Method, apparatus, device and medium for evaluating device operating behavior

Abstract

The application discloses a device working condition behavior evaluation method, device, equipment and medium, the method comprises the following steps: using the battery related data of a target device under at least one working condition behavior, respectively, to extract the first characteristic value of the key characteristic item corresponding to each working condition behavior, wherein the first characteristic value of the key characteristic item corresponding to the working condition behavior represents the parameter characteristics of the battery of the target device under the working condition behavior; and obtaining the reference characteristic value corresponding to each key characteristic item, which is obtained by analyzing the second characteristic value of the key characteristic item corresponding to the reference device of different battery health degrees; compare the first characteristic value of each key characteristic item with the corresponding reference characteristic value to determine whether the working condition behavior corresponding to each key characteristic item affects the battery health degree of the target device. The above scheme can qualitatively evaluate whether the working condition behavior of the target device affects the battery health degree.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of batteries, and in particular to a method and device for evaluating device working condition behavior, and a medium. BACKGROUND

[0002] Battery health is the ratio of the current capacity of a battery to its original design capacity, and is an indicator of battery performance and capacity state, usually expressed in percentage form. Battery health has an important influence on the battery life and charging speed, and is very important for improving user experience. However, the prior art cannot qualitatively analyze whether a specific device working condition behavior will affect the battery health. SUMMARY

[0003] The present application provides at least a method and device for evaluating device working condition behavior, which can qualitatively evaluate whether the working condition behavior of a target device affects the battery health.

[0004] The present application provides a method for evaluating device working condition behavior, comprising: based on battery-related data of a target device under at least one working condition behavior, extracting first feature values of key feature items corresponding to each working condition behavior, wherein the first feature values of the key feature items corresponding to the working condition behavior represent the parameter characteristics of the battery of the target device under the working condition behavior; and obtaining reference feature values corresponding to each key feature item, which are obtained by analyzing second feature values of the key feature items of a plurality of reference devices corresponding to different battery health; and comparing the first feature values of each key feature item with the corresponding reference feature values to determine whether the working condition behavior corresponding to each key feature item affects the battery health of the target device.

[0005] In the above scheme, the feature values of the reference devices with different battery health are analyzed to determine under what circumstances the feature values will affect the battery health, and the reference feature values representing whether the battery health is affected are obtained, which are used for comparison with the first feature values of the target vehicle, thereby qualitatively evaluating whether the working condition behavior of the target device affects the battery health.

[0006] In some embodiments, the battery-related data under the working condition behavior includes battery time-series data of a plurality of battery parameters under the working condition behavior; the battery time-series data of the battery parameters includes parameter values of the battery parameters at a plurality of historical time points when the working condition behavior occurs; based on the battery-related data of the target device under at least one working condition behavior, the first feature value of the key feature corresponding to each working condition behavior is extracted, including: for each working condition behavior, the plurality of battery time-series data is divided into at least one time-series data group corresponding to the working condition behavior, each time-series data group including battery time-series data of at least one battery parameter; for each time-series data group corresponding to the working condition behavior, at least one statistic is performed on the time-series data group under the working condition behavior to obtain the first feature value of at least one key feature corresponding to the working condition behavior.

[0007] In the above scheme, at least one statistical method is used to extract the key features from the data, which can be used to realize qualitative judgment.

[0008] In some embodiments, the first feature value of the key feature includes at least one of the following: a time representation feature value and a parameter value representation feature value corresponding to the time-series data group, the time representation feature value being obtained by performing time statistics on the battery time-series data in the time-series data group, and the parameter value representation feature value being obtained by performing statistics on the parameter values contained in the battery time-series data in the time-series data group.

[0009] In the above scheme, two feature extraction methods are provided, one of which is to directly perform statistics on the parameter values of the time-series data and the other is to perform time statistics on the time-series data, which can flexibly extract the key features to reflect the working condition behavior of the device.

[0010] In some embodiments, before the first feature value of the key feature corresponding to each working condition behavior is extracted based on the battery-related data of the target device under at least one working condition behavior, the method further includes: obtaining original battery data collected by the battery of the target device, the original battery data including a plurality of parameter values of a plurality of battery parameters corresponding to different historical time points within a first historical time period; based on the working condition behavior of the target device within the first historical time period, the original battery data is divided to obtain battery-related data under each working condition behavior, wherein the battery-related data under the working condition behavior includes parameter values of each battery parameter at a plurality of historical time points when the working condition behavior occurs.

[0011] In the above scheme, the original data is divided into battery-related data corresponding to each working condition behavior, and then each working condition behavior can be analyzed respectively to realize qualitative evaluation of different working condition behaviors.

[0012] In some embodiments, before the first feature value of each key feature item corresponding to each working condition behavior is extracted based on the battery-related data of the target device under each working condition behavior, the method further comprises: constructing a battery feature item set, wherein the battery feature item set comprises a plurality of battery feature items; analyzing the relationship between each battery feature item in the battery feature item set and the battery health degree; and selecting, as the key feature item, a battery feature item related to the battery health degree from the battery feature item set based on the relationship.

[0013] In the above scheme, the battery feature item set is constructed in advance, and the feature items in the set are screened to select the feature items affecting the battery health degree to participate in subsequent qualitative evaluation, which can avoid ineffective evaluation of feature items that do not affect the battery health degree and improve efficiency.

[0014] In some embodiments, obtaining the reference feature value corresponding to each key feature item comprises: obtaining second feature values of each key feature item corresponding to a plurality of reference devices of different battery health degrees; clustering the second feature values of each key feature item corresponding to the plurality of reference devices based on the battery health degree to obtain a first feature cluster and a second feature cluster, wherein the first feature cluster comprises second feature values of each key feature item corresponding to a first reference device, and the second feature cluster comprises second feature values of each key feature item corresponding to a second reference device, the first reference device being each reference device belonging to a first battery health degree level, and the second reference device being each reference device belonging to a second battery health degree level; for each key feature item, counting the second feature values of the key feature item corresponding to each first reference device in the first feature cluster to obtain a first statistical value, and counting the second feature values of the key feature item corresponding to each second reference device in the second feature cluster to obtain a second statistical value; and obtaining the reference feature value corresponding to the key feature item based on the first statistical value and the second statistical value.

[0015] In the above scheme, the reference devices are clustered according to the battery health degree to distinguish the reference devices with high health degree and the reference devices with low health degree, and the difference between the feature values of the reference devices with high and low health degrees can be analyzed based thereon, so that the reference feature value for qualitative evaluation can be obtained.

[0016] In some embodiments, the first statistical value is a cluster center of the second feature values of the key feature item in the first feature cluster, and the second statistical value is a cluster center of the second feature values of the key feature item in the second feature cluster.

[0017] In the above scheme, the cluster center is used as the representative of the cluster, which can accurately represent the feature value situation of the two clusters with high and low health degrees, and thus the accurate reference feature value can be obtained.

[0018] In some embodiments, based on the first statistical value and the second statistical value, the reference feature value corresponding to the key feature item is obtained by calculating a distance between the first statistical value and the second statistical value as the reference feature value corresponding to the key feature item.

[0019] In the above scheme, the distance between the two can represent the difference in performance of the equipment with high and low health degrees on the key feature item, and can be used to distinguish whether the working condition behavior of the target equipment affects the health degree.

[0020] In some embodiments, the first feature value of each key feature item is compared with the corresponding reference feature value to determine whether the working condition behavior corresponding to each key feature item affects the battery health degree of the target equipment, including: in response to the first feature value of the key feature item and the corresponding reference feature value satisfying a first size relationship, determining that the working condition behavior corresponding to the key feature item affects the battery health degree of the target equipment; and in response to the first feature value of the key feature item and the corresponding reference feature value satisfying a second size relationship, determining that the working condition behavior corresponding to the key feature item does not affect the battery health degree of the target equipment.

[0021] In the above scheme, on the basis of setting the reference feature value, the qualitative evaluation of the working condition behavior of the target equipment is realized by judging the size relationship between the first feature value and the reference feature value.

[0022] In some embodiments, the first feature value of the key feature item is determined based on the parameter value of the battery parameter corresponding to the key feature item in the second historical time period; and after comparing the first feature value of each key feature item with the corresponding reference feature value to determine whether the working condition behavior corresponding to each key feature item affects the battery health degree of the target equipment, the method further includes: in response to determining that the working condition behavior corresponding to the key feature item affects the battery health degree of the target equipment, analyzing the feature value change of the associated time sequence feature of the key feature item in time sequence to determine the influence degree of the working condition behavior corresponding to the key feature item on the battery health degree, wherein the associated time sequence feature of the key feature item includes third feature values of the associated feature items corresponding to different time units, and the associated feature items are the key feature item or other feature items, and the different time units are located in the second historical time period, and the third feature value of the time unit is determined based on the parameter value of the battery parameter corresponding to the associated feature item in the time unit.

[0023] In the above scheme, the working condition behavior corresponds to the associated time sequence feature, and analyzing the feature value change of the associated time sequence feature can reflect the change of the influence of the corresponding working condition behavior on the battery health degree.

[0024] In some embodiments, the first feature value of the key feature item is obtained by statistically processing parameter values of the battery parameter corresponding to the key feature item in the second historical time period; the third feature value of the time unit is obtained by statistically processing parameter values of the battery parameter corresponding to the associated feature item in the time unit; the associated time sequence feature of the key feature item includes the third feature values of the key feature item corresponding to different time units respectively; the feature value change of the associated time sequence feature of the key feature item in the time sequence is analyzed to determine the influence degree of the working condition behavior corresponding to the key feature item on the battery health degree, including: clustering the associated time sequence feature of the key feature item to determine the frequency change of the target time period of the working condition behavior corresponding to the key feature item in the second historical time period relative to other time periods; based on the frequency change, it is determined that the working condition behavior corresponding to the key feature item in the target time period increases or decreases the influence on the battery health degree.

[0025] In the above scheme, by clustering the associated time sequence feature, the change of the feature value in the time sequence can be determined, and then the influence change of the battery health degree in different time periods can be judged.

[0026] In some embodiments, after comparing the first feature value of each key feature item with the corresponding reference feature value to determine whether the working condition behavior corresponding to each key feature item affects the battery health degree of the target device, the method further includes: providing the user with the influence evaluation result of the working condition behavior corresponding to the key feature item on the battery health degree; providing the user with the feature data corresponding to the key feature item.

[0027] In the above scheme, the influence evaluation result and the corresponding feature data can be prompted to the user, so as to help the user understand the influence of the working condition behavior on the battery health degree.

[0028] In some embodiments, the working condition behavior includes at least one of charging, discharging and standing; the battery related data under the working condition behavior includes at least one parameter value of a plurality of battery parameters respectively at the time of occurrence of the working condition behavior, and the plurality of battery parameters include at least one of current, voltage, state of charge and temperature; the target device is a vehicle.

[0029] The application provides an evaluation device for device working condition behavior, comprising an extraction module, a reference acquisition module and a comparison module, the extraction module is used for extracting battery-related data of a target device under at least one working condition behavior, and a first feature value of a key feature item corresponding to each working condition behavior is extracted, wherein the first feature value of the key feature item corresponding to the working condition behavior represents a parameter feature of the battery of the target device under the working condition behavior; the reference acquisition module is used for acquiring a reference feature value corresponding to each key feature item, the reference feature value is obtained by analyzing second feature values of the key feature items corresponding to a plurality of reference devices with different battery health degrees; and the comparison module is used for comparing the first feature value of each key feature item with the corresponding reference feature value to determine whether the working condition behavior corresponding to each key feature item affects the battery health degree of the target device.

[0030] The application provides a computer readable storage medium, which stores program instructions, and the program instructions are executed by a processor to implement the above method.

[0031] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, but not limiting the application. BRIEF DESCRIPTION OF DRAWINGS

[0032] The accompanying drawings incorporated in the specification and forming a part of it illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.

[0033] Figure 1 is a flowchart of an evaluation method for device working condition behavior provided by some embodiments of the application;

[0034] Figure 2 is a framework diagram of an evaluation device for device working condition behavior provided by some embodiments of the application;

[0035] Figure 3 is a framework diagram of an electronic device provided by some embodiments of the application;

[0036] Figure 4 is a framework diagram of a computer readable storage medium provided by some embodiments of the application. DETAILED DESCRIPTION

[0037] The scheme of the embodiments of the application will be described in detail below with reference to the accompanying drawings.

[0038] In the following description, specific details are set forth in order to provide a thorough understanding of the application, but the application is not limited to the specific details described.

[0039] The term "and / or", used in the present document, only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone. In addition, the character " / " in the present document generally represents an "or" relationship between the front and rear associated objects. In addition, "multiple" in the present document means two or more than two. In addition, the term "at least one" in the present document means any one of multiple or any combination of at least two of multiple, for example, including at least one of A, B and C can mean including any one or more elements selected from the set consisting of A, B and C.

[0040] Battery health is the ratio of the current capacity of the battery to its original design capacity, which is an indicator of the performance and capacity state of the battery, usually expressed in percentage. Battery health has an important influence on the endurance time and charging speed of the battery, and is very important for improving user experience. It is impossible to qualitatively analyze whether the specific device working condition behavior will affect the battery health in the prior art.

[0041] The researchers of the present application found that the same feature item of the device with different battery health is different, and the analysis of the feature item of the device with different health can determine under what circumstances the feature item will affect the battery health, and obtain the threshold value of the key feature item representing the influence on the battery health. The first feature value of the target vehicle is compared, so as to qualitatively evaluate whether the working condition behavior of the target device affects the battery health.

[0042] The evaluation method of the device working condition behavior disclosed in the embodiments of the present application can be applied to any device provided with a battery, for example, the device can be a new energy vehicle and the like. Wherein, the type of the battery provided on the device is not limited here, and exemplarily can be a lithium ion battery and the like.

[0043] Referring to Figure 1 , Figure 1 is a flowchart of the evaluation method of the device working condition behavior provided by some embodiments of the present application. Specifically, the method comprises:

[0044] Step S110: Based on the battery related data of the target device under at least one working condition behavior, the first feature value of the key feature item corresponding to each working condition behavior is extracted.

[0045] Wherein, the target device can be any device provided with a battery. The working condition behavior can represent the working state of the battery provided on the target device.

[0046] In some embodiments, the working condition behavior can include at least one of charging, discharging and standing.

[0047] In a specific application scenario, the working condition behaviors include charging, discharging and standing. Further, the charging, discharging and standing as the working condition behaviors can be further subdivided into a plurality of sub-working condition behaviors. Exemplarily, the specific sub-working condition behaviors can be further subdivided on the basis of charging, discharging and standing, for example, charging can be divided into fast charging and ordinary charging. A working condition behavior can also be subdivided from different angles to obtain the subdivision results under each angle, for example, charging can be divided into fast charging and ordinary charging according to the charging current, and charging can be divided into low-temperature charging and high-temperature charging according to the charging temperature.

[0048] In some embodiments, the battery-related data can include a plurality of battery parameters. Among them, the aforementioned battery parameters can include at least one of current, voltage, state of charge, and temperature.

[0049] It can be understood that the battery parameters can be directly collected from the battery, such as current, voltage and temperature, or the battery parameters can be indirectly collected, that is, calculated from the directly collected data, such as state of charge (SOC).

[0050] In a specific application scenario, the battery on the target device is continuously working for a period of time, and the battery parameters can be continuously obtained during the working process of the battery to obtain a plurality of parameter values of the battery parameters, for example, the current, voltage and temperature at different times during its working period, and further, the state of charge corresponding to each time.

[0051] The battery parameters can be associated with the working condition behaviors. The battery-related data under a certain working condition behavior can include a plurality of battery parameters when the working condition behavior occurs. For example, the battery-related data of charging can include charging current, charging voltage, charging temperature, and state of charge during charging.

[0052] Each working condition behavior can correspond to its own key feature item, and a working condition behavior can correspond to a plurality of key feature items, such as charging key feature item, discharging key feature item and standing key feature item. The key feature items of each working condition behavior can be determined in advance, and the feature values of the target device corresponding to each key feature item can be extracted from the battery-related data of the target device to represent the parameter characteristics of the battery of the target device under each working condition behavior. Specifically, each working condition behavior is processed separately, and the first feature value of the key feature item corresponding to the working condition behavior can be extracted from the battery-related data of the target device under the working condition behavior, and the first feature value can represent the parameter characteristics of the battery of the target device under the working condition behavior.

[0053] In a specific application scenario, taking the charging condition as an example, the first feature value of each charging key feature item corresponding to charging is extracted by using the battery-related data of the target device during charging.

[0054] Step S120: Obtain the reference feature value corresponding to each key feature item.

[0055] Each key feature item has a reference feature value corresponding thereto, and the reference feature value is obtained by analyzing the second feature values of the key feature items corresponding to a plurality of reference devices. The battery health degrees of the plurality of reference devices are different. The battery characteristics set on the reference devices are consistent or similar to the battery characteristics set on the target device, so that the analysis of the feature values of the reference devices can be applicable to the target device.

[0056] The key feature item is a feature item that is associated with the battery health degree, and the performance of the reference devices with different health degrees on the key feature item, i.e., the second feature value, can determine the feature value of the key feature item that affects the battery health degree, and obtain the reference feature value of the corresponding key feature item. The reference feature value can be used to indicate that the critical value of the key feature item that affects the battery health degree.

[0057] Step S130: Compare the first feature value of each key feature item with the corresponding reference feature value to determine whether the condition behavior corresponding to each key feature item affects the battery health degree of the target device.

[0058] For each key feature item, the first feature value of the target device and the corresponding reference feature value are compared to determine whether the condition behavior corresponding to the key feature item affects the battery health degree of the target device.

[0059] It should be noted that after obtaining the first feature value and the corresponding reference feature value, the sizes of the two can be compared to determine whether the corresponding condition behavior affects the battery health degree of the target device.

[0060] In some embodiments, in response to the first feature value of the key feature item and the corresponding reference feature value satisfying a first size relationship, it is determined that the condition behavior corresponding to the key feature item affects the battery health degree of the target device, and in response to the second feature value of the key feature item and the corresponding reference feature value satisfying a second size relationship, it is determined that the condition behavior corresponding to the key feature item does not affect the battery health degree of the target device.

[0061] The size relationship that may occur between the first feature value and the corresponding reference feature value is divided into a first size relationship and a second size relationship, the former corresponds to the condition behavior affecting the battery health degree of the target device, and the latter corresponds to the condition behavior not affecting the battery health degree of the target device.

[0062] In a specific application scenario, the first size relationship indicates that the first feature value is greater than or equal to the corresponding reference feature value, and the second size relationship indicates that the first feature value is less than the corresponding reference feature value.

[0063] It should be noted that the first size relationship corresponding to each key feature item is predetermined. The first size relationship corresponding to different key feature items can be different. For example, for feature A, when the first feature value is greater than the corresponding reference feature value, the corresponding working condition behavior affects the battery health of the target device, and when the first feature value is less than or equal to the corresponding reference feature value, the corresponding working condition behavior does not affect the battery health of the target device. For feature B, when the first feature value is less than or equal to the corresponding reference feature value, the corresponding working condition behavior affects the battery health of the target device, and when the first feature value is greater than the corresponding reference feature value, the corresponding working condition behavior does not affect the battery health of the target device.

[0064] It can be understood that before the first feature value is extracted, the key feature items corresponding to each working condition behavior can be predetermined. There are a large number of battery-related feature items, and the feature items that affect the battery health can be selected as key feature items.

[0065] In some embodiments, a battery feature item set is constructed, the set including a plurality of battery feature items, the relationship between each battery feature item in the set and the battery health is analyzed, and the battery feature items related to the battery health are selected from the set as key feature items based on the foregoing relationship.

[0066] It should be noted that the battery feature item set can be constructed around the working condition behavior. For example, the battery feature items corresponding to each working condition behavior are constructed respectively to form the battery feature item set.

[0067] It can be understood that the method provided by the embodiments of the present application can be executed by the target processing device, wherein the foregoing steps of constructing the battery feature item set and selecting the key feature items, and the step of determining the reference feature value can be executed by other processing devices, and the key feature items, the reference feature values and the like data are provided to the target processing device by the other processing devices.

[0068] In some application scenarios, a DOE (Design of Experiments) comparison analysis experiment or battery EIS (Electrochemical Impedance Spectroscopy) data collection and analysis can be set up to determine whether the battery feature items affect the battery health. The battery EIS data can include polarization resistance and polarization capacitance.

[0069] In some application scenarios, the key feature item corresponding to charging can include fast charging proportion, ambient temperature. The key feature item corresponding to standing can include standing minimum temperature.

[0070] It should be noted that the feature value corresponding to the key feature item can be extracted in a certain feature extraction manner from the battery related data of the actual device under the working condition behavior, such as the first feature value corresponding to the target device under a certain working condition behavior, and the second feature value corresponding to the reference device under a certain working condition behavior.

[0071] The feature extraction manner corresponding to the key feature item can also be determined in advance. The feature extraction manner can include a time range and a data processing manner. The time range can be used to limit the battery related data in which time range to extract the feature value. The data processing manner can be used to limit which battery related data in the foregoing time range is processed in which way to obtain the feature value. Further, the data processing manner can include which one or more battery parameters in the battery related data is used for processing, and can also include the processing means of the one or more battery parameters, for example, selecting the maximum value, calculating the average value, judging whether it is greater than a threshold value, etc.

[0072] In some application scenarios, taking the fast charging proportion as an example, the time range corresponding to the key feature item can be determined as the whole life cycle, or one year before the current time, that is, the battery related data in the whole life cycle of the device, or the battery related data in the range of one year before the current time is used for feature extraction. The charging current in the foregoing time range can be selected, and whether the current charging is fast charging is judged according to the size of the charging current. The ratio of the fast charging times to the total charging times can be calculated to obtain the feature value corresponding to the fast charging proportion.

[0073] In some application scenarios, the charging current in the foregoing time range can be selected, and whether it is fast charging is judged according to the size of the charging current. The ratio of the fast charging time to the total charging time can be calculated to obtain the feature value corresponding to the fast charging proportion.

[0074] In some application scenarios, taking the standing temperature minimum value as an example, the time range corresponding to the key feature item can be determined as the whole life cycle, and the battery related data in the whole life cycle of the device is used for feature extraction. The standing temperature in the foregoing time range can be selected, the standing temperature minimum value in a standing process is selected, and the average value of all standing temperature minimum values in the foregoing time range is counted as the feature value corresponding to the standing temperature minimum value.

[0075] Further, the key feature item corresponding to the working condition behavior can be further divided into a key feature item corresponding to a sub-working condition behavior, for example, the fast charging proportion can correspond to fast charging. In the case where there is such an association, it can be further determined whether the sub-working condition behavior affects the battery health degree.

[0076] In some embodiments, before extracting the first feature value from the battery-related data, the method further comprises: obtaining raw battery data collected by the battery of the target device; and dividing the raw battery data to obtain battery-related data under each working condition behavior based on the working condition behavior of the target device in the first historical time period.

[0077] The raw battery data includes a plurality of parameter values of a plurality of battery parameters corresponding to different historical time points in the first historical time period.

[0078] The working condition behavior of the target device in the first historical time period is determined in advance. For example, the working condition behavior of the target device can be determined by the current of the battery. Specifically, when there is no current or the current is close to zero, it is considered to be in a static state; when the battery current is positive, it is considered to be in a discharging state; and when the current is negative, it is considered to be in a charging state.

[0079] After the working condition behavior at different historical time points is determined, the raw battery data can be divided according to the working condition behavior to obtain battery-related data under each working condition behavior. The battery-related data under a certain working condition behavior includes a plurality of historical time points of each battery parameter under the working condition behavior.

[0080] It should be noted that the working condition behavior of the device can switch during the working process. For example, the first historical time period can be divided into four parts, the working condition behavior of the first part is charging, the working condition behavior of the second part is static, the working condition behavior of the third part is discharging, and the working condition behavior of the fourth part is charging. The battery parameter values of the historical time points in the first part and the battery parameter values of the historical time points in the fourth part are divided into battery-related data corresponding to charging. The battery parameter values of the historical time points in the second part are divided into battery-related data corresponding to static. The battery parameter values of the historical time points in the third part are divided into battery-related data corresponding to discharging.

[0081] The parameter values of a battery parameter at a plurality of historical time points when a certain working condition behavior occurs can jointly constitute battery time series data of the battery parameter under the working condition behavior. The battery-related data under a certain working condition behavior includes battery time series data of all battery parameters under the working condition behavior.

[0082] In some embodiments, the first feature value of the key feature item corresponding to each working condition behavior is extracted based on the battery related data of the target device under the at least one working condition behavior, including: for each working condition behavior, dividing a plurality of battery time series data into at least one time series data group corresponding to the working condition behavior, each time series data group including battery time series data of at least one battery parameter; for each time series data group corresponding to the working condition behavior, performing at least one statistical operation on the time series data group under the working condition behavior to obtain the first feature value of at least one key feature item corresponding to the working condition behavior.

[0083] It can be understood that the battery related data under a working condition behavior includes battery time series data of each battery parameter under the working condition behavior. The battery time series data of each battery parameter can be used to form at least one time series data group, and one time series data group can include battery time series data of one or more battery parameters.

[0084] In a specific application scenario, the battery time series data of each battery parameter can form a time series data group respectively. The battery time series data of two or more battery parameters can also form the same time series data group.

[0085] In some embodiments, the first feature value of the key feature item corresponding to each working condition behavior is extracted based on the battery related data of the target device under the at least one working condition behavior, including: for each working condition behavior, dividing a plurality of battery time series data into at least one time series data group corresponding to the working condition behavior, each time series data group including battery time series data of at least one battery parameter; for each time series data group corresponding to the working condition behavior, performing at least one statistical operation on the time series data group under the working condition behavior to obtain the first feature value of at least one key feature item corresponding to the working condition behavior.

[0086] The first feature value can be characterized from at least one of a time dimension and a parameter value dimension. In some embodiments, the first feature value of the key feature item includes at least one of a time characterization feature value and a parameter value characterization feature value corresponding to the time series data group. The time characterization feature value is obtained by performing time statistics on the battery time series data in the time series data group, and the parameter value characterization feature value is obtained by performing parameter value statistics on the battery time series data in the time series data group.

[0087] In some application scenarios, the parameter value characterization feature value can include a minimum resting temperature.

[0088] In some application scenarios, the parameter value characterization feature value includes a statistical parameter value of the battery parameter corresponding to the time series data group in at least one time dimension, for example, a minimum resting temperature within one month, a minimum resting temperature within one quarter, etc.

[0089] In some application scenarios, the time characteristic value comprises at least one of a duration of the working condition behavior and a proportion of the working condition behavior. Exemplarily, the time characteristic value can comprise a fast charging proportion, which can be obtained according to a ratio of a fast charging time to a total charging duration, or a number of fast charging times in a period of time.

[0090] In some embodiments, obtaining the reference characteristic value corresponding to each key feature item can comprise: obtaining a second characteristic value corresponding to each key feature item of each reference device in the plurality of reference devices. Clustering the second characteristic values of each key feature item of the plurality of reference devices based on the battery health degrees, to obtain a first feature cluster and a second feature cluster. For each key feature item, counting the second characteristic values of the key feature item of each first reference device in the first feature cluster to obtain a first statistical value, and counting the second characteristic values of the key feature item of each second reference device in the second feature cluster to obtain a second statistical value. Obtaining the reference characteristic value corresponding to the key feature item based on the first statistical value and the second statistical value.

[0091] In the method, the battery health degrees of the plurality of reference devices are different. For each reference device, its battery health degree and a piece of characteristic value data can be obtained, and the characteristic value data comprises a second characteristic value corresponding to each key feature item of the reference device. In the clustering, the clustering is based on the battery health degrees of the reference devices, and a piece of characteristic value data corresponding to one reference device is taken as a single object for clustering, and all objects are clustered.

[0092] In some implementation scenarios, a clustering method with a computable cluster center, represented by K-means, can be used.

[0093] Before clustering, the number of clusters can be determined in advance. Specifically, it is determined in advance that all objects need to be divided into two clusters, that is, the battery health degrees of the reference devices are divided into two levels, the first feature cluster corresponds to a first battery health degree level, and the second feature cluster corresponds to a second battery health degree level. One of the first battery health degree level and the second battery health degree level represents a high battery health degree, and the other represents a low battery health degree.

[0094] The first feature cluster comprises the characteristic value data of the first reference device, that is, the second characteristic values corresponding to each key feature item of the first reference device, and the second feature cluster comprises the characteristic value data of the second reference device, that is, the second characteristic values corresponding to each key feature item of the second reference device. The first reference device is each reference device belonging to the first battery health degree level, and the second reference device is each reference device belonging to the second battery health degree level.

[0095] After the clustering of the characteristic value data is completed, the calculation of the reference characteristic value is performed for each key feature item respectively.

[0096] In some application scenarios, the battery health of 100 reference devices and the second feature values of the respective key feature items are obtained respectively. The feature value data of the 100 reference devices are divided into two clusters according to the battery health, to obtain a first feature cluster and a second feature cluster. The first feature cluster contains feature value data of 77 first reference devices, and the second feature cluster contains feature value data of 33 second reference devices. Each key feature item is processed respectively. The second feature values of the key feature item A corresponding to the 77 first reference devices are counted to obtain a first statistical value of the key feature item A, and the second feature values of the key feature item A corresponding to the 33 second reference devices are counted to obtain a second statistical value of the key feature item A. The reference feature value of the key feature item A is obtained by using the first statistical value and the second statistical value of the key feature item A.

[0097] In some embodiments, the first statistical value and the second statistical value can be obtained by calculating the cluster center.

[0098] In some application scenarios, the first statistical value is the cluster center of the second feature values of the corresponding key feature item in the first feature cluster, and the second statistical value is the cluster center of the second feature values of the corresponding key feature item in the second feature cluster.

[0099] In some embodiments, the reference feature value of the key feature item can be obtained by calculating the distance between the first statistical value and the second statistical value corresponding to the key feature item as the reference feature value of the key feature item.

[0100] In some application scenarios, the Euclidean distance between the first statistical value and the second statistical value corresponding to the key feature item can be calculated, and the Euclidean distance is taken as the reference feature value of the key feature item.

[0101] In some application scenarios, the first statistical value and the second statistical value are denoted as x1 and x2 respectively. The Euclidean distance is denoted as d, and the Euclidean distance can be obtained by the following formula:

[0102]

[0103] In some application scenarios, the Mahalanobis distance between the first statistical value and the second statistical value corresponding to the key feature item can be calculated, and the Mahalanobis distance is taken as the reference feature value of the key feature item.

[0104] In some application scenarios, for a key feature item, the cluster center of the second feature values in the first feature cluster is denoted as P, and the cluster center of the second feature values in the second feature cluster is denoted as Q. The Mahalanobis distance D M (P, Q) can be obtained by the following formula:

[0105]

[0106] S indicates the covariance of the data set. In the embodiments of the present application, the covariance of the second characteristic value of the key feature item corresponding to all reference devices can be determined.

[0107] It should be noted that, when determining whether the corresponding working condition behavior affects the battery health of the target device by using the reference characteristic value, in addition to the reference characteristic value, the first size relationship and the second size relationship corresponding to the key feature item also need to be determined.

[0108] In some application scenarios, the reference characteristic value of the key feature item A is 60%, and it is also necessary to determine that under the condition that the first characteristic value is greater than or equal to 60%, it can be determined that the working condition behavior affects the battery health of the target device. Under the condition that the first characteristic value is less than 60%, it can be determined that the working condition behavior does not affect the battery health of the target device.

[0109] In some embodiments, the first size relationship and the second size relationship can be determined based on the size relationship between at least two of the statistical value of the high battery health class cluster, the statistical value of the low battery health class cluster, and the reference characteristic value.

[0110] For example, if the statistical value of the high battery health class cluster is greater than the statistical value of the low battery health class cluster, it indicates that the characteristic value corresponding to the high battery health is higher, and the characteristic value corresponding to the low battery health is lower. Then the first size relationship includes that the first characteristic value is less than the reference characteristic value, and the second size relationship includes that the first characteristic value is greater than the reference characteristic value.

[0111] If the statistical value of the high battery health class cluster is less than the statistical value of the low battery health class cluster, it indicates that the characteristic value corresponding to the high battery health is lower, and the characteristic value corresponding to the low battery health is higher. Then the first size relationship includes that the first characteristic value is greater than the reference characteristic value, and the second size relationship includes that the first characteristic value is less than the reference characteristic value.

[0112] If the statistical value of the low battery health class cluster is greater than the reference characteristic value, and / or, the statistical value of the high battery health class cluster is less than the reference characteristic value, it indicates that the characteristic value corresponding to the low battery health is higher, and the characteristic value corresponding to the high battery health is lower. Then the first size relationship includes that the first characteristic value is greater than the reference characteristic value, and the second size relationship includes that the first characteristic value is less than the reference characteristic value.

[0113] It should be noted that the original battery data includes a plurality of parameter values at different historical time points in the first historical time period. When extracting the first characteristic value, a second historical time period can be selected from the first historical time period, and the extraction can be performed based on the parameter values in the second historical time period. The second historical time period can be equal to or less than the first historical time period. The second historical time periods corresponding to different key feature items can be different.

[0114] In some embodiments, in a case where it is determined that the key feature item corresponds to a working condition behavior that affects the battery health of the target device, the change in the feature value of the associated time sequence feature in time sequence can also be analyzed to determine the influence degree of the working condition behavior corresponding to the key feature item on the battery health. The associated time sequence feature includes third feature values of the associated feature item corresponding to different time units, and the associated feature item can be the key feature item itself or other feature items.

[0115] In some implementation scenarios, the associated feature item of the fast charging ratio can be the fast charging ratio or other feature items related to fast charging.

[0116] Analyzing the change in the feature value of the associated time sequence feature in time sequence can determine the change in the working condition behavior corresponding to the key feature item in time sequence, and the change in the working condition behavior is related to the influence degree on the battery health, so the change in the influence degree of the working condition behavior on the battery health can be determined.

[0117] The second historical time period can be divided into multiple time units, and the associated time sequence feature includes third feature values of the associated feature item corresponding to different time units. For each time unit, the third feature value of the time unit can be determined by using the parameter value of the battery parameter corresponding to the associated feature item in the time unit. The battery parameter corresponding to the associated feature item is the battery parameter used to obtain the feature item.

[0118] The length of the time unit can be determined as needed, for example, it can be one time, one day, one month, one year, etc.

[0119] It should be noted that the parameter value at different historical time points in the second historical time period itself is time series data, which can be directly used as time sequence features, or the foregoing parameter values can be processed in at least one time dimension to obtain time sequence features of different granularities. The lengths of the time units corresponding to different granularities are different. The sequence features to be used can be selected as needed.

[0120] In some implementation scenarios, the change in the feature value of the associated time sequence feature in time sequence can be analyzed by clustering all third feature values included in the associated time sequence feature.

[0121] The first feature value of the key feature item is a parameter value of the battery parameter corresponding to the key feature item in the second historical time period, and the third feature value of a time unit is obtained based on the parameter value of the battery parameter corresponding to the associated feature item in the time unit.

[0122] In some embodiments, the associated time sequence feature of the key feature item includes third feature values corresponding to different time units respectively; the feature value change of the associated time sequence feature of the key feature item in time sequence is analyzed to determine the influence degree of the working condition behavior corresponding to the key feature item on the battery health degree, including: clustering the associated time sequence feature of the key feature item to determine the frequency change of the working condition behavior corresponding to the key feature item in a target time period in the second historical time period relative to other time periods; based on the frequency change, it is determined that the working condition behavior corresponding to the key feature item in the target time period increases or decreases the influence on the battery health degree.

[0123] In some implementation scenarios, the key feature item can reflect the frequency of the working condition behavior, for example, the fast charging proportion can reflect the frequency of the fast charging behavior.

[0124] The clustering of the associated time sequence feature of the key feature item can determine whether there is a change in the time dimension of the associated time sequence feature, that is, whether there is a change in the time dimension of the working condition behavior corresponding to the key feature item. Further, the working condition behavior change time corresponding to the key feature item can be obtained, whether there is a change in the recent period (for example, within the last three months).

[0125] In some implementation scenarios, the associated time sequence feature of the key feature item is clustered, and if multiple clusters can be obtained, it indicates that there is a change in the time dimension of the corresponding working condition behavior. Each cluster can correspond to a time period, and each cluster can be used as a target time period to compare with other time periods to compare the feature value change of the target time period relative to other time periods, so as to determine the change of the influence degree of the corresponding working condition behavior on the battery health degree.

[0126] In some application scenarios, the fast charging proportion is taken as an example for illustration, and the corresponding associated time sequence feature can be the fast charging proportion of each month in the second historical time period. The fast charging proportions of each month in the second historical time period are clustered to obtain the feature value change. Further, it can be analyzed whether there is a change in the time dimension of the fast charging behavior in the second historical time period.

[0127] In some application scenarios, the clustering of the associated time sequence feature of a single vehicle can use a clustering method of distinguishable time sequence features represented by DBSCAN (Density-Based Spatial Clustering of Applications with Noise) to determine whether the time sequence distribution of the vehicle has clusters and outliers, so as to determine the recent behavior change of the vehicle.

[0128] In some implementation scenarios, the target device can be a vehicle provided with a battery.

[0129] In some embodiments, the method can further include providing the user with an influence evaluation result of the working condition behavior corresponding to the key feature item on the battery health.

[0130] In some embodiments, the method can further include providing the user with feature data corresponding to the key feature item. The feature data can be obtained by time-weighting the feature values of the key feature item. For example, a second historical time period is divided into two time periods, the feature values of the key feature item corresponding to the two time periods are obtained, the time lengths of the time periods are used as weights, and the feature values corresponding to the time periods are weighted and summed to obtain the feature data for display.

[0131] A certain time threshold can be set to divide the two time periods. For example, the time threshold can be set to a certain time length from the current time to divide the recent data and the long-term data.

[0132] In a specific application scenario, raw battery data reported by a vehicle-side BMS (battery management system) can be obtained from a cloud database, and the data can be cleaned of outliers and invalid values, and then divided into charging, discharging, and standing behavior period data.

[0133] In a specific application scenario, the key feature item is the fast charging ratio, and a clustering analysis is performed to obtain a division threshold y. If the fast charging ratio of a vehicle is higher than y, it is considered that the fast charging behavior of the vehicle affects the battery health.

[0134] In a specific application scenario, the key feature item is the fast charging ratio, and it is determined whether the associated time sequence feature of the fast charging ratio has obvious clustering. If there is clustering, it indicates that the fast charging behavior has changed in the time dimension. Further, it can be determined whether the fast charging ratio of the vehicle has increased or decreased in the recent period. If the fast charging ratio of the vehicle is 4% in the previous 8 months and 90% in the recent 4 months, it can be found that the fast charging behavior of the vehicle is more frequent in the recent four months, and the influence on the battery health is further expanded.

[0135] In a specific application scenario, if the fast charging ratio of the vehicle is 90% in the previous 8 months and 4% in the recent 4 months, the weighted fast charging ratio of the vehicle is 0.9*8 / 12+0.4*4 / 12 = 73.3% according to time weighting. The above data can be provided to the user for viewing, so that the user can understand the vehicle condition.

[0136] Referring to Figure 2 , Figure 2 is a schematic diagram of a framework of an equipment working condition behavior evaluation device provided by some embodiments of the present application.

[0137] The device working condition behavior evaluation apparatus 20 comprises an extraction module 21, a reference acquisition module 22 and a comparison module 23. The extraction module 21 is configured to extract a first feature value of a key feature item corresponding to each working condition behavior based on battery-related data of the target device under at least one working condition behavior, respectively, wherein the first feature value of the key feature item corresponding to the working condition behavior represents a parameter feature of the battery of the target device under the working condition behavior; the reference acquisition module 22 is configured to acquire a reference feature value corresponding to each key feature item, wherein the reference feature value is obtained by analyzing second feature values of the key feature items corresponding to different battery health degrees of a plurality of reference devices; and the comparison module 23 is configured to compare the first feature value of each key feature item with the corresponding reference feature value, respectively, to determine whether the working condition behavior corresponding to each key feature item affects the battery health degree of the target device.

[0138] Referring to Figure 3 , Figure 3 is a framework schematic diagram of an electronic device provided by some embodiments of the present application.

[0139] The electronic device 30 comprises a memory 31 and a processor 32, and the processor 32 is configured to execute program instructions stored in the memory 31 to implement any of the device working condition behavior evaluation methods described above. In a specific implementation scenario, the electronic device 30 can include but is not limited to a computer device, an electrical device, a microcomputer, a desktop computer, a server, and in addition, the electronic device 30 can also include a notebook computer, a tablet computer and other mobile devices, which are not limited herein.

[0140] Specifically, the processor 32 is configured to control itself and the memory 31 to implement any of the device working condition behavior evaluation methods described above. The processor 32 can also be referred to as a CPU (Central Processing Unit). The processor 32 can be an integrated circuit chip with processing capability. The processor 32 can also be a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. In addition, the processor 32 can be implemented by an integrated circuit chip together.

[0141] Referring to Figure 4 , Figure 4 is a framework schematic diagram of a computer-readable storage medium provided by some embodiments of the present application.

[0142] The computer readable storage medium 40 stores program instructions 41 capable of being run by the processor, the program instructions 41 being executed by the processor to implement the evaluation method of any of the device operating behaviors described above.

[0143] The above description of the various embodiments tends to emphasize differences between the various embodiments, and the same or similar parts can be referred to each other, and for brevity, will not be repeated here.

[0144] In several embodiments provided in the present application, it should be understood that the disclosed method and device can be implemented in other ways. For example, the above-described device implementation is only schematic; for example, the division of the modules or units is only a logical function division, and an actual implementation can have another division manner, for example, a unit or component can be combined or integrated into another subsystem, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed mutual elements can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0145] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that makes a contribution to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program codes that can be stored in the medium.

Claims

1. A method of evaluating equipment operating behavior, characterized by, The method comprises: extracting first feature values of key feature items corresponding to each of the working conditions based on battery-related data of the target device under each of the working conditions, wherein the battery-related data under the working conditions comprises battery time series data of a plurality of battery parameters under the working conditions; the battery time series data of the battery parameters comprises parameter values of the battery parameters at a plurality of historical time points when the working conditions occur; the key feature items are feature items associated with battery health, and the first feature values of the key feature items corresponding to the working conditions represent parameter characteristics of the battery of the target device under the working conditions; and obtaining reference feature values corresponding to each of the key feature items, which are obtained by analyzing second feature values of the key feature items corresponding to a plurality of reference devices with different battery health; comparing the first feature values of each of the key feature items with the corresponding reference feature values to determine whether the working conditions corresponding to each of the key feature items affect the battery health of the target device; The extracting of the first feature values of the key feature items corresponding to each of the working conditions based on the battery-related data of the target device under each of the working conditions comprises: For each of the working conditions, the battery time series data of the plurality of battery parameters is divided into at least one set of time series data corresponding to the working condition, and each set of the time series data comprises battery time series data of at least one of the battery parameters; For each of the time series data sets corresponding to the working condition, at least one statistic is performed on the time series data set under the working condition to obtain the first feature values of at least one key feature item corresponding to the working condition.

2. The method of claim 1, wherein, The first feature values of the key feature items comprise at least one of a time representation feature value and a parameter value representation feature value corresponding to the time series data set, the time representation feature value is obtained by performing time statistics on the battery time series data in the time series data set, and the parameter value representation feature value is obtained by performing parameter value statistics on the battery time series data in the time series data set.

3. The method of claim 1, wherein, Before the extracting of the first feature values of the key feature items corresponding to each of the working conditions based on the battery-related data of the target device under each of the working conditions, the method further comprises: obtaining original battery data collected by the battery of the target device, wherein the original battery data comprises a plurality of parameter values of a plurality of battery parameters corresponding to different historical time points within a first historical time period; dividing the original battery data based on the working conditions of the target device within the first historical time period to obtain battery-related data under each of the working conditions, wherein the battery-related data under the working conditions comprises parameter values of each of the battery parameters at a plurality of historical time points when the working conditions occur; and and / or, before the extracting of the first feature values of the key feature items corresponding to each of the working conditions based on the battery-related data of the target device under each of the working conditions, the method further comprises: constructing a battery feature item set, wherein the battery feature item set comprises a plurality of battery feature items; analyzing a relationship between each battery feature item in the battery feature item set and a battery health degree; selecting, based on the relationship, the battery feature item related to the battery health degree from the battery feature item set as the key feature item.

4. The method of claim 1, wherein, The acquiring of the reference feature value corresponding to each of the key feature items comprises: acquiring second feature values of each of the key feature items corresponding to a plurality of reference devices of different battery health degrees; based on the battery health degree, clustering the second feature values of each of the key feature items corresponding to the plurality of reference devices to obtain a first feature cluster and a second feature cluster, wherein the first feature cluster comprises the second feature values of each of the key feature items corresponding to the first reference device, the second feature cluster comprises the second feature values of each of the key feature items corresponding to the second reference device, the first reference device is each of the reference devices belonging to a first battery health degree level, and the second reference device is each of the reference devices belonging to a second battery health degree level; for each of the key feature items, counting the second feature values of the key feature item corresponding to each of the first reference devices in the first feature cluster to obtain a first statistical value, and counting the second feature values of the key feature item corresponding to each of the second reference devices in the second feature cluster to obtain a second statistical value; based on the first statistical value and the second statistical value, obtaining the reference feature value corresponding to the key feature item.

5. The method of claim 4, wherein, The first statistical value is the cluster center of the second feature values corresponding to the key feature item in the first feature cluster, and the second statistical value is the cluster center of the second feature values corresponding to the key feature item in the second feature cluster; and / or, the obtaining of the reference feature value corresponding to the key feature item based on the first statistical value and the second statistical value comprises: calculating the distance between the first statistical value and the second statistical value as the reference feature value corresponding to the key feature item.

6. The method of claim 1, wherein, The comparing of the first feature value of each of the key feature items with the corresponding reference feature value to determine whether the working condition behavior corresponding to each of the key feature items affects the battery health degree of the target device comprises: in response to the first feature value of the key feature item and the corresponding reference feature value satisfying a first size relationship, determining that the working condition behavior corresponding to the key feature item affects the battery health degree of the target device; in response to the first feature value of the key feature item and the corresponding reference feature value satisfying a second size relationship, determining that the working condition behavior corresponding to the key feature item does not affect the battery health degree of the target device.

7. The method of claim 1, wherein, The first feature value of the key feature item is determined by the parameter value of the battery parameter corresponding to the key feature item in a second historical time period; after the comparing of the first feature value of each of the key feature items with the corresponding reference feature value to determine whether the working condition behavior corresponding to each of the key feature items affects the battery health degree of the target device, the method further comprises: In response to determining that the working condition behavior corresponding to the key feature item affects the battery health of the target device, a change in a feature value of a related time sequence feature of the key feature item in time sequence is analyzed to determine an influence degree of the working condition behavior corresponding to the key feature item on the battery health, wherein the related time sequence feature of the key feature item includes third feature values of the related feature items corresponding to different time units, the related feature items are the key feature item or other feature items, the different time units are located in the second historical time period, and the third feature value of the time unit is determined based on a parameter value of a battery parameter corresponding to the related feature item in the time unit.

8. The method of claim 7, wherein, The first feature value of the key feature item is a parameter value of a battery parameter corresponding to the key feature item in a second historical time period; and the third feature value of the time unit is a parameter value of a battery parameter corresponding to the related feature item in the time unit. And / or, the analysis of the change in the feature value of the related time sequence feature of the key feature item in time sequence to determine the influence degree of the working condition behavior corresponding to the key feature item on the battery health includes: Clustering the related time sequence feature of the key feature item to determine a frequency change of the working condition behavior corresponding to the key feature item in a target time period in the second historical time period relative to other time periods; Based on the frequency change, determining that the working condition behavior corresponding to the key feature item in the target time period increases or decreases the influence on the battery health.

9. The method of claim 1, wherein, After the comparison of the first feature value of each key feature item with the corresponding reference feature value to determine whether the working condition behavior corresponding to each key feature item affects the battery health of the target device, the method further includes: providing the user with an influence evaluation result of the working condition behavior corresponding to the key feature item on the battery health; and / or, providing the user with feature data corresponding to the key feature item.

10. The method of claim 1, wherein, The working condition behavior includes at least one of charging, discharging, and standing; And / or, the battery-related data under the working condition behavior includes at least one parameter value of a plurality of battery parameters when the working condition behavior occurs, and the plurality of battery parameters include at least one of current, voltage, state of charge, and temperature. And / or, the target device is a vehicle.

11. An apparatus for evaluating the behavior of a device in operation, characterized by The method includes: The extraction module is configured to extract a first feature value of a key feature item corresponding to each of the at least one working condition behavior based on battery-related data of the target device under the at least one working condition behavior, respectively. The battery-related data under the working condition behavior includes battery time series data of a plurality of battery parameters under the working condition behavior, respectively. The battery time series data includes parameter values of the battery parameters at a plurality of historical time points when the working condition behavior occurs. The key feature item is a feature item associated with the battery health degree. The first feature value of the key feature item corresponding to the working condition behavior represents a parameter feature of the battery of the target device under the working condition behavior. The extraction of the first feature value of the key feature item corresponding to each of the at least one working condition behavior based on the battery-related data of the target device under the at least one working condition behavior, respectively, includes: for each of the at least one working condition behavior, dividing the battery time series data of the plurality of battery parameters into at least one time series data group corresponding to the working condition behavior, each of the time series data groups including battery time series data of at least one of the battery parameters; and for each of the time series data groups corresponding to the working condition behavior, performing at least one statistical operation on the time series data group under the working condition behavior to obtain the first feature value of at least one key feature item corresponding to the working condition behavior; and The reference acquisition module is configured to acquire a reference feature value corresponding to each of the key feature items, the reference feature value being obtained by analyzing second feature values of the key feature items corresponding to a plurality of reference devices with different battery health degrees. The comparison module is configured to compare the first feature value of each of the key feature items with the corresponding reference feature value, respectively, to determine whether the working condition behavior corresponding to each of the key feature items affects the battery health degree of the target device.

12. An electronic device, comprising: The memory and the processor, the memory has a program instruction stored thereon, and the program instruction is executed by the processor to execute the method in any one of claims 1 to 10.

13. A computer readable storage medium having stored thereon program instructions, wherein, The program instruction is executed by the processor to implement the method in any one of claims 1 to 10.

Images