as the internal resistance of the battery) may also be used to determine the first sensitive information, but the aging-sensitive item, such as the internal resistance of the battery, is not only sensitive to the aging process of the battery, but also sensitive to the temperature of the battery, the remaining capacity of the battery, and other information, that is, selecting the internal resistance of the battery as the aging-sensitive item may reduce the accuracy of the first sensitive information. Therefore, items more sensitive to the battery aging process are prioritized on the selection of aging sensitive items, and extraction of features when the battery is nearly fully charged or nearly empty is avoided as much as possible.

S120, determining a first intermediate capacity of the battery according to the trained base model and the first sensitive information;

specifically, the first sensitive information is input into a trained base model, and the base model outputs a first intermediate capacity of the battery, wherein the first intermediate capacity represents the capacity of the battery corresponding to the current first sensitive information, but the first intermediate capacity is obtained by calculating the sensitive information and has a certain deviation from the real capacity of the battery.

The training process of the base model is described above with reference to fig. 2, and fig. 2 is a flowchart illustrating steps of training the base model according to an embodiment of the present disclosure, where the process includes, but is not limited to, steps S200 to S230.

S200, acquiring aging experiment information of the battery;

specifically, the battery is subjected to accelerated aging treatment under laboratory conditions, and aging experimental information of the battery during the aging treatment is obtained. Aging processes include, but are not limited to, applying high and low temperatures to the battery, or cycling the battery using high currents, or overcharging, overdischarging, etc. the battery. Under laboratory conditions, the required complete battery test data can be acquired.

It will be appreciated that different types or models of batteries will have different data representations under the same test environment. If a certain amount of experimental data already exists in the current model battery, the experimental data can be directly used as the aging test information of the battery without being obtained through experiments again.

S210, determining second sensitive information according to the aging experiment information and the aging sensitive items;

specifically, the second sensitive information is extracted from the aging test information, similar to the first sensitive information above, in accordance with the aging sensitive item sensitive to the aging of the battery. Referring to fig. 3, fig. 3 is a graph of a capacity increment trace provided in the example of the present application, in which, as shown in fig. 3, the horizontal axis represents the voltage of the battery (in mV), and the vertical axis represents the capacity of the battery (in Ah/V) per unit voltage. Four second sensitive information can be extracted as shown in fig. 2: the horizontal black dotted line represents the peak amplitude (represented by IC ^ pk), the horizontal green dotted line represents 85% of the peak amplitude, the red vertical dotted line represents the peak voltage (represented by V ^ pk), the purple region represents the partial capacity of the cell contained within + -15 mV of the peak voltage of the curve (represented by Area 1), the pink region represents the partial capacity of the cell contained within a trace on the curve having an amplitude greater than 85% of the peak amplitude (represented by Area2), then the second sensitive information is IC ^ pk, V ^ pk, Area1, and Area 2.

S220, determining the experimental capacity of the battery according to the aging experimental information;

specifically, since the aging test information obtained in the laboratory environment is relatively complete, corresponding to the second sensitive information, the corresponding battery capacity can be found from the aging test information, and the capacity is referred to as the test capacity. The experimental capacity represents the capacity of the battery in an aging experiment, taking a large current as an example of accelerated aging treatment, the experimental capacity of the battery in the current cycle can be determined corresponding to each charge and discharge cycle in a large current environment, and of course, the experimental capacity has a mapping relation with second sensitive information in the current cycle.

S230, completing training of the base model according to the mapping relation between the second sensitive information and the experimental capacity;

in particular, it was mentioned above that the experimental capacity of the battery is mapped to the second sensitive information in the current cycle, and therefore the base model can be trained using this mapping. In the embodiment of the application, the base model is composed of three layers of feedforward neural networks, and the Bayesian regularization algorithm is adopted to train the networks. The neural network activation function is selected to have no saturation effect, and needs to be as simple as possible and have local linear characteristics. Therefore, the generalization performance of the neural network can be effectively improved, and therefore, a ReLu function can be selected, and similar functions also include Gaussian-ReLU, E-LU, Leaky-ReLU and the like. The number of hidden layer nodes of the network is 20 and biases are applied to the hidden layers and the output layers. In the example of using the capacity increment trajectory as the aging sensitive item, the above-mentioned four second sensitive information (refer to IC ^ pk, V ^ pk, Area1, and Area2) are input to the base model, and the base model outputs the experimental capacity corresponding to the battery. And finishing the training of the base model by using the determined second sensitive information and the experimental capacity.

Through the steps S200-S240, the aging test information of the battery is utilized to complete the training of the base model, and the main significance of the base model is that the model embodies the mapping relation between the aging sensitive information of the battery and the experimental capacity in the aging process under the laboratory condition, so that the sensitive information data in the industrial database can be input by utilizing the base model, and the capacity of the battery can be determined. However, due to incomplete data in the industrial database and objective deviation of the aging performance of the battery caused by difference between the laboratory environment and the industrial environment, further correction of the data obtained by the base model is required.

The following begins to explain step S130 in fig. 1.

S130, determining the predicted capacity of the battery according to the trained migration model and the first intermediate capacity;

specifically, the first intermediate capacity is input into a trained migration model, and the migration model outputs the predicted capacity of the battery. Through the base model and the migration model, the method and the device generate predicted capacity data of a large number of batteries by using incomplete data in the industrial database.

Through steps S100 to S130, the embodiment of the present application utilizes incomplete data in the industrial database through the base model and the migration model to generate a large amount of predicted capacity, thereby saving time for performing a large amount of battery aging experiments in a laboratory. The generated predicted capacity data of the battery is beneficial to improving the accuracy, reliability and generalization capability of the battery health management system based on data driving to the estimation of the battery capacity.

The above training process of the migration model refers to fig. 4, and fig. 4 is a flowchart illustrating steps of training the migration model according to the embodiment of the present application, where the process includes, but is not limited to, steps S400 to S440.

S400, obtaining battery maintenance information;

specifically, the industry regularly maintains the battery to obtain battery maintenance information, which is relatively complete battery parameter information in the industrial environment. It should be noted that, in an industrial environment, the maintenance frequency of the battery in the whole life cycle is very low, that is, the battery maintenance information only contains a few times of maintenance data with a small number.

S410, determining third sensitive information according to the battery maintenance information and the aging sensitive items;

specifically, third sensitive information is extracted from the battery maintenance information according to the aging sensitive item, the third sensitive information represents the aging sensitive information in the battery aging process in the industrial environment, and the third sensitive information is similar to the first sensitive information and the second sensitive information and is not repeated here.

S420, determining the calibration capacity of the battery according to the battery maintenance information;

specifically, since the battery maintenance information includes the more complete battery parameters, corresponding to the third sensitive information, the corresponding battery capacity can be found from the battery maintenance information, and the capacity is referred to as the calibration capacity of the battery. The calibration capacity represents the battery capacity of the battery in the industrial environment, the battery is normally used, and the calibration capacity in the current cycle can be determined every charge and discharge cycle, so that a mapping relation exists between the calibration capacity and the third sensitive information in the current cycle, and the mapping relation can comparatively reflect the normal aging process of the battery in the industrial environment.

S430, determining a second intermediate capacity of the battery according to the trained base model and the third sensitive information;

specifically, the base model is trained through the method steps in fig. 2, and since the base model is a mapping for representing sensitive information to the battery capacity, the third sensitive information is input to the base model, and the base model outputs the capacity of the battery, which is referred to as a second intermediate capacity. And the second intermediate capacity represents the battery capacity corresponding to the third sensitive information in the laboratory environment.

S440, completing the training of the migration model according to the mapping relation between the calibration capacity and the second intermediate capacity;

specifically, it is mentioned above that since the aging performance of the battery has an objective deviation due to the difference between the laboratory environment and the industrial environment, the second intermediate capacity generated by the base model in step S440 has a large deviation from the actual capacity (i.e., the calibration capacity) of the battery, and therefore, it is necessary to use the migration model to map the second intermediate capacity to the calibration capacity, that is, to use a small number of accurate aging process data of the battery in the industrial environment to migrate the base model from the laboratory environment to the industrial environment, and to use the migration model to complete the mapping of the intermediate capacity to the actual capacity of the battery, so that the combination of the base model and the migration model is suitable for use in the industrial database.

Through steps S400 to S440, the training of the migration model is completed by using the battery maintenance information, and the main significance of the migration model is that the model represents a mapping relationship between the second intermediate capacity generated by the base model and the actual calibration capacity of the battery under the industrial condition, so that the actual capacity of the battery can be generated by inputting the intermediate capacity generated by the base model by using the migration model.

The following begins to describe steps S140-S150 in fig. 1.

S140, when the first sensitive information is missing, determining a missing cycle number of the missing first sensitive information according to the industrial database;

specifically, as can be seen from the above steps S100 to S130, the predicted capacity data of the battery can be generated by the base model and the migration model by extracting the first sensitive information from the industrial database. However, there still exists some data with serious damage in the industrial database, for example, in the test data of the current battery cycle charging and discharging, the first sensitive information cannot be extracted, and then the cycle number missing the first sensitive information in the test cycle of the current battery is determined, and these cycle numbers are called missing cycle numbers.

S150, determining the completion capacity according to the trained completion model and the missing cycle number;

specifically, the obtained missing cycle number is input into a trained completion model, and the completion model outputs the battery capacity corresponding to the missing cycle number, which is called the completion capacity. By means of the completion model, the embodiment of the application realizes completion of partial seriously incomplete data in the industrial database.

Reference may be made to fig. 5 for a training process of the completion model, where fig. 5 is a flowchart illustrating steps of training the completion model according to an embodiment of the present application, and the process includes, but is not limited to, steps S500 to S520.

S500, determining a normal cycle number according to an industrial database;

specifically, it is mentioned above that the cycle number missing the first sensitive information in the test cycle of the current battery is referred to as a missing cycle number, and similarly, the cycle number not missing the first sensitive information is referred to as a normal cycle number. And extracting a plurality of normal cycle serial numbers from the industrial database. It should be noted that the extracted normal cycle number is suitable before and after the missing cycle number, because the cycle before and after the missing cycle number is similar to the cycle condition of the missing sensitive information. For example, if the capacity corresponding to the current battery cycle of 100 to 105 is lost, a completion model can be established by using data of 90 to 100 cycles and 106 to 120 cycles.

S510, determining a predicted capacity according to the normal cycle number;

specifically, in the cycle corresponding to the normal cycle number, by the method for predicting the battery capacity based on the model migration provided in the embodiment of the present application, the predicted capacity of the battery can be generated through the aging sensitive information, so that the predicted capacity can be determined according to the normal cycle number.

And S520, finishing training of the complete model according to the mapping relation between the normal sequence number and the predicted capacity.

Specifically, the completion model is trained through the mapping relation between the normal serial number and the predicted capacity, the cyclic serial number is input into the completion model, and the capacity of the battery is output by the model. The completion model generally uses an interpolation model, and the interpolation method includes, but is not limited to, linear interpolation, polynomial interpolation, neural network interpolation, and the like.

Through steps S550-S520, the completion model is trained using the sequence numbers of the normal cycles, and when the completion model training is completed, the missing cycle sequence numbers are input to the completion model to generate the corresponding predicted capacity, so that the completion of the battery capacity data can be completed even if part of the industrial data is seriously missing.

Referring to fig. 6, fig. 6 is a battery capacity trajectory diagram provided in an embodiment of the present application, and fig. 6 corresponds to the method for predicting battery capacity based on model migration illustrated in fig. 1. As shown in fig. 6, the horizontal axis represents the cycle number, and the vertical axis represents the battery capacity. The blue curve represents a first intermediate capacity (represented by Base model) generated by the Base model using first sensitive information of the industrial database, the red star points (total 3, represented by Labeled samples) represent battery maintenance information, the migration model is used to characterize a mapping of the first intermediate capacity to a battery calibration capacity, and the orange dashed line represents a predicted capacity (represented by Actual capacity) of the battery generated by the migration model using the first intermediate capacity of the Base model.

In summary, according to the battery capacity prediction method based on model migration provided by the embodiment of the present application, an industrial database of a battery is first obtained, and first sensitive information is determined from the industrial database according to an aging sensitive item of battery aging characteristics in battery parameters; inputting the first sensitive information into the trained base model to complete the mapping from the first sensitive information to the first intermediate capacity of the battery; and inputting the first intermediate capacity into the trained migration model to complete the mapping from the first intermediate capacity to the predicted capacity. And for the data with serious defects in the industrial database, completing by using a completion model. According to the method and the device, incomplete data in the industrial database are utilized through the base model and the migration model, a large amount of predicted capacity is generated, and the time for carrying out a large amount of battery aging experiments in a laboratory is saved. The generated predicted capacity data of the battery is beneficial to improving the accuracy, reliability and generalization capability of the battery health management system based on data driving to the estimation of the battery capacity.

Referring to fig. 7, fig. 7 is a schematic diagram of a battery capacity prediction system based on model migration according to an embodiment of the present application, where the system 700 includes an obtaining module 710, a sensitive information extracting module 720, an intermediate capacity determining module 730, and a battery capacity estimating module 740. The acquisition module is used for acquiring an industrial database of the battery; the sensitive information extraction module is used for determining first sensitive information according to an industrial database and an aging sensitive item of the battery; the middle capacity determining module is used for determining a first middle capacity of the battery according to the trained base model and the first sensitive information; the battery capacity estimation module is used for determining the predicted capacity of the battery according to the trained migration model and the first intermediate capacity.

Referring to fig. 8, fig. 8 illustrates an apparatus according to an embodiment of the present application, where the apparatus 800 includes at least one processor 810 and at least one memory 820 for storing at least one program; in fig. 8, a processor and a memory are taken as an example.

The processor and memory may be connected by a bus or other means, such as by a bus in FIG. 8.

The memory, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs as well as non-transitory computer executable programs. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and these remote memories may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

Another embodiment of the present application also provides an apparatus that may be used to perform the control method as in any of the embodiments above, for example, performing the method steps of fig. 1 described above.

The above-described embodiments of the apparatus are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, i.e. may be located in one place, or may also be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.

The embodiment of the application also discloses a computer storage medium, wherein a program executable by a processor is stored, and the program executable by the processor is used for realizing the battery capacity prediction method based on model migration when being executed by the processor.

One of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as is well known to those of ordinary skill in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media as known to those skilled in the art.

While the preferred embodiments of the present invention have been described, the present invention is not limited to the above embodiments, and those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention, and such equivalent modifications or substitutions are included in the scope of the present invention defined by the claims.