CN116482557A

CN116482557A - Battery management method and device based on internal resistance of battery, electronic equipment and medium

Info

Publication number: CN116482557A
Application number: CN202310286376.2A
Authority: CN
Inventors: 潘兵; 王加龙
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2023-03-20
Filing date: 2023-03-20
Publication date: 2023-07-25

Abstract

The embodiment of the specification provides a battery management method based on internal resistance of a battery, which comprises the following steps: acquiring internal resistance data corresponding to at least one battery in the battery pack; respectively determining dynamic internal resistance thresholds corresponding to the batteries in the at least one battery according to the internal resistance data; the dynamic internal resistance threshold is an internal resistance threshold which is dynamically determined along with the change of the internal resistance data corresponding to each battery; based on the historical operation data of each battery in a specified time window, respectively predicting a predicted internal resistance value corresponding to each battery, and comparing the predicted internal resistance value with the dynamic internal resistance threshold; and outputting a fault alarm corresponding to the target battery in response to the fact that the numerical relation between the predicted internal resistance value corresponding to any one of the batteries and the dynamic internal resistance threshold value corresponding to the target battery meets a preset condition. In the process, faults can be found in advance, the judging accuracy is improved, the implementation is easy, and the operation and maintenance cost is reduced.

Description

Battery management method and device based on internal resistance of battery, electronic equipment and medium

Technical Field

The present disclosure relates to the field of battery technologies, and in particular, to a battery management method, device, electronic apparatus, and medium based on internal resistance of a battery.

Background

With the development of battery technology, more and more batteries are widely applied, for example, for enterprises, millions of storage batteries can be configured in the data center to serve as standby power sources, so as to cope with sudden states such as mains supply interruption and the like, and economic losses caused by the power failure of the data center are prevented.

In general, in order to prevent a battery from malfunctioning, it is necessary to perform daily inspection of these storage batteries by operation and maintenance personnel, check whether the battery has problems such as swelling or leakage, and replace the malfunctioning battery in time. In addition, the operation and maintenance personnel also need to periodically perform discharge tests on the battery to detect the performance of the battery, so that the battery with aging or abnormality is eliminated.

However, on the one hand, most faults need to rely on periodic discharge tests and only the battery that has failed can be detected, and cannot be predicted in advance; on the other hand, due to the excessive number of batteries, the operation and maintenance work needs to consume a large amount of manpower and material resources, and in consideration of the operation and maintenance cost, the discharge test is usually performed once a half year, which results in a relatively lagging operation and maintenance work.

Disclosure of Invention

In view of this, one or more embodiments of the present disclosure provide a battery management method, apparatus, electronic device, and medium based on internal resistance of a battery, to solve the problems in the related art.

In order to achieve the above object, one or more embodiments of the present disclosure provide the following technical solutions:

according to a first aspect of embodiments of the present specification, there is provided a battery management method based on internal resistance of a battery, including:

acquiring internal resistance data corresponding to at least one battery in the battery pack;

respectively determining dynamic internal resistance thresholds corresponding to each battery in the at least one battery according to the internal resistance data; the dynamic internal resistance threshold is an internal resistance threshold which is dynamically determined along with the change of the internal resistance data corresponding to each battery;

based on the historical operation data of each battery in a specified time window, respectively predicting a predicted internal resistance value corresponding to each battery, and comparing the predicted internal resistance value with the dynamic internal resistance threshold;

and outputting a fault alarm corresponding to the target battery in response to the fact that the numerical relation between the predicted internal resistance value corresponding to any one of the batteries and the dynamic internal resistance threshold value corresponding to the target battery meets a preset condition.

According to a second aspect of embodiments of the present specification, there is provided a battery management device based on internal resistance of a battery, comprising:

the acquisition module is used for acquiring internal resistance data corresponding to at least one battery in the battery pack;

a determining module for determining dynamic internal resistance thresholds corresponding to each battery in the at least one battery according to the internal resistance data; the dynamic internal resistance threshold is an internal resistance threshold which is dynamically determined along with the change of the internal resistance data corresponding to each battery;

the prediction module is used for respectively predicting the predicted internal resistance value corresponding to each battery based on the historical operation data of each battery in a specified time window and comparing the predicted internal resistance value with the dynamic internal resistance threshold;

and the alarm module is used for responding to the fact that the numerical relation between the predicted internal resistance value corresponding to any one target battery in the batteries and the dynamic internal resistance threshold value corresponding to the target battery meets the preset condition, and outputting fault alarm corresponding to the target battery.

According to a third aspect of embodiments of the present specification, there is provided an electronic device comprising a communication interface, a processor, a memory and a bus, the communication interface, the processor and the memory being interconnected by the bus;

The memory stores machine readable instructions and the processor performs the method by invoking the machine readable instructions.

According to a fourth aspect of embodiments of the present description, there is provided a machine-readable storage medium storing machine-readable instructions which, when invoked and executed by a processor, implement the above-described method.

The technical scheme provided by the embodiment of the specification can comprise the following beneficial effects:

according to the technical scheme, the dynamic internal resistance threshold value corresponding to each battery is determined based on the internal resistance data of the battery, the internal resistance of the battery is predicted, and the predicted result is compared with the respective dynamic internal resistance threshold value to determine whether an alarm is required. In the process, on one hand, the fault battery can be found in advance through the prediction of the internal resistance, so that the operation and maintenance work is more timely; on the other hand, by respectively determining the respective dynamic internal resistance threshold values according to the internal resistance data based on the dynamic change of each battery, the threshold value is more reasonable, the difference between batteries is considered, and the accuracy of fault judgment is improved. Meanwhile, the scheme is easy to implement, and the operation and maintenance cost is greatly reduced.

Drawings

Fig. 1 is a schematic diagram of initial values of internal resistances of different types of batteries according to an exemplary embodiment of the present disclosure;

fig. 2 is a flowchart of a battery management method based on internal resistance of a battery according to an exemplary embodiment of the present disclosure;

fig. 3 is a flowchart of still another battery management method based on internal resistance of a battery according to an exemplary embodiment of the present disclosure;

fig. 4 is a schematic structural diagram of an electronic device in which a battery management device based on internal resistance of a battery according to an exemplary embodiment of the present disclosure is located;

fig. 5 is a block diagram of a battery management device based on internal resistance of a battery according to an exemplary embodiment of the present disclosure.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of the present specification. Rather, they are merely examples of apparatus and methods consistent with aspects of one or more embodiments of the present description as detailed in the accompanying claims.

It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than described in this specification. Furthermore, individual steps described in this specification, in other embodiments, may be described as being split into multiple steps; while various steps described in this specification may be combined into a single step in other embodiments.

In the related art, the internal resistance of the battery can represent the aging degree of the battery, and the internal resistance value of the battery is different under different service life states, so that the capacity and the service life of the battery are influenced by the internal resistance of the battery.

Therefore, when detecting, operation and maintenance personnel can judge faults based on the internal resistance of the battery, and judge the faults through an expert threshold value given by battery manufacturers and battery specialists, and if the internal resistance is higher than the expert threshold value, the battery can be considered to be faulty and needs to be replaced.

However, the expert threshold is a fixed value, and cannot show the difference between the internal resistances of the batteries in practical application.

Taking fig. 1 as an example, fig. 1 is a schematic diagram of initial values of internal resistances of different types of batteries according to an exemplary embodiment of the present disclosure. As shown in fig. 1, on the one hand, there is a small difference in internal resistance even for the same type of battery, and on the other hand, the difference in internal resistance may be large between different types of batteries.

It can be understood that when the number of batteries is large and the number of models is large, due to different internal resistance characteristics of the batteries, a reasonable threshold value is difficult to set, if the threshold value is set too high, battery faults are difficult to find, and if the threshold value is set low, false alarms are easy to generate. In addition, the method cannot predict the battery failure in advance.

In view of this, the present disclosure provides a technical solution for determining whether an alarm is required by determining a dynamic internal resistance threshold corresponding to each battery based on internal resistance data of the battery, and comparing a prediction result with each dynamic internal resistance threshold by predicting internal resistances of the batteries.

When implemented, internal resistance data for each cell may be obtained for at least one cell in the battery.

In one example, the battery management platform may monitor the operation state of the battery and periodically sample the operation state to obtain historical operation internal resistance data of at least one battery, where the historical operation data may at least include internal resistance data corresponding to the battery, so as to obtain a historical change condition of the internal resistance of the battery.

Then, dynamic internal resistance thresholds corresponding to respective ones of the at least one battery may be determined based on the above internal resistance data, respectively.

In one example, the battery management platform may determine a reasonable dynamic internal resistance threshold for each of the at least one battery according to the historical change condition of the internal resistance of the battery, where the determined internal resistance threshold may be different due to the possible difference of the internal resistance data of the battery with the same model, and thus is referred to as a dynamic internal resistance threshold.

Further, based on historical operation data of each battery in a specified time window, predicted internal resistance values corresponding to each battery can be predicted respectively, and the predicted internal resistance values are compared with a dynamic internal resistance threshold.

In one example, historical operating data of each battery over a specified time window may be input into a pre-trained machine learning model to respectively predict predicted internal resistance values corresponding to each battery and compare the predicted internal resistance values to a dynamic internal resistance threshold.

Then, a fault alarm corresponding to the target battery can be output in response to the numerical relation between the predicted internal resistance value corresponding to any one of the batteries and the dynamic internal resistance threshold value corresponding to the target battery meeting a preset condition.

In one example, the preset condition may be a multiple relationship, and by comparing the predicted internal resistance value with the dynamic internal resistance threshold, a fault alarm corresponding to the target battery may be output when the predicted internal resistance value of any one of the target batteries reaches a specified multiple of the dynamic internal resistance threshold corresponding to the target battery.

The battery management method based on the internal resistance of the battery of the present specification will be described in detail with reference to the accompanying drawings.

Referring to fig. 2, fig. 2 is a flowchart of a battery management method based on internal resistance of a battery according to an exemplary embodiment of the present disclosure. As shown in fig. 2, the method comprises the following steps:

step 201, obtaining internal resistance data corresponding to at least one battery in a battery pack;

step 202, respectively determining dynamic internal resistance thresholds corresponding to each battery in the at least one battery according to the internal resistance data; the dynamic internal resistance threshold is an internal resistance threshold which is dynamically determined along with the change of the internal resistance data corresponding to each battery;

step 203, respectively predicting predicted internal resistance values corresponding to the batteries based on the historical operation data of the batteries in a specified time window, and comparing the predicted internal resistance values with the dynamic internal resistance threshold;

and 204, outputting a fault alarm corresponding to the target battery in response to the fact that the numerical relation between the predicted internal resistance value corresponding to any one of the batteries and the dynamic internal resistance threshold value corresponding to the target battery meets a preset condition.

When a large-scale battery is deployed, the battery is generally divided into a plurality of battery packs, each of which includes a plurality of batteries. Of course, the battery pack may include the entire number of batteries disposed, which is not limited in this specification.

The battery management method may also be referred to as a predictive maintenance method for a battery. Predictive maintenance refers to judging the change trend of the state of equipment in the future according to the current state of the equipment and making a maintenance plan when the equipment is in operation.

In this embodiment, internal resistance data corresponding to at least one battery in the battery pack may be acquired.

As can be seen from the foregoing, as the service life of the battery increases, the internal resistance of the battery changes, and the internal resistance affects the service state of the battery. In addition, there is a difference between the respective batteries, resulting in a difference in internal resistance data of the respective batteries. Therefore, the internal resistance data of each battery needs to be monitored.

In one embodiment shown, the above method may be applied to a battery management platform for monitoring the operating state of the battery;

further, when the internal resistance data corresponding to at least one battery in the battery pack is obtained, historical operation data of at least one battery in the battery pack can be obtained, wherein the historical operation data at least comprises the internal resistance data corresponding to at least one battery in the battery pack, which is obtained by periodic sampling by the battery management platform.

For example, for at least one battery in the battery pack, the battery management platform may obtain historical operation data of the battery by monitoring an operation state of the battery and sampling the operation state through a period specified by an operation and maintenance person.

The value of the internal resistance of the battery can represent the aging degree of the battery, and can reflect the health state of the battery, for example, the internal resistance of the battery is different when the battery is in different service life states. Therefore, at least the internal resistance data is included in the above-described historical operation data.

In one embodiment shown, the battery management platform includes a predictive maintenance PM platform, or a monitoring platform corresponding to a data center.

For example, the PM platform for predictive maintenance or the monitoring platform corresponding to the data center may perform the above battery management method on the battery to perform predictive maintenance, collect state data during battery operation, determine a trend of change in the state of the battery in the future according to the current state of the battery, and make a maintenance plan.

After the historical data of the battery is obtained, the abnormal sampling points in the historical data can be filtered, and the abnormal data can be removed.

In one embodiment, the internal resistance average value of the battery corresponding to the model may be determined according to the model to which each battery in at least one battery in the battery pack belongs, and the abnormality filtering may be performed on the internal resistance data of which the internal resistance value is higher than a specified multiple of the internal resistance average value in the internal resistance data and the internal resistance data of which the internal resistance value is lower than a specified percentage of the internal resistance average value in the internal resistance data based on the internal resistance average value;

Determining whether internal resistance data with the increase amplitude of the internal resistance value reaching a threshold value compared with the internal resistance value corresponding to one or more previous sampling periods exists in the internal resistance data; and if so, carrying out anomaly filtering on the internal resistance data.

For example, assuming that there are multiple types of batteries in the battery pack, for all the batteries of type a, the internal resistance average value of all the batteries of type a may be calculated from the internal resistance data in the historical operation data of all the batteries of type a.

Assuming that the internal resistance average value of all the batteries of the model A is Q, filtering out internal resistance data with internal resistance values higher than the appointed multiple of the internal resistance average value in the internal resistance data when abnormal filtering is carried out based on the internal resistance average value, for example, filtering out internal resistance data with internal resistance values exceeding 5 times of the average value Q in all the batteries of the model A; the internal resistance data with the internal resistance value lower than the specified percentage of the internal resistance mean value can be filtered, for example, the internal resistance data with the internal resistance value lower than 10% of the mean value Q is filtered out from all batteries of the same type A.

Also, in addition to the above-described internal resistance average value, abnormality filtering may be performed based on the relative degree of change in the internal resistance data.

Specifically, the anomaly filtering may be performed by determining whether there is internal resistance data in which an increase amplitude of the internal resistance value reaches a threshold value in the internal resistance data as compared with the internal resistance value corresponding to the previous one or more sampling periods.

For example, for the data of multiple sampling periods, the internal resistance data of the current sampling period may be determined, and compared with the increasing amplitude of the internal resistance value corresponding to one or more previous sampling periods, assuming that the threshold value is 30%, when the increasing amplitude reaches 40%, the internal resistance data of the current sampling period may be determined to be abnormal data, and filtering may be performed.

In this embodiment, according to the above internal resistance data, the dynamic internal resistance threshold value corresponding to each of the at least one battery may be determined, respectively.

For example, the battery management platform may determine a reasonable dynamic internal resistance threshold for each of the at least one battery based on historical changes in internal resistance of the battery.

It should be noted that, since the threshold value determined between different batteries of the same model may be different, and the threshold value determined according to the difference of the historical operation data may also be different for the same battery, the internal resistance threshold value is dynamically changed, and thus may be referred to as a dynamic internal resistance threshold value.

When the dynamic internal resistance threshold is determined, the internal resistance change of the battery and the use duration have a larger relationship, the longer the use time is, the lower the health degree is, the larger the generated internal resistance change is, and the shorter the use time is, the healthier the battery is, and the smaller the internal resistance change is, so that the determination can be separately performed based on the use duration of the battery.

In one embodiment, for each battery in at least one battery in the battery pack, in response to the running duration of the battery reaching a first duration, the internal resistance data obtained by periodically sampling the battery in the first duration is ordered, and the internal resistance value reaching a preset quantile in the internal resistance data is used as a dynamic internal resistance threshold corresponding to the battery according to the ordering result;

for each battery in at least one battery in the battery pack, responding to the fact that the running time of the battery is smaller than the first time, calculating an average value of the internal resistances of the batteries based on internal resistance data obtained by periodically sampling the batteries in the second time, and taking the average value as a dynamic internal resistance threshold corresponding to the batteries; wherein the second duration is less than the first duration.

For example, for each battery in at least one battery in the battery pack, the service duration of the battery, that is, the running duration, may be determined, and if the running duration of the battery reaches 2 years, then if the running duration of the battery reaches 2 years, the historical running data obtained by periodically sampling the battery within 2 years may be obtained, and the internal resistance data in the historical running data may be ranked. Then, according to the sorting result, the internal resistance value reaching the preset quantile in the internal resistance data is used as the dynamic internal resistance threshold value corresponding to the battery.

For example, if the quantile is 10%, the internal resistance data can be ranked from small to large according to the size of the internal resistance data, the internal resistance data in the first 10% of the ranking result is selected, and the maximum internal resistance value in the first 10% of the internal resistance data is used as the dynamic internal resistance threshold corresponding to the battery.

Continuing with the example, assuming the second duration is 1 month, if the battery is operated for less than 2 years, historical operating data from periodic sampling of the battery for one month may be obtained. Because short-term data is different from long-term data and has no large change, the short-term data can be directly averaged without sorting, and the calculated average value is used as a dynamic internal resistance threshold value corresponding to the battery.

It should be noted that, when determining the dynamic internal resistance threshold value corresponding to each battery, a certain correction may also be performed.

In one embodiment, the sorting may be performed based on the dynamic internal resistance threshold corresponding to each battery in at least one battery in the battery pack, and the dynamic internal resistance threshold reaching a preset quantile is used as the internal resistance threshold corresponding to the battery pack according to the sorting result;

and correcting the battery with the dynamic internal resistance threshold lower than the internal resistance threshold corresponding to the battery pack, and correcting the dynamic internal resistance threshold of the battery to the internal resistance threshold corresponding to the battery pack.

For example, assuming that the quantile is 10%, if there are 100 batteries of the same type in a certain battery pack, after determining the dynamic internal resistance thresholds corresponding to the 100 batteries respectively, the dynamic internal resistance thresholds may be ranked from small to large, and the dynamic internal resistance threshold located in the first 10% in the ranking result is selected, and the maximum value in the dynamic internal resistance thresholds of the first 10% is used as the internal resistance threshold corresponding to the battery pack.

Further, for the first 10% of the dynamic internal resistance threshold, since it is lower than the internal resistance threshold corresponding to the battery pack, the first 10% of the dynamic internal resistance threshold may be corrected to the internal resistance threshold corresponding to the battery pack. And for the dynamic internal resistance threshold value higher than the internal resistance threshold value corresponding to the battery pack, no correction is needed.

In this embodiment, the predicted internal resistance values corresponding to the respective batteries may be predicted based on the historical operation data of the respective batteries within the specified time window, and the predicted internal resistance values may be compared with the dynamic internal resistance threshold.

The above-mentioned time window refers to a critical time period playing an important role in prediction, for example, when it is desired to predict the internal resistance of the battery after one month in the future, the time window may be selected to be the last two months, and the specific selection of the time window may be determined by those skilled in the art according to actual needs, which is not limited in this specification.

Note that the specific time window corresponding to each battery may or may not be identical, and may be determined by one skilled in the art according to actual needs, which is not limited in this specification.

For example, in the prediction, historical operation data of each battery in a specified time window may be input into a machine learning model trained in advance to predict predicted internal resistance values corresponding to each battery, and the predicted internal resistance values may be compared with a dynamic internal resistance threshold.

It is worth to say that, since the machine learning model does not directly give out whether the battery has a fault or not, but gives out the predicted internal resistance value of the battery, the machine learning model has strong interpretation, is easily accepted by operation and maintenance personnel, and is easy to popularize.

In one embodiment, the historical operating data of the battery over a specified time window further includes one or more of battery voltage, battery temperature, and current data.

For example, the operating state of the battery includes various data, which may include one or more of battery voltage, battery temperature, and current data in addition to critical internal battery resistance, in a specified time window.

Further, the historical operation data of each battery in the appointed time window can be input into a machine learning model which is trained in advance so as to respectively predict the predicted internal resistance value corresponding to each battery;

the machine learning model is characterized by comprising one or more of internal resistance of a battery, temperature of the battery, voltage of the battery, internal resistance threshold value of the battery, service life of the battery and internal resistance change rate during training.

For example, in extracting the effective feature by the feature engineering, one or more of the internal resistance of the battery, the battery temperature, the battery voltage, the internal resistance threshold of the battery, the battery in-use period, and the internal resistance change rate may be used.

In order to improve the efficiency of management and maintenance, the battery that may have an abnormality may be determined by data screening, and prediction may be performed for the battery that may have an abnormality.

In one embodiment, the historical operation data of each battery in the specified time window may be screened to determine the target historical operation data corresponding to the battery in the floating state;

calculating the internal resistance increase rate of the battery according to the internal resistance data in the target historical operation data;

And in response to the internal resistance increase rate reaching a preset threshold, and at least one internal resistance value in the internal resistance data in the target historical operation data reaches a specified multiple of the dynamic internal resistance threshold, predicting a predicted internal resistance value corresponding to each battery based on the internal resistance data in the target historical operation data.

For example, historical operation data of each battery in a specified time window may be screened to determine target historical operation data corresponding to the battery in a float state.

The floating state refers to a state in which the battery is charged with a small current and a constant voltage. Float charging is a way to protect the battery by supplying a small amount of current to the battery to compensate for the loss of local action of the battery pack so that it can often remain in a charge-satisfactory state without overcharging, protecting the life of the battery.

It should be noted that, since the internal resistance values of the batteries are different when the batteries are in different electric quantity states, especially the batteries in a discharging state, the internal resistance is unstable, and the floating state is opposite, at this time, the internal resistance of the batteries is relatively stable, and the batteries have reference significance and can be used as a measurement standard, therefore, the battery data which are not in the floating state need to be filtered.

Further, in the illustrated embodiment, the historical operation data of each battery in the specified time window may be filtered, and data of the historical operation data of each battery, where the battery current is less than 0, may be filtered, and data of the historical operation data of each battery, where the battery voltage is less than the specified threshold, may be filtered, so as to determine the target historical operation data corresponding to the battery in the floating charge state.

In one example, when historical operation data of each battery in a specified time window is screened, data with a battery current smaller than 0 in the historical operation data and data with a battery voltage smaller than a specified threshold (for example, a battery with a rated voltage of 12V and a voltage threshold of 13.1V) can be filtered, and the filtered historical operation data is target historical operation data corresponding to the battery in a floating charge state.

Continuing with the example, since the internal resistance of the battery is normally low, but as the operation time increases, the internal resistance of the battery gradually increases until the use is affected, the target historical operation data can be determined by calculating the internal resistance increase rate in addition to the target historical operation data based on the float state.

In one example, the internal resistance increase rate of the battery in two months may be calculated according to the historical operation data of the battery in two months, if the internal resistance increase rate reaches a preset threshold value (for example, the preset threshold value may be 0.5), it may be determined whether at least one internal resistance value (for example, the internal resistance value of the last sampling point) exists in the historical operation data of the two months, whether a specified multiple (for example, 1.1 times) of the dynamic internal resistance threshold value determined for the battery is reached, and if the internal resistance increase rate reaches, the historical operation data may be determined as the target historical operation data.

It should be noted that, since the batteries having different rated voltages have large differences in internal resistances, different machine learning models can be set for the batteries having different rated voltages.

In one embodiment, a plurality of machine learning models for predicting the internal resistance of the battery are deployed on the battery management platform, wherein the machine learning models are respectively corresponding to different rated voltages;

further, the historical operation data of each battery in the specified time window may be input to a pre-trained machine learning model corresponding to the rated voltage of each battery, and the predicted internal resistance value corresponding to each battery may be predicted.

In one example, for different batteries with rated voltage of 2V and rated voltage of 12V, prediction may be performed using different trained machine learning models, and at the time of prediction, historical operation data of the battery with rated voltage of 2V within a specified time window may be input to a pre-trained machine learning model corresponding to the rated voltage of the battery to predict a predicted internal resistance value corresponding to the battery.

It should be noted that, when the machine learning model is trained, the training samples of the model can be determined based on the data screening method, and because the screened training samples are data in a floating state and have a growing trend, the accuracy of model training can be improved, and model overfitting is avoided.

The machine learning model may be a machine learning model such as LightGBM (Light Gradient Boosting Machine) or XGBoost (eXtreme Gradient Boosting), and is not limited in this specification.

In this embodiment, a fault alarm corresponding to the target battery may be output in response to a value relationship between a predicted internal resistance value corresponding to any one of the respective batteries and a dynamic internal resistance threshold corresponding to the target battery meeting a preset condition.

In one example, the preset condition may be a multiple relationship, and by comparing the predicted internal resistance value with the dynamic internal resistance threshold, a fault alarm corresponding to the target battery may be output when the predicted internal resistance value of any one of the target batteries reaches N times the dynamic internal resistance threshold corresponding to the target battery. The value of N may be determined by an operator according to an operation and maintenance standard, which is not limited in this specification.

In still another example, the preset condition may be an inequality formula, and the predicted internal resistance value and the dynamic internal resistance threshold value are substituted into the formula to determine whether the inequality formula is satisfied, and if so, a fault alarm corresponding to the target battery is output.

Referring to fig. 3, fig. 3 is a flowchart of another battery management method based on internal resistance of a battery according to an exemplary embodiment of the present disclosure, which is applied to a battery management platform. As shown in fig. 3, the method comprises the following steps:

s301, historical operation data of at least one battery in the battery pack is acquired.

The historical operation data at least comprises internal resistance data corresponding to at least one battery in the battery pack, which is obtained by periodic sampling of the battery management platform.

S302, performing anomaly filtering on the internal resistance data.

In the foregoing, on the one hand, the abnormality filtering may be performed based on the internal resistance average value of the cells of the same type, and the internal resistance average value of the cells corresponding to the type may be determined according to the type to which each cell of at least one cell of the battery pack belongs, and based on the internal resistance average value, abnormality filtering may be performed on internal resistance data, in which the internal resistance value is higher than a specified multiple of the internal resistance average value, and internal resistance data, in which the internal resistance value is lower than a specified percentage of the internal resistance average value, in the internal resistance data;

on the other hand, abnormality filtering may also be performed based on the relative degree of change in the internal resistance data. Determining whether internal resistance data with the increase amplitude of the internal resistance value reaching a threshold value compared with the internal resistance value corresponding to one or more previous sampling periods exists in the internal resistance data; and if so, carrying out anomaly filtering on the internal resistance data.

S303, determining a dynamic internal resistance threshold corresponding to each battery in the at least one battery.

The dynamic internal resistance threshold is an internal resistance threshold which is dynamically determined along with the change of the internal resistance data corresponding to each battery.

From the foregoing, in determining the dynamic internal resistance threshold, since the internal resistance change of the battery and the use period have a larger relationship, the longer the use period, the lower the health, the larger the generated internal resistance change, and the shorter the use period, the healthier the battery, and the smaller the internal resistance change, the determination can be made separately based on the use period of the battery.

The method comprises the steps that for each battery in at least one battery in the battery pack, internal resistance data obtained by periodically sampling the battery in a first time period is sequenced according to the time period of the battery operation reaching the first time period, and an internal resistance value reaching a preset quantile in the internal resistance data is used as a dynamic internal resistance threshold corresponding to the battery according to a sequencing result;

Further, as can be seen from the foregoing, for the determined dynamic internal resistance threshold value corresponding to each battery, a certain correction may be performed in units of battery packs.

The method comprises the steps that sorting can be conducted based on dynamic internal resistance thresholds corresponding to all batteries in at least one battery in the battery pack, and the dynamic internal resistance thresholds reaching preset quantiles are used as the internal resistance thresholds corresponding to the battery pack according to sorting results; and correcting the battery with the dynamic internal resistance threshold lower than the internal resistance threshold corresponding to the battery pack, and correcting the dynamic internal resistance threshold of the battery to the internal resistance threshold corresponding to the battery pack.

S304, historical operation data of each battery in a specified time window is obtained.

Wherein the historical operating data may include one or more of battery voltage, battery temperature, and current data in addition to critical internal battery resistance.

In addition, if the data is missing more than a certain percentage, the data can be directly discarded, and if the data is not missing, linear interpolation filling can be performed to repair the data.

S305, screening target historical operation data corresponding to the battery in a floating charge state.

As can be seen from the foregoing, since the internal resistance values of the batteries are different when the batteries are in different electric states, especially the batteries in a discharging state, the internal resistance is unstable, and the floating state is opposite, at this time, the internal resistance of the batteries is relatively stable, and the batteries have a reference meaning and can be used as a measurement standard, therefore, the battery data not in the floating state needs to be filtered.

During filtering, historical operation data of each battery in a specified time window can be filtered, data, in which battery current is smaller than 0, in the historical operation data of each battery is filtered, and data, in which battery voltage is smaller than a specified threshold, in the historical operation data of each battery is filtered, so that target historical operation data corresponding to the battery in a floating charge state are determined.

S306, screening out target historical operation data with an internal resistance increasing trend.

As can be seen from the foregoing, since the internal resistance of the battery is normally low, but as the operation time increases, the internal resistance of the battery gradually increases until the use is affected, the target historical operation data can be determined by calculating the internal resistance increase rate in addition to the target historical operation data based on the float state.

During screening, the internal resistance growth rate of the battery can be calculated according to the internal resistance data in the target historical operation data; and predicting a predicted internal resistance value of the battery based on the internal resistance data in the target historical operation data in response to the internal resistance increase rate reaching a preset threshold and at least one internal resistance value in the internal resistance data in the target historical operation data reaching a specified multiple of the dynamic internal resistance threshold.

S307, predicting the predicted internal resistance value corresponding to each battery.

As can be seen from the foregoing, since the batteries having different rated voltages have large differences in internal resistance, the batteries having different rated voltages can be predicted based on different machine learning models.

In the prediction, the historical operation data of each battery in a specified time window can be input into a pre-trained machine learning model corresponding to the rated voltage of each battery, and the predicted internal resistance value corresponding to each battery is predicted.

S308, comparing the predicted internal resistance value with a dynamic internal resistance threshold.

S309, outputting a fault alarm.

From the foregoing, it is possible to output a fault alarm corresponding to the target battery in response to the numerical relationship between the predicted internal resistance value corresponding to any one of the target batteries and the dynamic internal resistance threshold value corresponding to the target battery meeting a preset condition.

The implementation process is specifically described in the implementation process of the corresponding steps in the battery management method based on the internal resistance of the battery, and relevant parts are referred to in the description of the method implementation mode, and are not repeated here.

In an exemplary embodiment of the present specification, there is also provided an apparatus capable of implementing the above method.

Fig. 4 is a schematic block diagram of an apparatus according to an exemplary embodiment. Referring to fig. 4, at the hardware level, the device includes a processor 402, an internal bus 404, a network interface 406, a memory 408, and a nonvolatile memory 410, although other hardware requirements are possible. One or more embodiments of the present description may be implemented in a software-based manner, such as by the processor 402 reading a corresponding computer program from the non-volatile memory 410 into the memory 408 and then running. Of course, in addition to software implementation, one or more embodiments of the present disclosure do not exclude other implementation manners, such as a logic device or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but may also be hardware or a logic device.

Referring to fig. 5, in a software embodiment, a battery management device 500 based on internal resistance of a battery is provided. As shown in fig. 5, the apparatus 500 includes:

the acquiring module 501 acquires internal resistance data corresponding to at least one battery in the battery pack;

a determining module 502, configured to determine dynamic internal resistance thresholds corresponding to each of the at least one battery according to the internal resistance data; the dynamic internal resistance threshold is an internal resistance threshold which is dynamically determined along with the change of the internal resistance data corresponding to each battery;

a prediction module 503, configured to predict predicted internal resistance values corresponding to the respective batteries based on historical operation data of the respective batteries within a specified time window, and compare the predicted internal resistance values with the dynamic internal resistance threshold;

and the alarm module 504 is used for outputting fault alarms corresponding to the target batteries in response to the fact that the numerical relation between the predicted internal resistance value corresponding to any one of the batteries and the dynamic internal resistance threshold value corresponding to the target battery meets preset conditions.

Optionally, the method is applied to a battery management platform, and the battery management platform is used for monitoring the running state of the battery;

The obtaining module 501 further:

and acquiring historical operation data of at least one battery in the battery pack, wherein the historical operation data at least comprises internal resistance data corresponding to the at least one battery in the battery pack, which is obtained by periodic sampling by the battery management platform.

Optionally, the apparatus 500 further includes:

a first abnormality filtering module 505 (not shown in the figure), which determines an internal resistance average value of the batteries corresponding to the model according to the model to which each of at least one battery in the battery pack belongs, and performs abnormality filtering on internal resistance data, of which the internal resistance value is higher than a specified multiple of the internal resistance average value, among the internal resistance data, and on internal resistance data, of which the internal resistance value is lower than a specified percentage of the internal resistance average value, based on the internal resistance average value;

a second abnormality filtering module 506 (not shown in the figure) that determines whether there is internal resistance data in which an increase amplitude of the internal resistance value reaches a threshold value as compared to the internal resistance value corresponding to the previous one or more sampling periods; and if so, carrying out anomaly filtering on the internal resistance data.

Optionally, the determining module 502 further:

for each battery in at least one battery in the battery pack, responding to the time length of the battery operation reaching a first time length, sequencing internal resistance data obtained by periodically sampling the battery in the first time length, and taking the internal resistance value reaching a preset quantile in the internal resistance data as a dynamic internal resistance threshold corresponding to the battery according to the sequencing result;

Optionally, the apparatus 500 further includes:

a correction module 507 (not shown in the figure) is configured to sort the batteries based on the dynamic internal resistance threshold corresponding to each battery in at least one battery in the battery pack, and take the dynamic internal resistance threshold reaching a preset quantile as the internal resistance threshold corresponding to the battery pack according to the sorting result; and correcting the battery with the dynamic internal resistance threshold lower than the internal resistance threshold corresponding to the battery pack, and correcting the dynamic internal resistance threshold of the battery to the internal resistance threshold corresponding to the battery pack.

Optionally, the historical operating data of the battery within a specified time window further comprises one or more of battery voltage, battery temperature and current data;

the prediction module 503 further:

the historical operation data of each battery in a specified time window is input into a machine learning model which is trained in advance so as to respectively predict the predicted internal resistance value corresponding to each battery; the machine learning model is characterized by comprising one or more of internal resistance of a battery, temperature of the battery, voltage of the battery, internal resistance threshold value of the battery, service life of the battery and internal resistance change rate during training.

Optionally, the apparatus 500 further includes:

a first screening module 508 (not shown in the figure) screens the historical operation data of each battery in the specified time window to determine the target historical operation data corresponding to the battery in the floating charge state;

a second screening module 509 (not shown) for calculating an internal resistance increase rate of the battery based on the internal resistance data in the target historical operating data; and predicting predicted internal resistance values corresponding to the batteries based on the internal resistance data in the target historical operation data in response to the internal resistance increase rate reaching a preset threshold and at least one internal resistance value in the internal resistance data in the target historical operation data reaching a specified multiple of the dynamic internal resistance threshold.

Optionally, the first screening module 508 further:

and screening the historical operation data of each battery in a specified time window, filtering out the data of which the battery current is smaller than 0 in the historical operation data of each battery, and filtering out the data of which the battery voltage is smaller than a specified threshold in the historical operation data of each battery, so as to determine the target historical operation data corresponding to the battery in a floating charge state.

Optionally, a plurality of machine learning models for predicting the internal resistance of the battery, which correspond to different rated voltages respectively, are deployed on the battery management platform;

the prediction module 503 further:

and inputting the historical operation data of each battery in a specified time window into a pre-trained machine learning model corresponding to the rated voltage of each battery, and respectively predicting the predicted internal resistance value corresponding to each battery.

Optionally, the battery management platform includes a predictive maintenance PM platform, or a monitoring platform corresponding to a data center.

The implementation process of the functions and roles of each module in the apparatus 500 is specifically shown in the implementation process of the corresponding steps in the battery management method based on the internal resistance of the battery, and the relevant parts only need to be referred to in the description of the method implementation mode, which is not repeated here.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical modules, i.e., may be located in one place, or may be distributed over a plurality of network modules. Some or all of the units or modules may be selected according to actual needs to achieve the purposes of the present description. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer, which may be in the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or a combination of any of these devices.

In a typical configuration, a computer includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, read only compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage, quantum memory, graphene-based storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by the computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The terminology used in the one or more embodiments of the specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of the specification. As used in this specification, one or more embodiments and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used in one or more embodiments of the present description to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of one or more embodiments of the present description. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.

The foregoing description of the preferred embodiment(s) is (are) merely intended to illustrate the embodiment(s) of the present invention, and it is not intended to limit the embodiment(s) of the present invention to the particular embodiment(s) described.

Claims

1. A battery management method based on internal resistance of a battery, comprising:

2. The method of claim 1, applied to a battery management platform for monitoring an operational status of the battery;

the obtaining internal resistance data corresponding to at least one battery in the battery pack includes:

3. The method of claim 1, further comprising, prior to separately determining dynamic internal resistance thresholds corresponding to respective ones of the at least one battery from the internal resistance data:

determining an internal resistance average value of batteries corresponding to the model according to the model of each battery in at least one battery pack, and performing abnormal filtering on internal resistance data with internal resistance values higher than a designated multiple of the internal resistance average value in the internal resistance data and internal resistance data with internal resistance values lower than a designated percentage of the internal resistance average value in the internal resistance data based on the internal resistance average value;

4. The method of claim 1, the determining, from the internal resistance data, dynamic internal resistance thresholds corresponding to respective ones of the at least one battery, respectively, comprising:

5. The method of claim 4, the method further comprising:

sorting based on the dynamic internal resistance threshold corresponding to each battery in at least one battery in the battery pack, and taking the dynamic internal resistance threshold reaching a preset quantile as the internal resistance threshold corresponding to the battery pack according to the sorting result;

6. The method of claim 1, the historical operating data of the battery over a specified time window further comprising one or more of battery voltage, battery temperature, and current data;

the predicting the predicted internal resistance value corresponding to each battery based on the historical operation data of each battery in the appointed time window includes:

7. The method of claim 1, further comprising, prior to predicting respective predicted internal resistance values for the respective batteries based on historical operating data of the respective batteries within a specified time window:

Screening the historical operation data of each battery in a specified time window, and determining target historical operation data corresponding to the battery in a floating charge state;

8. The method of claim 7, wherein the screening the historical operation data of each battery in the specified time window to determine the target historical operation data corresponding to the battery in the floating charge state comprises:

9. The method of claim 2, wherein a plurality of machine learning models for predicting internal resistance of the battery are deployed on the battery management platform, each corresponding to a different rated voltage;

10. The method of claim 2, the battery management platform comprising a predictive maintenance PM platform, or a monitoring platform corresponding to a data center.

11. A battery management apparatus based on internal resistance of a battery, comprising:

12. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to implement the method of any of claims 1-10 by executing the executable instructions.

13. A machine-readable storage medium having stored thereon machine-readable instructions which, when executed by a processor, implement the steps of the method of any of claims 1-10.