CN115805810A - Battery failure prediction method, apparatus, device, storage medium, and program product - Google Patents

Abstract

The present application relates to a battery failure prediction method, apparatus, device, storage medium and program product. The method comprises the following steps: acquiring state characteristic data of a target battery in multiple periods, performing correlation analysis processing on the state characteristic data in the multiple periods to obtain correlation analysis processing results, and determining a fault prediction result of the target battery according to the correlation analysis processing results; wherein the state characteristic data in each cycle comprises a plurality of different battery characteristic data. In the method, because the real-time state characteristics in each period have a plurality of different battery characteristic data, equivalently, the battery characteristic data under a plurality of periods are combined for analysis, so that whether the battery has faults or not is predicted from the battery characteristics of a plurality of dimensions, and the accuracy of the battery fault prediction result is improved.

Description

Battery failure prediction method, device, equipment, storage medium and program product

Technical Field

The present application relates to the field of battery technologies, and in particular, to a method, an apparatus, a device, a storage medium, and a program product for predicting a battery failure.

Background

With the rapid development of the battery industry, batteries are more and more widely used, for example, in many new energy vehicles.

However, with the increasing of the sales volume of new energy vehicles, the battery capacity and the charging technology, more and more batteries have safety problems, for example, if the batteries are charged high or low, certain potential safety hazards exist, and therefore, the failure monitoring of the batteries in the vehicles is performed, and safety accidents are avoided.

However, there is a lack of a method for accurately predicting a battery failure in the related art.

Disclosure of Invention

In view of the above, it is necessary to provide a battery failure prediction method, device, apparatus, storage medium, and program product, which can accurately predict whether a battery has a failure.

In a first aspect, the present application provides a battery failure prediction method, including:

acquiring state characteristic data of a target battery in a plurality of periods; the state characteristic data in each period comprise a plurality of different battery characteristic data;

performing correlation analysis processing on the state characteristic data in a plurality of periods to obtain correlation analysis processing results;

and determining a fault prediction result of the target battery according to the correlation analysis processing result.

In the embodiment of the application, state characteristic data of a target battery in multiple periods are obtained, correlation analysis processing is carried out on the state characteristic data in the multiple periods to obtain correlation analysis processing results, and then a fault prediction result of the target battery is determined according to the correlation analysis processing results; wherein the state characteristic data in each cycle comprises a plurality of different battery characteristic data. In the embodiment, because the state characteristic data in multiple cycles of the battery are subjected to the correlation analysis processing, and the real-time state characteristic in each cycle has multiple different battery characteristic data, the multiple different battery characteristic data in the multiple cycles of the battery are equivalently combined for analysis, so that whether the battery has a fault or not is predicted from the battery characteristics in multiple dimensions, and the accuracy of the battery fault prediction result is improved.

In one embodiment, performing correlation analysis processing on the state feature data in multiple cycles to obtain a correlation analysis processing result includes:

and inputting the state characteristic data in the multiple periods into the fault prediction model, and performing correlation analysis processing on the state characteristic data in the multiple periods through the fault prediction model to obtain a correlation analysis processing result.

In the embodiment of the application, the state characteristic data is subjected to correlation analysis through the fault prediction model, and the fault prediction model is constructed through historical data in advance, so that when the fault of the battery is predicted, the fault prediction result of the target battery can be obtained by only calling the fault prediction model and analyzing the acquired state characteristic data in a plurality of periods, and the fault detection efficiency of the target battery is improved; the fault prediction model is constructed in advance through historical data, the historical data are some characteristic data belonging to the battery, and the characteristic data can accurately reflect the characteristics of the battery, so that the fault of the battery is predicted through the fault prediction model, and the accuracy of a fault prediction result of the battery is also ensured.

In one embodiment, performing correlation analysis processing on state feature data in multiple cycles through a fault prediction model to obtain a correlation analysis processing result, including:

acquiring abnormal battery characteristic distribution information in each period and abnormal battery characteristic change trend information in a plurality of periods according to the state characteristic data in the plurality of periods;

and performing correlation analysis processing on the battery characteristic abnormal distribution information in each period and the battery characteristic abnormal change trend information in a plurality of periods to obtain a correlation analysis processing result.

In the embodiment of the application, according to the state characteristic data in a plurality of periods, the battery characteristic abnormal distribution information in each period is obtained, the battery characteristic abnormal change trend information in a plurality of periods is obtained, and the battery characteristic abnormal distribution information in each period and the battery characteristic abnormal change trend information in the plurality of periods are subjected to correlation analysis processing to obtain a correlation analysis processing result. In the embodiment, the battery characteristic abnormal distribution information in each period of the target battery and the battery characteristic abnormal change trend information in a plurality of periods are comprehensively analyzed, and joint analysis is performed on the information in a plurality of dimensions, so that the accuracy of target battery fault prediction in the follow-up process is improved.

In one embodiment, the fault prediction model includes a classification model and a time sequence model, and the obtaining of the abnormal distribution information of the battery characteristics in each period and the abnormal change trend information of the battery characteristics in a plurality of periods according to the state characteristic data in the plurality of periods includes:

inputting a plurality of different battery characteristic data in the state characteristic data in each period into a classification model to obtain battery characteristic abnormal distribution information in each period;

and inputting a plurality of different battery characteristic data in the state characteristic data in each period into the time sequence model to obtain abnormal variation trend information of the battery characteristics in a plurality of periods.

In the embodiment of the application, a plurality of different battery characteristic data in the state characteristic data in each period are input into the classification model to obtain battery characteristic abnormal distribution information in each period, and a plurality of different battery characteristic data in the state characteristic data in each period are input into the time sequence model to obtain battery characteristic abnormal change trend information in a plurality of periods. In the embodiment, the state characteristic data in a plurality of periods are analyzed through the classification model and the time sequence model respectively to obtain battery characteristic abnormal distribution information in each period and battery characteristic abnormal variation trend in each period, and the battery characteristic data in two dimensions in each period and in each period are considered comprehensively, so that the accuracy of a target battery fault prediction result obtained subsequently is ensured.

In one embodiment, the failure prediction result includes that the target battery has a failure or does not have a failure, and the method further includes:

under the condition that the target battery has a fault, acquiring at least one abnormal battery cell with high charging and discharging risks in the target battery;

acquiring fault risk types of each abnormal battery cell;

and determining the fault risk degree of each abnormal battery cell according to the fault risk type.

In the embodiment of the application, under the condition that a target battery has a fault, at least one abnormal battery cell with high charging and discharging risks in the target battery is obtained, fault risk types of the abnormal battery cells are obtained, and then fault risk degrees of the abnormal battery cells are determined according to the fault risk types. In the embodiment, the fault risk types of the abnormal electric cores in the target battery are obtained, and the fault risk degree of the abnormal electric cores is determined based on the fault risk types, so that the fault accuracy of the target battery is further improved, and the target battery is convenient to overhaul subsequently.

In one embodiment, the acquiring the fault risk types of the abnormal battery cells includes:

for any abnormal electric core, dividing the state characteristic data of the abnormal electric core to obtain the internal resistance characteristic data and the capacity characteristic data of the abnormal electric core;

determining the fault risk type of the abnormal cell as the internal resistance abnormality under the condition that the abnormality exists in the internal resistance characteristic data of the abnormal cell;

and under the condition that the abnormal capacity characteristic data of the abnormal battery cell exists, determining the fault risk type of the abnormal battery cell as capacity abnormality.

In the embodiment of the application, for any abnormal battery cell, state characteristic data of the abnormal battery cell is divided to obtain internal resistance characteristic data and capacity characteristic data of the abnormal battery cell, and when abnormality exists in the internal resistance characteristic data of the abnormal battery cell, the fault risk type of the abnormal battery cell is determined to be internal resistance abnormality; and under the condition that the abnormal capacity characteristic data of the abnormal battery cell exists, determining the fault risk type of the abnormal battery cell as capacity abnormality. In this embodiment, by dividing the real-time state data of the abnormal electrical core into the internal resistance type characteristic data and the capacity type characteristic data, the accuracy of identifying whether the abnormal electrical core is abnormal in internal resistance or abnormal in capacity is improved, the fault type of the abnormal electrical core is determined, and the abnormal electrical core is convenient to overhaul.

In one embodiment, determining the fault risk degree of each abnormal battery cell according to the fault risk type includes:

for any abnormal battery cell, determining target battery characteristic data from the state characteristic data of the abnormal battery cell according to the fault risk type of the abnormal battery cell, wherein the target battery characteristic data comprises at least one piece of battery characteristic data;

and determining the fault risk degree of the abnormal electric core according to the target battery characteristic data of the abnormal electric core.

In the embodiment of the application, for any abnormal electric core, target battery characteristic data is determined from the state characteristic data of the abnormal electric core according to the fault risk type of the abnormal electric core, and then the fault risk degree of the abnormal electric core is determined according to the target battery characteristic data of the abnormal electric core; wherein the target battery characteristic data comprises at least one battery characteristic data. In this embodiment, the fault risk degree of the abnormal electric core is determined according to at least one battery characteristic data selected from the state characteristic data of the abnormal electric core, so that the accuracy of determining the fault risk degree of the abnormal electric core is improved.

In one embodiment, determining target battery characteristic data from state characteristic data of an abnormal cell according to a fault risk type of the abnormal cell includes:

determining target battery characteristic data of the abnormal electric core from the internal resistance characteristic data in the state characteristic data of the abnormal electric core under the condition that the fault risk type of the abnormal electric core is abnormal internal resistance;

and under the condition that the fault risk type of the abnormal battery cell is abnormal in capacity, determining target battery characteristic data of the abnormal battery cell from capacity characteristic data in the state characteristic data of the abnormal battery cell.

In the embodiment of the application, when the fault risk type of the abnormal battery cell is that the internal resistance is abnormal, the target battery characteristic data of the abnormal battery cell is determined from the internal resistance characteristic data in the state characteristic data of the abnormal battery cell, and when the fault risk type of the abnormal battery cell is that the capacity is abnormal, the target battery characteristic data of the abnormal battery cell is determined from the capacity characteristic data in the state characteristic data of the abnormal battery cell. In this embodiment, the accuracy of determining the fault risk degree of the abnormal electric core through the target battery characteristic data corresponding to the fault risk type of the abnormal electric core is improved.

In one embodiment, determining the fault risk degree of an abnormal cell according to the target battery characteristic data of the abnormal cell includes:

acquiring target battery characteristic data of other normal electric cores except the abnormal electric core in the target battery;

and transversely comparing the target battery characteristic data of the abnormal battery cell with the target battery characteristic data of other normal battery cells to determine the fault risk degree of the abnormal battery cell.

In the embodiment of the application, the target battery characteristic data of other normal electric cores in the target battery except the abnormal electric core is obtained, and the target battery characteristic data of the abnormal electric core and the target battery characteristic data of other normal electric cores are transversely compared to determine the fault risk degree of the abnormal electric core. In the embodiment, the accuracy of determining the fault risk degree of the abnormal battery cell is ensured by transversely comparing the abnormal data with the normal data.

In one embodiment, if the target battery characteristic data is an internal resistance value, transversely comparing the target battery characteristic data of the abnormal cell with the target battery characteristic data of other normal cells to determine a fault risk degree of the abnormal cell, includes:

acquiring the internal resistance value of the abnormal cell and the median internal resistance value of the abnormal cell and other normal cells;

obtaining the ratio of the internal resistance value to the internal resistance value of the median;

and determining the fault risk degree of the abnormal battery cell according to the ratio and a preset plurality of different fault risk degree grade ranges.

In the embodiment of the application, the internal resistance value of the abnormal electric core is acquired, the median internal resistance value of the abnormal electric core and other normal electric cores is acquired, the ratio of the internal resistance value to the median internal resistance value is acquired, and the fault risk degree of the abnormal electric core is determined according to the ratio and the preset fault risk degree grade range of a plurality of differences. According to the method, the internal resistance value of the abnormal battery cell and the median internal resistance values of other normal battery cells are considered, the fault risk degree of the abnormal battery cell is determined according to the ratio and the preset different fault risk degree grade ranges, and the accuracy of the determined fault risk degree of the abnormal battery cell is guaranteed.

In a second aspect, the present application also provides a battery failure prediction apparatus, including:

the acquisition module is used for acquiring state characteristic data of the target battery in a plurality of periods; the state characteristic data in each period comprise a plurality of different battery characteristic data;

the analysis module is used for performing correlation analysis processing on the state characteristic data in a plurality of periods to obtain correlation analysis processing results;

and the determining module is used for determining a fault prediction result of the target battery according to the correlation analysis processing result.

In a third aspect, an embodiment of the present application provides a computer device, which includes a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the method provided in any one of the foregoing first aspects when executing the computer program.

In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method provided in any one of the embodiments in the first aspect.

In a fifth aspect, the present application further provides a computer program product, including a computer program, where the computer program implements the steps of the method provided in any one of the foregoing embodiments of the first aspect when executed by a processor.

The above description is only an overview of the technical solutions of the present application, and the present application may be implemented in accordance with the content of the description so as to make the technical means of the present application more clearly understood, and the detailed description of the present application will be given below in order to make the above and other objects, features, and advantages of the present application more clearly understood.

Drawings

Various additional advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like parts are designated by like reference numerals throughout the drawings. In the drawings:

FIG. 1 is a diagram of an exemplary embodiment of a battery failure prediction method;

FIG. 2 is a schematic flow chart diagram illustrating a method for battery failure prediction in one embodiment;

FIG. 3 is a schematic flow chart diagram of a battery failure prediction method in another embodiment;

FIG. 4 is a schematic flow chart diagram of a battery failure prediction method in another embodiment;

FIG. 5 is a schematic flow chart diagram of a battery failure prediction method in another embodiment;

FIG. 6 is a schematic flow chart diagram illustrating a battery failure prediction method according to another embodiment;

FIG. 7 is a schematic flow chart diagram illustrating a battery failure prediction method according to another embodiment;

FIG. 8 is a schematic flow chart diagram of a battery failure prediction method in another embodiment;

FIG. 9 is a schematic flow chart diagram of a battery failure prediction method in another embodiment;

FIG. 10 is a schematic flow chart diagram illustrating a battery failure prediction method according to another embodiment;

FIG. 11 is a schematic flow chart diagram illustrating a method for predicting battery failure in another embodiment;

FIG. 12 is a schematic flow chart diagram of a battery failure prediction method in another embodiment;

FIG. 13 is a block diagram showing the construction of a battery failure prediction apparatus according to an embodiment;

FIG. 14 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following examples are merely used to more clearly illustrate the technical solutions of the present application, and therefore are only examples, and the protection scope of the present application is not limited thereby.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "including" and "having," and any variations thereof, in the description and claims of this application and the description of the above figures are intended to cover non-exclusive inclusions.

In the description of the embodiments of the present application, the technical terms "first", "second", and the like are used only for distinguishing different objects, and are not to be construed as indicating or implying relative importance or implicitly indicating the number, specific order, or primary-secondary relationship of the technical features indicated. In the description of the embodiments of the present application, "a plurality" means two or more unless specifically defined otherwise. Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.

In the description of the embodiments of the present application, the term "plurality" refers to two or more (including two).

In practical application, the battery charging and discharging phenomena can seriously affect the service performance of the battery. In the related technology, the phenomenon of high and low charging of the battery is monitored, and the battery data is mainly obtained and subjected to data analysis to determine whether the battery data is within a safety threshold value or not, and when the phenomenon of high and low charging of the battery occurs, the battery data exceeds the safety threshold value.

The applicant researches and discovers that the monitoring of whether the battery is charged or not through comparing the battery data with the safety threshold is that the monitoring can be carried out on the battery which is charged or not, and the monitoring can not be used for predicting whether the battery is charged or not.

Based on the above considerations, in order to predict the phenomena of battery charging and discharging, the inventors have conducted extensive research and propose a battery failure prediction method, which determines the failure prediction result of the battery by performing correlation analysis on a plurality of different battery characteristic data of the battery in a plurality of cycles.

In the battery fault prediction method, the battery characteristic distribution information in each period and the battery characteristic variation trend in a plurality of periods are subjected to correlation analysis, so as to determine whether the battery is likely to generate the charging and discharging phenomena, and the accuracy of the battery fault prediction result is improved.

Of course, it should be understood that the technical effects that can be achieved by the battery fault prediction method provided in the embodiment of the present application are not limited thereto, and other technical effects can also be achieved, for example, the high-low charging phenomenon of the battery is predicted, and the safety risk caused by the high-low charging phenomenon of the battery is reduced. Specific examples of the technical effects achieved in the embodiments of the present application can be found in the following embodiments.

It should be noted that the battery fault prediction scheme provided by the present application is applicable to all new energy related fields, including but not limited to batteries in new energy vehicles and energy storage and quick change modules, and is not limited in this embodiment of the present application.

The battery disclosed in the embodiment of the present application can be used in electric devices such as vehicles, ships or aircrafts, but not limited thereto.

The embodiment of the application provides an electric device using a battery as a power supply, wherein the electric device can be but is not limited to a mobile phone, a tablet, a notebook computer, an electric toy, an electric tool, a battery car, an electric automobile, a ship, a spacecraft and the like. The electric toy may include a stationary or mobile electric toy, such as a game machine, an electric car toy, an electric ship toy, an electric airplane toy, and the like, and the spacecraft may include an airplane, a rocket, a space shuttle, a spacecraft, and the like.

For convenience of description, an electric device according to an embodiment of the present application is described below, and as shown in fig. 1, fig. 1 is a schematic structural diagram of an electric device 100 provided in the embodiment of the present application, in which a target battery 102 is a power source of the electric device 100.

Next, the failure warning of the target battery will be described using a controller in the electric device as an execution subject (the controller is not shown in fig. 1).

In one embodiment, as shown in fig. 2, a battery failure prediction method is provided, which is described by taking an example of applying the method to a controller of the electric device in fig. 1, and includes the following steps:

s201, acquiring state characteristic data of a target battery in multiple periods; wherein the state characteristic data in each cycle comprises a plurality of different battery characteristic data.

The target battery is a battery needing fault identification, and at least comprises one battery cell; the state characteristic data may include state characteristic data of the target battery, and may also include state characteristic data of a battery cell in the target battery, where the battery characteristic data includes a plurality of state characteristic data.

Therefore, when the target battery needs to be subjected to fault identification, the state characteristic data of the target battery in a plurality of cycles can be acquired, and the prediction result of the fault of the battery can be judged according to the state characteristic data in the plurality of cycles.

Each of the plurality of periods may be a preset time duration, and the preset time duration is variable, for example, 1 day, 3 days, 7 days, or a charging period, a discharging period, or a charging and discharging period of the whole battery; the plurality of periods may be consecutive periods; it is to be understood that the period is preset, the period does not change any more in the process of acquiring the state characteristic data of a plurality of periods, and the plurality of periods includes at least two periods, and in the embodiment of the present application, the specific number of periods is not limited in the present application.

The battery characteristic data includes: charging high frequency, discharging low frequency, differential pressure current slope, differential pressure State of Charge (SOC) slope, internal resistance, etc.

Wherein, the charging frequency represents the frequency of the charging phenomenon in a preset period; the lowering frequency represents a frequency at which a lowering phenomenon occurs within a preset one period; the pressure difference represents the pressure difference of all the cells in the battery in one period; the voltage difference current slope represents the slope of a curve between the voltage difference value and the current value of a battery cell in the battery; the differential pressure SOC slope represents the slope of a curve between the cell differential pressure and the SOC value of the battery in one period; the internal resistance value represents the internal resistance value of all the battery cells in one cycle of the battery.

Optionally, the voltage difference may be a difference between a maximum voltage of the battery cell and a minimum voltage of the battery cell in one period, and the voltage difference may also be a difference between an average maximum voltage of the battery cell and an average minimum voltage of the battery cell in one period; the internal resistance value may be an average value of all the battery cells of the battery in one cycle, or may be an internal resistance of each battery cell calculated according to the measured current and voltage of the battery cell, and then an internal resistance of the battery calculated based on the internal resistances of the battery cells.

In the embodiment of the present application, the calculation method of the battery characteristic data is not limited, and may be determined according to actual requirements.

In one embodiment, initial characteristic data of each battery cell in the target battery in multiple cycles can be acquired through a sensor, and then state characteristic data of the target battery in multiple cycles is determined based on the initial characteristic data of each battery cell and the calculation mode; the state characteristic data of the target battery in a plurality of cycles can be directly measured or indirectly calculated.

And S202, performing correlation analysis processing on the state characteristic data in a plurality of periods to obtain a correlation analysis processing result.

The charging and discharging phenomena of the battery cannot be directly determined through a variable test value, can be represented only through a correlation phenomenon, and can be comprehensively judged through state characteristic data in a plurality of periods by utilizing a plurality of correlation characteristics; the judgment mode can be to integrate the change trends of the state characteristic data in different periods and judge whether the amplification trend exists or not.

Therefore, after the state characteristic data of the target battery in a plurality of periods are acquired, the state characteristic data in the plurality of periods can be processed by using a correlation analysis mode, and a correlation analysis processing result is obtained.

Wherein, the correlation analysis can find the correlation or relativity among data in a large amount of data sets; therefore, the relevance or correlation among the state characteristic data in a plurality of cycles can be found through the relevance analysis so as to obtain more accurate analysis results.

In one embodiment, the state feature data in multiple cycles are subjected to association analysis processing through a preset association rule to obtain an association analysis processing result, wherein the association rule can be an association relationship between the state feature data, for example, if an association exists between the a feature data and the B feature data, the B feature data can be predicted through the value of the a data feature, so that the state feature data in the multiple cycles can be analyzed according to the preset association rule to analyze whether the state feature data in the multiple cycles meet the preset association rule or not to obtain the association analysis processing result; the a-feature data includes at least one feature and the B-feature data includes at least one feature.

For example, if the state characteristic data includes the pressure difference and the internal resistance value, the pressure difference and the internal resistance value in multiple cycles may be subjected to correlation analysis processing to determine a correlation analysis processing result; specifically, the pressure difference and the internal resistance of the target battery have a correlation relationship, correspondingly, the pressure difference is in a range A, the internal resistance value is in a range B, correlation analysis processing is carried out on the pressure difference and the internal resistance value, when the pressure difference and the internal resistance value of the target battery are not in the range A and not in the range B, the target battery is determined to have an abnormal condition, a correlation analysis processing result of the target battery is determined according to the pressure difference and the internal resistance value of the target battery and the deviation value in the corresponding preset range, and the correlation analysis processing result can be a numerical value.

And S203, determining a fault prediction result of the target battery according to the correlation analysis processing result.

The failure prediction result of the target battery may be whether the target battery will be charged or discharged, and if the target battery has the charging or discharging phenomenon, the failure of the target battery may be caused.

The correlation analysis processing result may be a data characteristic value obtained by comprehensively processing the state characteristic data in multiple cycles, where the data characteristic value may be a value in a range from 0 to 1, and the severity of the fault is determined by the distribution position of the data characteristic value in the range from 0 to 1, where a closer data characteristic value to 0 indicates a smaller fault, and a closer data characteristic value to 1 indicates a more serious fault; for example, if the data characteristic value is 0, it may indicate that the target battery is not charged or discharged, and if the data characteristic value is 1, it may indicate that the target battery is faulty and the fault risk is serious.

Optionally, the data characteristic value may also be compared with a preset characteristic threshold, and a fault prediction result of the target battery is determined according to the comparison result; specifically, if the characteristic data value is greater than a preset characteristic threshold value, determining that the target battery has a fault; otherwise, the target battery is not malfunctioning.

In the battery fault prediction method, state characteristic data of a target battery in a plurality of periods are obtained, correlation analysis processing is carried out on the state characteristic data in the plurality of periods to obtain a correlation analysis processing result, and then a fault prediction result of the target battery is determined according to the correlation analysis processing result; wherein the state characteristic data in each cycle comprises a plurality of different battery characteristic data. In the embodiment, because the state characteristic data in multiple cycles of the battery are subjected to the correlation analysis processing, and the real-time state characteristic in each cycle has multiple different battery characteristic data, the multiple different battery characteristic data in the multiple cycles of the battery are equivalently combined for analysis, so that whether the battery has a fault or not is predicted from the battery characteristics in multiple dimensions, and the accuracy of the battery fault prediction result is improved.

In one embodiment, performing correlation analysis processing on the state feature data in multiple cycles to obtain a correlation analysis processing result includes: and inputting the state characteristic data in the multiple periods into the fault prediction model, and performing correlation analysis processing on the state characteristic data in the multiple periods through the fault prediction model to obtain a correlation analysis processing result.

The manner of performing the correlation analysis processing on the state characteristic data in the multiple periods may be a manner of using a neural network model, and specifically, the state characteristic data of the battery in the multiple periods is input into a fault prediction model, and the state characteristic data in the multiple periods is analyzed by the fault prediction model to obtain a correlation analysis processing result.

Optionally, the fault prediction model is pre-constructed through historical data, and the construction process of the fault prediction model may be to obtain historical state feature data of the sample battery in multiple cycles, and then train the initial fault prediction model through the historical state feature data until the initial fault prediction model meets a preset convergence condition, so as to obtain the fault prediction model.

It should be noted that, in the embodiment of the present application, a network architecture adopted in constructing the fault prediction model is not limited.

In the following, how to perform the association analysis processing on the state feature data in multiple cycles through the failure prediction model is described with an embodiment, in an embodiment, as shown in fig. 3, the performing the association analysis processing on the state feature data in multiple cycles through the failure prediction model to obtain an association analysis processing result includes the following steps:

s301, according to the state characteristic data in a plurality of periods, acquiring abnormal distribution information of the battery characteristics in each period, and acquiring abnormal change trend information of the battery characteristics in a plurality of periods.

Analyzing the state characteristic data in each period, and determining battery characteristic abnormal distribution information in each period, where the battery characteristic abnormal distribution information indicates an abnormal condition of the battery characteristic data of the target battery in the period, and may include abnormal battery characteristic data, a battery core identifier corresponding to the abnormal battery characteristic data, and the like.

Analyzing the state characteristic data of a plurality of periods to determine battery characteristic abnormal change trend information of the battery characteristic data between the periods, wherein the battery characteristic abnormal change trend information represents abnormal change conditions of the battery characteristic data between the periods, and the method can comprise the following steps: in a plurality of weeks, the change tendency of certain battery characteristic data is abnormal, and there is a sudden change or the like.

The manner of obtaining the battery characteristic abnormal distribution information in each cycle may be to analyze a plurality of battery characteristic data in the cycle, take the battery characteristic data including the charging frequency and the discharging frequency as an example, respectively judge whether the charging frequency and the discharging frequency of the target battery in the cycle meet the safety frequency, and determine the battery characteristic abnormal distribution condition of the charging frequency and the discharging frequency in the cycle.

The battery characteristic abnormal change trend information of the plurality of week periods may be obtained by, for example, taking the battery characteristic data of the target battery as including the charging frequency and the discharging frequency, respectively determining the abnormal change trends of the charging frequency and the discharging frequency of the target battery in the week periods, and determining whether abrupt changes exist in the charging frequency and the discharging frequency to determine the battery characteristic abnormal change trend information of the plurality of week periods.

And S302, performing correlation analysis processing on the battery characteristic abnormal distribution information in each period and the battery characteristic abnormal change trend information in a plurality of periods to obtain correlation analysis processing results.

And performing correlation analysis on the obtained battery characteristic abnormal distribution information in each period and the battery characteristic abnormal change trend information in a plurality of periods to obtain a correlation analysis processing result.

For example, taking the high-frequency as an example, if there is abnormal distribution information in the high-frequency, but the high-frequency has no abnormal change trend in a plurality of cycles, the abnormal distribution information in the high-frequency may be analyzed to determine the correlation analysis processing result in the high-frequency, and if the abnormal distribution information in the high-frequency is not obvious, the correlation analysis processing result in the high-frequency may be determined to be normal.

If the abnormal distribution information exists in the charging frequency, but the charging frequency does not have the abnormal change trend in a plurality of weeks, but the abnormal distribution information of the charging frequency is obvious, the correlation analysis processing result of the charging frequency can be determined to be abnormal.

If the battery characteristic data comprises a charging frequency and a discharging frequency, the correlation analysis processing result of the charging frequency and the correlation analysis processing result of the discharging frequency can be comprehensively analyzed to obtain the final correlation analysis processing result.

In one embodiment, the battery characteristic abnormal distribution information in each period of the target battery corresponds to one data distribution abnormal value, the battery characteristic abnormal change trend in a plurality of periods corresponds to one data change abnormal value, the data distribution abnormal value and the data change abnormal value are subjected to comprehensive analysis, and the correlation analysis result of the target battery is determined; for example, the sum of the data distribution abnormal value and the data change abnormal value is determined as the correlation analysis processing result, or the weighted value of the data distribution abnormal value and the data change abnormal value is determined as the correlation analysis processing result, or the like.

According to the battery fault prediction method, the battery characteristic abnormal distribution information in each period and the battery characteristic abnormal change trend information in a plurality of periods are obtained according to the state characteristic data in the plurality of periods, and the battery characteristic abnormal distribution information in each period and the battery characteristic abnormal change trend information in the plurality of periods are subjected to correlation analysis processing to obtain a correlation analysis processing result. According to the method, battery characteristic abnormal distribution information in each period of the target battery and battery characteristic abnormal change trend information in a plurality of periods are comprehensively analyzed, joint analysis is carried out on information in a plurality of dimensions, and accuracy of target battery fault prediction in the follow-up process is improved.

If the fault prediction model includes a classification model and a time sequence model, in an embodiment, as shown in fig. 4, the obtaining of the abnormal distribution information of the battery characteristics in each cycle and the abnormal variation trend information of the battery characteristics in a plurality of cycles according to the state characteristic data in the plurality of cycles includes the following steps:

and S401, inputting a plurality of different battery characteristic data in the state characteristic data in each period into a classification model to obtain battery characteristic abnormal distribution information in each period.

Classifying a plurality of different battery characteristic data in the state characteristic data in each period through a classification model, and determining battery characteristic abnormal distribution information in the plurality of battery characteristic data in each period, wherein the battery characteristic abnormal distribution information can comprise abnormal battery characteristic data in each period and corresponding battery core identification.

The method for determining the abnormal battery feature distribution information in each period through the classification model may be that a plurality of different battery feature data in the state feature data in a plurality of periods are simultaneously input into the classification model to obtain the abnormal battery feature distribution information in each period.

Alternatively, a plurality of different battery feature data in the state feature data in each cycle may be input to the classification model, and the battery feature abnormal distribution information in each cycle may be obtained.

The construction process of the classification model can be that historical state characteristic data of the sample battery in a plurality of periods are obtained, then the initial classification model is trained through the historical state characteristic data until the initial classification model meets a preset convergence condition, and the classification model is obtained.

S402, inputting a plurality of different battery characteristic data in the state characteristic data in each period into a time sequence model to obtain abnormal change trend information of the battery characteristics in a plurality of periods.

The time sequence model is a model for researching the self variation trend of corresponding data by taking time as an independent variable, and battery characteristic data in a plurality of continuous periods can be analyzed through a time sequence to obtain the time sequence model; analyzing the state characteristic data in a plurality of periods through a time sequence model, and determining battery characteristic abnormal change trend information in the plurality of periods, wherein the battery characteristic abnormal change trend information can comprise battery characteristic data with abnormal change trends in the plurality of periods and corresponding battery core identifications.

The method for determining the abnormal change trend information of the battery characteristics in the plurality of week periods through the time series model may be that a plurality of different battery characteristic data in the state characteristic data in the plurality of week periods are simultaneously input into the time series model, and the abnormal change trend information of the battery characteristics in the plurality of week periods is obtained through analysis of the time series model.

Alternatively, different battery characteristic data for a plurality of cycles may be input to the time-series model, and the battery characteristic data having an abnormal change tendency between the cycles may be obtained.

The time sequence model can be constructed by obtaining historical state characteristic data of a sample battery in a plurality of periods, and then training the initial time sequence model through the historical state characteristic data until the initial time sequence model meets a preset convergence condition, so as to obtain the time sequence model.

In the battery fault prediction method, a plurality of different battery characteristic data in the state characteristic data in each period are input into the classification model to obtain the battery characteristic abnormal distribution information in each period, and a plurality of different battery characteristic data in the state characteristic data in each period are input into the time sequence model to obtain the battery characteristic abnormal change trend information in a plurality of periods. According to the method, the state characteristic data in a plurality of periods are analyzed through the classification model and the time sequence model respectively, battery characteristic abnormal distribution information in each period and battery characteristic abnormal change trend in the period are obtained, battery characteristic data in two dimensions in the period and in the period are considered comprehensively, and accuracy of a target battery fault prediction result obtained subsequently is guaranteed.

If the failure prediction result includes that the target battery has a failure or does not have a failure, in an embodiment, as shown in fig. 5, the embodiment includes the following steps:

s501, under the condition that the target battery has faults, at least one abnormal battery cell with high charging and discharging risks in the target battery is obtained.

The prediction result comprises that the target battery has a fault or does not have a fault, if the fault prediction result is that the target battery has a fault, the corresponding fault prediction result also comprises a battery cell identifier with a charging height lowering phenomenon in the target battery, and the battery cell identifier corresponds to an abnormal battery cell; therefore, the battery cell identification with high charging and discharging risks in the target battery can be directly obtained from the fault prediction result; the abnormal battery cell comprises at least one.

Optionally, the manner of acquiring at least one abnormal cell with a high charge and low discharge risk in the target battery may also be that feature data of each cell in the target battery in multiple cycles is first acquired, and then the abnormal cell is determined according to the feature data.

For example, if the characteristic data includes a voltage difference, a voltage difference current slope, and an internal resistance, the voltage difference current slope, and the internal resistance of each cell may be compared with preset safety conditions, and for any cell, if any one of the characteristic data of the voltage difference, the voltage difference current slope, and the internal resistance in the cell does not satisfy the preset safety conditions, the cell is determined to be an abnormal cell.

The safety condition may be that a numerical value corresponding to the corresponding feature data is smaller than a preset safety threshold, or larger than the preset safety threshold, or within a preset safety range.

In the present application, the existence of a fault in the target battery indicates that the target battery has a high charge or low charge phenomenon, and the existence of an abnormal cell indicates that there is a high charge or low charge risk.

And S502, acquiring fault risk types of the abnormal battery cells.

The fault risk corresponding to the abnormal electric core is a fault risk, and the caused fault types comprise internal resistance abnormality and capacity abnormality.

Therefore, after determining the abnormal cell in the target battery, the fault type causing the abnormal risk of the cell can be further determined.

In an embodiment, as shown in fig. 6, the acquiring the fault risk types of the abnormal battery cells includes the following steps:

and S601, dividing the state characteristic data of the abnormal battery cell aiming at any abnormal battery cell, and acquiring the internal resistance characteristic data and the capacity characteristic data of the abnormal battery cell.

And S602, determining the fault risk type of the abnormal cell as the internal resistance abnormality under the condition that the abnormality exists in the internal resistance characteristic data of the abnormal cell.

And S603, determining the fault risk type of the abnormal electric core as capacity abnormity under the condition that the capacity characteristic data of the abnormal electric core is abnormal.

Before determining the fault risk type of each abnormal cell, firstly, acquiring state characteristic data of each abnormal cell, and then dividing the state characteristic data through the fault risk type to obtain internal resistance type characteristic data and capacity type characteristic data, wherein the internal resistance type characteristic data is the state characteristic data capable of representing the phenomena of high charging and low charging caused by the abnormal internal resistance of the cell, and whether the fault risk type causing the abnormal cell is the abnormal internal resistance can be judged through the internal resistance type characteristic data; the capacity characteristic data represents state characteristic data capable of representing the phenomena of high charging and low charging caused by the abnormal capacity of the battery cell, and whether the fault risk type causing the abnormal battery cell is the abnormal capacity can be judged through the capacity characteristic data.

In the phenomena of high and low charging, if the characteristics of abnormal cell-to-normal cell median ratio, abnormal cell differential pressure and current slope absolute value amplification (large differential pressure under large current), rapid rise of abnormal cell voltage in a discharge back-flushing section and the like exist, determining the fault risk type of the abnormal cell as internal resistance type abnormality; in the phenomena of charging and discharging, the pressure difference is not directly related to the current, if the low SOC of the abnormal cell is the lowest voltage cell, the high SOC is the highest voltage cell, the absolute value of the pressure difference and the SOC slope is amplified (the pressure difference is related to the SOC), and the like, the fault risk type of the abnormal cell is determined to be capacity type abnormality.

Based on the above description, the internal resistance class characteristic data may include: the maximum voltage cell frequency, the differential pressure current slope, the internal resistance value, the internal resistance ratio of the abnormal cell to other normal cells, the absolute value of the differential pressure and the current slope, the cell voltage in the discharging and recharging stage and the like in the recharging stage.

The maximum voltage cell frequency in the recharging stage represents the frequency at which the cell is at the maximum voltage in the recharging stage; the internal resistance ratio of the abnormal cell to other normal cells can be expressed as the ratio of the internal resistance of the abnormal cell to the median internal resistance of other normal cells, and can also be expressed as the ratio of the internal resistance of the abnormal cell to the average internal resistance of other normal cells; the absolute value of the voltage difference and the current slope represents the absolute value of the slope of a curve between the voltage difference value and the current value of the battery cell, and the cell voltage in the discharge recharge stage represents the voltage of each battery cell in the battery in the discharge recharge stage.

The capacity class characterization data may include: the minimum frequency under the low SOC working condition without considering the current, the maximum frequency under the high SOC working condition without considering the current, the voltage difference, the current, the minimum voltage under the low SOC working condition, the maximum voltage under the high SOC working condition, the absolute values of the voltage difference and the SOC slope and the like.

The minimum frequency under the low SOC condition without considering the current represents the minimum frequency under the low charge condition of the battery cell without considering the current, the maximum frequency under the high SOC condition without considering the current represents the maximum frequency under the high charge condition of the battery cell without considering the current, and the current can be the current of each battery cell in the battery; the lowest voltage under the low SOC working condition represents the minimum voltage of the battery cell under the low charge quantity; the maximum voltage under the high SOC working condition represents the maximum voltage of the battery cell under the high charge quantity; the absolute value of the differential pressure and the SOC slope represents the absolute value of the slope of the curve between the cell differential pressure and the SOC value.

For any abnormal battery cell, if the internal resistance characteristic data of the abnormal battery cell is abnormal, determining that the fault risk type of the abnormal battery cell is the internal resistance abnormality; and if the abnormal cell exists in the capacity characteristic data of the abnormal cell, determining that the fault risk type of the abnormal cell is capacity abnormality.

The internal resistance characteristic data of the abnormal cell may be abnormal, and the capacity characteristic data of the abnormal cell may be abnormal.

For example, the internal resistance characteristic data of the battery cell may be set to a standard range, that is, one internal resistance characteristic data corresponds to one standard range, and if any one of the internal resistance characteristic data of the abnormal battery cell is not in the corresponding standard range, it is determined that the characteristic data is abnormal, and the fault risk type of the corresponding abnormal battery cell is internal resistance abnormality.

Correspondingly, the capacity characteristic data of the battery cell may also be set to a standard range, that is, one kind of capacity characteristic data corresponds to one standard range, if any one of the capacity characteristic data of the abnormal battery cell is not in the corresponding standard range, it is determined that the characteristic data is abnormal, and the fault risk type of the corresponding abnormal battery cell is capacity abnormality.

It can be understood that, for any abnormal cell, if any one of the internal resistance characteristic data of the abnormal cell is abnormal and any one of the capacity characteristic data of the abnormal cell is abnormal, it is determined that the abnormal cell is abnormal in both internal resistance and capacity.

And S503, determining the fault risk degree of each abnormal electric core according to the fault risk type.

And determining the fault risk degree of each abnormal electric core based on the determined fault risk types, and determining the fault risk degree of each abnormal electric core according to the fault risk types of the abnormal electric cores and the state characteristic data corresponding to the abnormal electric cores.

If the fault risk degree of each abnormal cell is an internal resistance abnormality, further determining the fault risk degree of each abnormal cell according to the internal resistance characteristic data of each abnormal cell, for example, presetting the risk grade of each characteristic data, and determining the risk grade corresponding to each internal resistance characteristic data as the fault risk degree of the abnormal cell.

If the fault risk degree of each abnormal cell is capacity abnormality, further determining the fault risk degree of each abnormal cell according to the capacity characteristic data of each abnormal cell, for example, presetting the risk grade of each characteristic data, and determining the risk grade corresponding to each capacity characteristic data as the fault risk degree of the abnormal cell.

Optionally, if two abnormal characteristic data exist in the abnormal battery cell and correspond to two risk levels, the abnormal characteristic data and the corresponding two risk levels can be comprehensively judged to determine the fault risk degree of the abnormal battery cell; for example, the two risk levels are weighted to determine the final fault risk degree.

If the internal resistance abnormality and the capacity abnormality exist in the fault risk types of the abnormal cells, the risk level corresponding to the internal resistance abnormality and the risk level corresponding to the capacity abnormality can be determined based on the above manner, then the final risk level of each abnormal cell is determined according to the weight of the internal resistance abnormality and the weight of the capacity abnormality, and the risk level of each abnormal cell is determined as the fault risk degree of each abnormal cell.

Optionally, the fault risk levels may include a light charge high-low event, a medium charge high-low event, and a severe charge high-low event.

In the battery fault prediction method, under the condition that the target battery has a fault, at least one abnormal battery cell with high charging and discharging risks in the target battery is obtained, the fault risk type of each abnormal battery cell is obtained, and then the fault risk degree of each abnormal battery cell is determined according to the fault risk type. In the embodiment, the fault risk types of the abnormal electric cores in the target battery are obtained, and the fault risk degree of the abnormal electric cores is determined based on the fault risk types, so that the fault accuracy of the target battery is further improved, and the target battery is convenient to overhaul subsequently.

In the following, how to determine the fault risk degree of each abnormal battery cell is described in detail through an embodiment, in one embodiment, as shown in fig. 7, determining the fault risk degree of each abnormal battery cell according to the fault risk type includes the following steps:

and S701, for any abnormal electric core, determining target battery characteristic data from the state characteristic data of the abnormal electric core according to the fault risk type of the abnormal electric core.

Wherein the target battery characteristic data comprises at least one battery characteristic data; namely, at least one piece of battery characteristic data is selected from the state characteristic data of the abnormal electric core, and the selected at least one piece of battery characteristic data is determined as target battery characteristic data.

In one embodiment, determining target battery characteristic data from state characteristic data of an abnormal cell according to a fault risk type of the abnormal cell includes: determining target battery characteristic data of the abnormal electric core from the internal resistance characteristic data in the state characteristic data of the abnormal electric core under the condition that the fault risk type of the abnormal electric core is abnormal internal resistance; and under the condition that the fault risk type of the abnormal battery cell is abnormal in capacity, determining target battery characteristic data of the abnormal battery cell from capacity type characteristic data in the state characteristic data of the abnormal battery cell.

Specifically, if the fault type of the abnormal electric core is that the internal resistance is abnormal, the corresponding internal resistance characteristic data includes: maximum voltage cell frequency, differential pressure current slope, internal resistance value, internal resistance ratio of abnormal cell to other normal cells, absolute value of differential pressure and current slope, cell voltage in discharge recharge stage, etc. in recharge stage; determining at least one characteristic data in the internal resistance characteristic data as target battery characteristic data of the abnormal battery core; for example, the target battery characteristic data of the abnormal cell may be a voltage difference current slope, an internal resistance value, and an internal resistance ratio of the abnormal cell to other normal cells.

If the fault type of the abnormal battery core is capacity abnormality, the corresponding capacity characteristic data comprises the following data: the minimum frequency under the low SOC working condition without considering the current, the maximum frequency under the high SOC working condition without considering the current, the pressure difference, the current, the lowest voltage under the low SOC working condition, the highest voltage under the high SOC working condition, the absolute values of the pressure difference and the SOC slope and the like; determining at least one characteristic data in the capacity characteristic data as target battery characteristic data of the abnormal battery core; for example, the target battery characteristic data of the abnormal cell may be a minimum frequency in a low SOC condition without considering the current, and a maximum frequency in a high SOC condition without considering the current.

It should be noted that the target battery characteristic data can represent the most critical data characteristics of the state and abnormal condition of the abnormal cell.

And S702, determining the fault risk degree of the abnormal electric core according to the target battery characteristic data of the abnormal electric core.

Because the target battery characteristic data can represent the state and abnormal condition of the abnormal electric core, the fault risk degree of the abnormal electric core can be determined according to the target battery characteristic data of the abnormal electric core.

The mode of determining the fault risk degree of the abnormal electrical core may be that an abnormal characteristic value of the abnormal electrical core is determined according to target battery characteristic data of the abnormal electrical core, where the abnormal characteristic value may be a value in a range from 0 to 1, and the fault risk degree is determined according to a distribution position of the abnormal characteristic value in the range from 0 to 1, where the closer the abnormal characteristic value is to 0, the lower the fault risk degree is indicated, and the closer the abnormal characteristic value is to 1, the higher the fault risk degree is indicated; for example, if the abnormal characteristic value is 0.1, it indicates that the failure risk of the abnormal cell is low, and if the abnormal characteristic value is 0.8, it indicates that the failure risk of the abnormal cell is high.

The method for determining the abnormal characteristic value of the abnormal electric core according to the target battery characteristic data of the abnormal electric core may be to calculate a deviation value between the target battery characteristic data of the abnormal electric core and a safety threshold, determine the abnormal characteristic value of the abnormal electric core according to the deviation value, for example, one abnormal characteristic value corresponds to one abnormal deviation range, and determine the abnormal characteristic value corresponding to the deviation range where the deviation value is located as the abnormal characteristic value of the abnormal electric core.

The failure risk degree of the abnormal cell may be determined by comparing the target battery characteristic data of the abnormal cell with a preset risk level range, determining which risk level range the target battery characteristic data of the abnormal cell is in, and determining the failure risk degree corresponding to the risk level range as the failure risk degree of the abnormal cell; wherein, the risk grade range corresponds to the fault risk degree one by one; for example, the risk level range may be a numerical range of the battery characteristic data, and the failure risk level may be a level range including primary, secondary, tertiary, and so on.

The mode of determining the fault risk degree of the abnormal electrical core may also be that the target battery characteristic data of the abnormal electrical core is input into a preset fault risk model, and the target battery characteristic data of the abnormal electrical core is analyzed through the fault risk model to obtain the fault risk degree of the abnormal electrical core.

In the battery fault prediction method, for any abnormal electric core, target battery characteristic data is determined from state characteristic data of the abnormal electric core according to the fault risk type of the abnormal electric core, and then the fault risk degree of the abnormal electric core is determined according to the target battery characteristic data of the abnormal electric core; wherein the target battery characteristic data comprises at least one battery characteristic data. According to the method, the fault risk degree of the abnormal electric core is determined according to at least one battery characteristic data selected from the state characteristic data of the abnormal electric core, and the accuracy of determining the fault risk degree of the abnormal electric core is improved.

In an embodiment, as shown in fig. 8, the specific implementation of determining the fault risk degree of the abnormal electrical core is provided below, and the determining the fault risk degree of the abnormal electrical core according to the target battery characteristic data of the abnormal electrical core includes the following steps:

and S801, acquiring target battery characteristic data of other normal cells except the abnormal cells in the target battery.

When the fault risk degree of the abnormal electric core is determined, the target battery characteristic data of other normal electric cores except the abnormal electric core in the target battery needs to be acquired, and then the target battery characteristic data of the abnormal electric core is compared with the target battery characteristic data of the normal electric core, so that the fault risk degree of the abnormal electric core can be more reasonably determined.

Based on the abnormal electric core in the target battery, the normal electric core in the target battery can be determined, and then the target battery characteristic data of the normal electric core is directly obtained from the database.

The battery characteristic data of each battery cell in the battery can be stored in a database, and the target battery characteristic data can be directly acquired through a sensor, for example, the voltage of the battery cell; or after battery data is acquired by the sensor, the battery data is processed to obtain target battery characteristic data, such as cell voltage difference; the target battery characteristic data includes at least one battery characteristic data.

S802, transversely comparing the target battery characteristic data of the abnormal electric core with the target battery characteristic data of other normal electric cores to determine the fault risk degree of the abnormal electric core.

The transverse comparison may be comparison of the same data, transverse comparison of the target characteristic data of the abnormal electric core and the target battery characteristic data of other normal electric cores, or comparison of the same target battery characteristic data of the abnormal electric core and other normal electric cores in the same period.

In one embodiment, if the target battery characteristic data is a cell voltage difference, the cell voltage difference is a cell voltage difference within a preset period, and therefore, the cell voltage difference of the abnormal cell may be compared with an average value of the cell voltage differences of other normal cells to determine a fault risk degree of the abnormal cell; the difference value between the average value of the cell voltage difference of the abnormal cell and the average value of the cell voltage differences of other normal cells can be calculated, and the fault risk degree is determined according to the absolute value of the difference value.

For example, if the absolute value of the difference is greater than a first preset threshold, it is determined that the fault risk degree of the abnormal electrical core is level 1, and if the absolute value of the difference is greater than a second preset threshold and less than or equal to the first preset threshold, it is determined that the fault risk degree of the abnormal electrical core is level 2, and if the absolute value of the difference is greater than a third preset threshold and less than or equal to the second preset threshold, it is determined that the fault risk degree of the abnormal electrical core is level 3, where the first preset threshold is greater than the second preset threshold, and the second preset threshold is greater than the third preset threshold.

In the battery fault prediction method, the target battery characteristic data of other normal battery cores in the target battery except the abnormal battery core is obtained, and the target battery characteristic data of the abnormal battery core and the target battery characteristic data of other normal battery cores are transversely compared to determine the fault risk degree of the abnormal battery core. In the method, the accuracy of determining the fault risk degree of the abnormal battery cell is ensured by transversely comparing the abnormal data with the normal data.

For example, if the target battery characteristic data is an internal resistance value, in an embodiment, as shown in fig. 9, the transversely comparing the target battery characteristic data of the abnormal cell with the target battery characteristic data of other normal cells to determine the fault risk degree of the abnormal cell includes the following steps:

s901, acquiring the internal resistance value of the abnormal cell and acquiring the internal resistance values of the median of the abnormal cell and other normal cells.

If the target battery characteristic data is the internal resistance value, the internal resistance value of the abnormal battery core and the median internal resistance values of other normal battery cores can be transversely compared, and therefore, the internal resistance value of the abnormal battery core and the median internal resistance values of other normal battery cores need to be obtained before comparison.

The internal resistance of the battery cell does not have a direct measured value, so the internal resistance of the battery cell needs to be obtained through calculation, and one of the ways of calculating the internal resistance can be that the voltage and the current of the battery cell are obtained through a sensor, and the internal resistance value of the battery cell is determined through the ratio of the voltage and the current to obtain the internal resistance value of each battery cell in the target battery; the method for calculating the internal resistance value of the battery cell is not limited, the internal resistance value can be calculated through different logics, and the internal resistance value of the battery cell is determined according to the internal resistance values calculated through different logics; for example, the average value of the internal resistance values calculated in different logics is determined as the internal resistance value of the cell.

And acquiring the internal resistance values of the abnormal battery cell and the internal resistance values of other normal battery cells from the internal resistance values of the battery cells, and then determining the median internal resistance value according to the internal resistance values of the other normal battery cells.

It can be understood that when the internal resistance value of the abnormal cell and the internal resistance values of other normal cells are obtained, other influence factors of the abnormal cell and other normal cells are the same.

And S902, acquiring the ratio of the internal resistance value to the internal resistance value of the median.

And calculating the ratio of the internal resistance value of the abnormal electric core to the internal resistance value of the median of other normal electric cores based on the internal resistance value of the abnormal electric core and the internal resistance values of the median of other normal electric cores.

And S903, determining the fault risk degree of the abnormal electric core according to the ratio and a plurality of preset different fault risk degree grade ranges.

Based on the ratio of the internal resistance value of the abnormal cell to the median internal resistance value of the normal cell, a fault risk degree grade range is preset, for example, the ratio range is 1.5 to 2, the fault risk degree grade is 1 grade, if the ratio range is 2.1 to 2.5, the fault risk degree is 2 grade, and the ratio range is 2.6 to 3, the fault risk degree grade is 3 grade.

If the ratio of the internal resistance value of the abnormal electric core to the median internal resistance values of other normal electric cores in the target battery is 1.6, it can be determined that the fault risk degree of the abnormal electric core is level 1.

It should be noted that the above-mentioned ratio range is only used as an example for illustration, and in practical applications, the present application does not limit the range of the plurality of different fault idle degree levels.

Optionally, for a certain abnormal electrical core L, if the internal resistance value of the abnormal electrical core L at the time T1 is X1, the median internal resistance values of other normal electrical cores are X2, and X1/X2= Q, then comparing Q with a plurality of preset threshold values, and different threshold values correspond to different fault risk degrees, so that the fault risk degree of the abnormal electrical core L can be determined, for example, the predicted fault risk degree includes that the abnormal electrical core L may have a slight high-low charging phenomenon, the abnormal electrical core L may have a medium high-low charging phenomenon, the abnormal electrical core L may have a more severe high-low charging phenomenon, and the abnormal electrical core L may have a very severe high-low charging phenomenon.

In one embodiment, whether the connecting piece between the battery cores is abnormal or not can be judged through the ratio of the internal resistance value of the abnormal battery core to the median internal resistance values of other normal battery cores; in addition, the mode of calculating the electric core internal resistance value is not limited in the application, the internal resistance value can be calculated by using various logics, mutual verification is carried out, and if the values are abnormal in one period, the abnormal phenomenon is regarded as an abnormal phenomenon; the internal resistance anomaly includes a connection anomaly.

According to the battery fault prediction method, the internal resistance value of the abnormal electric core is obtained, the median internal resistance value of the abnormal electric core and other normal electric cores is obtained, the ratio of the internal resistance value to the median internal resistance value is obtained, and the fault risk degree of the abnormal electric core is determined according to the ratio and a plurality of preset different fault risk degree grade ranges. According to the method, the internal resistance value of the abnormal battery cell and the median internal resistance values of other normal battery cells are considered, the fault risk degree of the abnormal battery cell is determined according to the ratio and the preset different fault risk degree grade ranges, and the accuracy of the determined fault risk degree of the abnormal battery cell is guaranteed.

In an embodiment, as shown in fig. 10, fig. 10 is a method for predicting a battery fault, first obtaining historical data in a big data platform, where the historical data is historical fault sample data, and establishing a data characteristic time sequence change and state time sequence change relation by using the historical fault sample data to obtain a fault monitoring model; and then, utilizing the fault monitoring model to perform real-time monitoring on the real-time data of the big data platform to obtain a real-time monitoring result, specifically, inputting the real-time data into the fault monitoring model, and analyzing the real-time data through the fault monitoring model to obtain the real-time monitoring result.

How to construct the fault monitoring model through the historical data is explained in detail below, as shown in fig. 11, abnormal internal resistance may be caused by abnormal connection pieces between the battery cells, and according to an abnormal fault mechanism, the internal resistance between the battery cells in the battery is abnormal, so that the charging height and the charging height of the battery cells are reduced, and the cruising mileage is reduced.

According to the mechanism process, determining the characteristics of historical data, specifically, judging whether the phenomena of high charging and low charging exist or not through high charging frequency, low charging frequency and pressure difference, and if the phenomena of high charging and low charging exist, determining a high charging cell number and a low charging cell number, namely an abnormal cell.

Then, judging whether the abnormal electric core causing the phenomena of high charging and low discharging is abnormal in internal resistance or capacity according to the characteristics of the battery; the battery features may include: a low voltage frequency of high current low SOC discharge, a high voltage frequency of high current charging, a high voltage frequency of high current recharge, a low voltage frequency of low SOC discharge, a high voltage frequency of high SOC charging, a differential voltage current slope, a differential voltage SOC correlation, and the like.

And transversely comparing the abnormal cell, calculating the median internal resistance ratio of the abnormal cell to other normal cells, and/or the median internal resistance ratio of the voltage change rate of the abnormal cell to the voltage change rate of other normal cells, and the like, and judging whether the internal resistance is abnormal.

And finally, based on the analysis process, extracting the characteristics of the historical fault sample corresponding to the historical data in cycles, and establishing data characteristic time sequence change and state time sequence change relation to construct a fault monitoring model.

Correspondingly, when the fault monitoring model monitors real-time data in real time, the fault monitoring model can extract the characteristic data of the real-time data in cycles, analyze the abnormal distribution of the characteristic data in the cycles and the abnormal change trend among the cycles, and comprehensively judge whether fault abnormality exists according to the abnormal distribution of the characteristic data in the cycles and the abnormal change trend among the cycles to obtain a real-time monitoring result; the fault monitoring model can comprise a classification model and a time sequence model, the classification model is used for analyzing the characteristic abnormal distribution in a period, and the time sequence model is used for analyzing the characteristic abnormal change trend in the period.

In an embodiment, the maximum voltage cell frequency, the differential voltage current slope, the internal resistance, the ratio of the internal resistances of the abnormal cell to other normal cells, and the like in the recharge stage may be used to determine whether the internal resistance abnormality occurs. The internal resistance can be determined by the ratio of the current jump electric voltage to the current, and can also be determined by the slope of the voltage and the current in a time window.

And comprehensively judging whether the capacity is abnormal or not by taking the low SOC working condition without considering the current as the minimum cell frequency and taking the high SOC working condition as the maximum cell frequency.

In one embodiment, as shown in fig. 11, this embodiment includes the steps of:

s1201, state characteristic data of the target battery in multiple periods are obtained, and the state characteristic data in each period comprise multiple different battery characteristic data.

And S1202, analyzing the state characteristic data in a plurality of periods to determine a fault prediction result of the target battery.

The fault prediction result comprises whether the target battery has a fault or does not have a fault, and whether the target battery has a charging height lowering phenomenon.

In the first case, the method is implemented by distinguishing a plurality of different models, and S1202 includes:

firstly, acquiring abnormal distribution of battery characteristics in each period according to a plurality of different battery characteristic data in each period; and acquiring abnormal variation trends of the battery characteristics in a plurality of periods according to the state characteristic data in the plurality of periods.

The battery characteristic abnormal distribution in each period is obtained by analyzing a plurality of different battery characteristic data in each period through a classification model constructed in advance according to historical data. And analyzing the state characteristic data in a plurality of periods through a time sequence model which is constructed in advance according to historical data, and acquiring the abnormal variation trend of the battery characteristics in the plurality of periods.

Then, a failure prediction result of the target battery is determined based on the abnormal distribution of the battery characteristics in each cycle and the abnormal variation tendency of the battery characteristics in the plurality of cycles.

In the second case, an overall model is adopted to implement the method, and then S1202 includes:

and inputting the state characteristic data in a plurality of periods into a fault prediction model which is constructed in advance according to historical data, and analyzing the abnormal distribution of the battery characteristics in the target battery period and the abnormal change trend of the battery characteristics in the period through the fault prediction model to obtain a fault prediction result of the target battery.

And S1203, determining a high-low battery cell from the target battery as an abnormal battery cell.

And S1204, distinguishing whether each abnormal electric core is abnormal in internal resistance or capacity aiming at each abnormal electric core.

Specifically, for any abnormal cell, multiple internal resistance characteristic data of the abnormal cell are extracted, and if an abnormal characteristic exists in the internal resistance characteristic data, the cell is determined to be abnormal in internal resistance.

And extracting a plurality of capacity type characteristic data of the abnormal battery cell aiming at any abnormal battery cell, and if the abnormal characteristics exist in the capacity type characteristic data (as long as one characteristic is abnormal), determining that the battery cell is abnormal in capacity.

And S1205, transversely comparing the abnormal electric core with other normal electric cores in the target battery to predict the fault risk degree of each abnormal electric core.

If the internal resistance is abnormal, selecting one or more characteristics from the internal resistance characteristic data to predict the fault risk degree of the abnormal battery cell; and if the capacity is abnormal, selecting one or more characteristics from the capacity characteristic data to predict the fault risk degree of the abnormal battery core.

For specific limitations of the battery failure prediction method provided in this embodiment, reference may be made to the above step limitations of each embodiment in the battery failure prediction method, which is not described herein again.

It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the application also provides a battery fault prediction device for realizing the battery fault prediction method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so the specific limitations in one or more embodiments of the battery failure prediction device provided below can be referred to the limitations on the battery failure prediction method in the foregoing, and details are not repeated here.

In one embodiment, as shown in fig. 13, there is provided a battery failure prediction apparatus including: an obtaining module 1301, an analyzing module 1302 and a determining module 1303, wherein:

an obtaining module 1301, configured to obtain state feature data of a target battery in multiple cycles; the state characteristic data in each period comprise a plurality of different battery characteristic data;

the analysis module 1302 is configured to perform correlation analysis processing on the state feature data in multiple cycles to obtain correlation analysis processing results;

and the determining module 1303 is configured to determine a failure prediction result of the target battery according to the correlation analysis processing result.

In one embodiment, the analysis module 1302 includes:

and the processing unit is used for inputting the state characteristic data in the multiple periods into the fault prediction model, and performing correlation analysis processing on the state characteristic data in the multiple periods through the fault prediction model to obtain a correlation analysis processing result.

In one embodiment, the processing unit comprises:

the first acquisition subunit is used for acquiring abnormal battery characteristic distribution information in each period according to the state characteristic data in the plurality of periods and acquiring abnormal battery characteristic change trend information in the plurality of periods;

and the processing subunit is used for performing correlation analysis processing on the battery characteristic abnormal distribution information in each period and the battery characteristic abnormal change trend information in a plurality of periods to obtain a correlation analysis processing result.

In one embodiment, the first acquisition subunit includes:

the first input subunit is used for inputting a plurality of different battery characteristic data in the state characteristic data in each period into the classification model to obtain battery characteristic abnormal distribution information in each period;

and the second input subunit is used for inputting a plurality of different battery characteristic data in the state characteristic data in each period into the time sequence model to obtain the abnormal variation trend information of the battery characteristics in a plurality of periods.

In one embodiment, the apparatus 1300 further comprises:

the abnormal cell determining module is used for acquiring at least one abnormal cell with a high charging and discharging risk in the target battery under the condition that the target battery has a fault;

the fault type determining module is used for acquiring fault risk types of the abnormal battery cells;

and the risk degree determining module is used for determining the fault risk degree of each abnormal electric core according to the fault risk type.

In one embodiment, the fault type determination module includes:

the dividing unit is used for dividing the state characteristic data of the abnormal electric core aiming at any abnormal electric core to obtain the internal resistance characteristic data and the capacity characteristic data of the abnormal electric core;

the first determining unit is used for determining that the fault risk type of the abnormal battery cell is internal resistance abnormality under the condition that abnormality exists in the internal resistance type characteristic data of the abnormal battery cell;

and the second determining unit is used for determining that the fault risk type of the abnormal cell is capacity abnormity under the condition that the capacity characteristic data of the abnormal cell is abnormal.

In one embodiment, the risk level determination module comprises:

the third determining unit is used for determining target battery characteristic data from the state characteristic data of the abnormal electric core according to the fault risk type of the abnormal electric core aiming at any abnormal electric core, wherein the target battery characteristic data comprises at least one battery characteristic data;

and the fourth determining unit is used for determining the fault risk degree of the abnormal electric core according to the target battery characteristic data of the abnormal electric core.

In one embodiment, the third determination unit includes:

the first determining subunit is used for determining target battery characteristic data of the abnormal electric core from the internal resistance characteristic data in the state characteristic data of the abnormal electric core under the condition that the fault risk type of the abnormal electric core is the internal resistance abnormality;

and the second determining subunit is used for determining target battery characteristic data of the abnormal battery cell from the capacity characteristic data in the state characteristic data of the abnormal battery cell under the condition that the fault risk type of the abnormal battery cell is abnormal capacity.

In one embodiment, the fourth determination unit includes:

the second acquiring subunit is used for acquiring target battery characteristic data of other normal battery cores except the abnormal battery core in the target battery;

and the third determining subunit is used for transversely comparing the target battery characteristic data of the abnormal electric core with the target battery characteristic data of other normal electric cores so as to determine the fault risk degree of the abnormal electric core.

In one embodiment, the third determining subunit includes:

the third acquiring subunit is used for acquiring the internal resistance value of the abnormal electric core and acquiring the median internal resistance values of the abnormal electric core and other normal electric cores;

the fourth obtaining subunit is used for obtaining the ratio of the internal resistance value to the internal resistance value of the median;

and the fourth determining subunit is used for determining the fault risk degree of the abnormal electric core according to the ratio and a preset plurality of different fault risk degree grade ranges.

The respective modules in the above battery failure prediction apparatus may be wholly or partially implemented by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 14. The computer device includes a processor, a memory, an Input/Output interface (I/O for short), and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store battery failure prediction data. The input/output interface of the computer device is used for exchanging data between the processor and an external device. The communication interface of the computer device is used for connecting and communicating with an external terminal through a network. The computer program is executed by a processor to implement a battery failure prediction method.

It will be appreciated by those skilled in the art that the configuration shown in fig. 14 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program.

The implementation principle and technical effect of each step implemented by the processor in this embodiment are similar to the principle of the battery failure prediction method described above, and are not described herein again.

In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.

In the embodiment, the implementation principle and the technical effect of each step implemented when the computer program is executed by the processor are similar to the principle of the battery failure prediction method described above, and are not described herein again.

In an embodiment, a computer program product is provided, comprising a computer program which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.

In the embodiment, the implementation principle and the technical effect of each step implemented when the computer program is executed by the processor are similar to the principle of the battery failure prediction method described above, and are not described herein again.

It should be noted that the data (including but not limited to data for analysis, stored data, displayed data, etc.) referred to in the present application are data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with relevant laws and regulations and standards of relevant countries and regions.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include a Read-Only Memory (ROM), a magnetic tape, a floppy disk, a flash Memory, an optical Memory, a high-density embedded nonvolatile Memory, a resistive Random Access Memory (ReRAM), a Magnetic Random Access Memory (MRAM), a Ferroelectric Random Access Memory (FRAM), a Phase Change Memory (PCM), a graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the various embodiments provided herein may be, without limitation, general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, or the like.

All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims (14)

1. A method of battery failure prediction, the method comprising:

acquiring state characteristic data of a target battery in a plurality of periods; the state characteristic data in each period comprise a plurality of different battery characteristic data;

performing correlation analysis processing on the state characteristic data in the multiple periods to obtain correlation analysis processing results;

and determining a fault prediction result of the target battery according to the correlation analysis processing result.

2. The method according to claim 1, wherein the performing correlation analysis processing on the state feature data in the plurality of cycles to obtain a correlation analysis processing result comprises:

and inputting the state characteristic data in the multiple periods into a fault prediction model, and performing correlation analysis processing on the state characteristic data in the multiple periods through the fault prediction model to obtain a correlation analysis processing result.

3. The method according to claim 2, wherein the performing, by the fault prediction model, a correlation analysis process on the state feature data in the plurality of cycles to obtain the correlation analysis process result comprises:

according to the state characteristic data in the multiple periods, acquiring battery characteristic abnormal distribution information in each period and acquiring battery characteristic abnormal change trend information in the multiple periods;

and performing correlation analysis processing on the abnormal distribution information of the battery characteristics in each period and the abnormal change trend information of the battery characteristics in the plurality of periods to obtain a correlation analysis processing result.

4. The method according to claim 3, wherein the fault prediction model comprises a classification model and a time sequence model, and the obtaining of the abnormal distribution information of the battery characteristics in each period and the abnormal variation trend information of the battery characteristics in the plurality of periods according to the state characteristic data in the plurality of periods comprises:

inputting a plurality of different battery characteristic data in the state characteristic data in each period into the classification model to obtain battery characteristic abnormal distribution information in each period;

and inputting a plurality of different battery characteristic data in the state characteristic data in each period into the time sequence model to obtain the abnormal variation trend information of the battery characteristics in the plurality of periods.

5. The method of any of claims 1 to 4, wherein the fault prediction result comprises a presence or absence of a fault in the target battery, the method further comprising:

under the condition that the target battery has a fault, acquiring at least one abnormal battery cell with high charge and discharge risks in the target battery;

acquiring the fault risk types of the abnormal battery cells;

and determining the fault risk degree of each abnormal battery cell according to the fault risk type.

6. The method of claim 5, wherein the fault risk types include internal resistance abnormality or capacity abnormality, and the obtaining the fault risk types of the abnormal battery cells includes:

for any abnormal battery cell, dividing the state characteristic data of the abnormal battery cell to obtain internal resistance characteristic data and capacity characteristic data of the abnormal battery cell;

determining the fault risk type of the abnormal battery cell as internal resistance abnormality under the condition that the internal resistance characteristic data of the abnormal battery cell is abnormal;

and under the condition that the abnormal capacity characteristic data of the abnormal electric core exists, determining that the fault risk type of the abnormal electric core is capacity abnormality.

7. The method of claim 5, wherein the determining the fault risk degree of each abnormal electrical core according to the fault risk type includes:

for any abnormal electric core, determining target battery characteristic data from state characteristic data of the abnormal electric core according to the fault risk type of the abnormal electric core, wherein the target battery characteristic data comprises at least one battery characteristic data;

and determining the fault risk degree of the abnormal battery cell according to the target battery characteristic data of the abnormal battery cell.

8. The method of claim 7, wherein the determining target battery characteristic data from the state characteristic data of the abnormal cell according to the fault risk type of the abnormal cell comprises:

determining target battery characteristic data of the abnormal electric core from the internal resistance characteristic data in the state characteristic data of the abnormal electric core under the condition that the fault risk type of the abnormal electric core is abnormal internal resistance;

and under the condition that the fault risk type of the abnormal electric core is abnormal in capacity, determining target battery characteristic data of the abnormal electric core from capacity characteristic data in the state characteristic data of the abnormal electric core.

9. The method of claim 7, wherein the determining the fault risk level of the abnormal cell according to the target battery characteristic data of the abnormal cell comprises:

acquiring target battery characteristic data of other normal cells except for the abnormal cells in the target battery;

and transversely comparing the target battery characteristic data of the abnormal electric core with the target battery characteristic data of the other normal electric cores to determine the fault risk degree of the abnormal electric core.

10. The method of claim 9, wherein if the target battery characteristic data is an internal resistance value, the transversely comparing the target battery characteristic data of the abnormal cell with the target battery characteristic data of the other normal cells to determine the fault risk level of the abnormal cell comprises:

acquiring the internal resistance value of the abnormal electric core and acquiring the median internal resistance values of the abnormal electric core and the other normal electric cores;

obtaining the ratio of the internal resistance value to the internal resistance value of the median;

and determining the fault risk degree of the abnormal electric core according to the ratio and a plurality of preset different fault risk degree grade ranges.

11. A battery failure prediction apparatus, the apparatus comprising:

the acquisition module is used for acquiring state characteristic data of the target battery in a plurality of periods; the state characteristic data in each period comprise a plurality of different battery characteristic data;

the analysis module is used for carrying out correlation analysis processing on the state characteristic data in the multiple periods to obtain correlation analysis processing results;

and the determining module is used for determining the fault prediction result of the target battery according to the correlation analysis processing result.

12. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor realizes the steps of the method of any one of claims 1 to 10 when executing the computer program.

13. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 10.

14. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 10 when executed by a processor.

Images