CN117761543A

CN117761543A - Multi-sparse observer fusion power battery abnormal voltage identification method and system

Info

Publication number: CN117761543A
Application number: CN202311832161.2A
Authority: CN
Inventors: 马建; 张昭; 赵轩; 马宇骋; 龚贤武; 相里康
Original assignee: Changan University
Current assignee: Changan University
Priority date: 2023-12-27
Filing date: 2023-12-27
Publication date: 2024-03-26

Abstract

The method comprises the steps of extracting single battery voltage data from historical data uploaded by an electric automobile and preprocessing the single battery voltage data; determining the length of a time window based on the preprocessed battery cell voltage data, and constructing a voltage fluctuation characteristic quantity; carrying out mode division on the voltage fluctuation characteristic quantity, and respectively constructing battery fluctuation characteristic quantity data sets in different modes; performing sparse processing on the voltage fluctuation characteristic quantity data sets in different modes, constructing a sparse observer set, performing condition judgment, and performing sparse observer fusion if the conditions are met to determine an abnormal voltage threshold; and carrying out abnormal voltage identification on the real-time data of the vehicle based on the abnormal voltage threshold. According to the invention, based on the voltage fluctuation state of the vehicle in different modes, the threshold value is determined for the abnormal voltage of the power battery by adopting a multi-sparse observer fusion method, so that early warning can be carried out on the vehicle with thermal runaway and potential fault risks, and the driving safety is improved.

Description

Multi-sparse observer fusion power battery abnormal voltage identification method and system

Technical Field

The invention belongs to the technical field of power battery safety, and particularly relates to a method and a system for identifying abnormal voltage of a fusion power battery by using a multi-sparse observer.

Background

With the continuous increase of global energy shortage, the popularity of electric vehicles is continuously increasing, and the electric vehicles are currently recognized as the key direction of future development of the automobile industry. Therefore, research on safety of the in-vehicle battery system is receiving increasing attention from industry and academia. The voltage is used as an external characterization parameter of the power battery and is an important index for evaluating the safety of a battery system. Numerous studies have shown that abnormal voltage fluctuations may indicate that the safety level of the battery system is dropping, and if left uncontrolled, may eventually lead to voltage failure or even thermal runaway. Therefore, accurate identification of voltage abnormality fluctuations is extremely important for early detection of battery system failures and for ensuring safe operation of an operating vehicle.

The abnormal voltage identification method based on the threshold value is widely applied to real-time safety monitoring of large-scale vehicles due to the advantage of short bicycle identification time. The determination of the anomaly threshold in the current study is often based on experimental results obtained for one or a few vehicles and remains constant throughout. However, different vehicles have different "abnormal voltage ranges" due to the large differences in battery specifications, driver driving habits, and operating environments. The constant abnormal voltage threshold for abnormality detection is difficult to popularize for large-scale vehicle monitoring. Therefore, the adaptive abnormal judgment threshold value is set according to different vehicle working conditions, and the method has important significance for quickly and accurately identifying abnormal voltage.

Disclosure of Invention

The invention aims to solve the problems in the prior art and provide a method and a system for identifying abnormal voltage of a power battery by fusing multiple sparse observers.

In order to achieve the above purpose, the present invention has the following technical scheme:

in a first aspect, a method for identifying abnormal voltages of a fusion power battery of a multi-sparse observer is provided, including:

extracting single battery voltage data from historical data uploaded by an electric automobile and preprocessing the single battery voltage data;

determining the length of a time window based on the preprocessed battery cell voltage data, and constructing a voltage fluctuation characteristic quantity;

carrying out mode division on the voltage fluctuation characteristic quantity, and respectively constructing battery fluctuation characteristic quantity data sets in different modes;

performing sparse processing on the voltage fluctuation characteristic quantity data sets in different modes, constructing a sparse observer set, performing condition judgment, and performing sparse observer fusion if the conditions are met to determine an abnormal voltage threshold;

and carrying out abnormal voltage identification on the real-time data of the vehicle based on the abnormal voltage threshold.

As a preferred scheme, according to the historical data of the electric automobile uploaded by national standard GB/T32960 technical specification of remote service and management system of electric automobile, the voltage data of the battery cell is extracted and preprocessed, and the preprocessing step comprises:

for time disorder and data repetition in sample data, sorting the data according to time and performing redundancy removal operation;

if the single battery voltage value list has a null value, the single battery voltage value list is regarded as invalid frame data;

if the highest monomer voltage value is greater than 6V, the frame data is determined to be invalid;

if the lowest cell voltage value is less than 0V, invalid frame data is determined.

As a preferable mode, the voltage fluctuation characteristic quantity is the voltage fluctuation condition of a single battery cell in a specific time period, and the calculation expression is as follows:

wherein x is _i Represents the voltage value at i time in the observation window, mu represents the average of the voltage values at all time in the observation window, n represents the number of samples in the observation window, std _i Representing the standard deviation of the voltage of the ith battery cell fixed time window;

the standard deviation is calculated after eliminating the monomer which exceeds the standard deviation of the battery cell voltage by more than 10 times of the median:

in Std _median Representing the median, std, of standard deviations of all cell voltage fixed time windows _std Representing standard deviation of fixed time window of all single batteries after being processed by Pre-standard method, std_standard _i The abnormal voltage fluctuation score of the ith battery cell after the treatment by the Pre-standard method is shown.

As a preferred embodiment, the step of performing mode division on the voltage fluctuation feature quantity includes: taking the median representation of the fluctuation characteristic quantity of all the battery monomers at the same time to represent the general fluctuation trend of the power battery pack at the corresponding time, and calling the general fluctuation trend as a safety mode; and taking the maximum value of all the battery monomer fluctuation characteristic quantities at the same time to represent the limit fluctuation trend of the power battery pack at the corresponding time, and calling the limit mode.

As a preferable scheme, the step of performing sparse processing on the voltage fluctuation feature quantity data sets in different modes to construct a sparse observer set includes:

for a battery fluctuation feature quantity set M in different modes, each object in the set is an observer; randomly extracting a plurality of objects from the battery fluctuation characteristic quantity set M to form an observer set, obtaining the number of observers required for specifying a confidence interval and a sampling error according to a statistical sampling theory, and calculating the following expression:

wherein k is the number of required observers, M is the number of objects in the battery fluctuation feature quantity set M, sigma is the standard deviation of the battery fluctuation feature quantity set M, epsilon is the limit error of non-repeated sampling, and Z is the confidence coefficient;

calculating Euclidean distance D of each observer and each object in the battery fluctuation feature quantity set M _i,j Constructing a distance matrix D, i.e. d= { D _i,j I e (1, 2,., m), j e (1, 2,., k); constructing an observation matrix P for storing indexes of x observers nearest to each object in the data set;

counting all index values in the observation matrix P, and creating a matrix U to represent the occurrence times of each observer, namely P = { P _j J e (1, 2,., k); setting an inertia threshold q according to the matrix U, and calculating the value of q according to the following formula:

q＝Q _ρ (U)

wherein Q is _ρ () is a quantile function;

when the number of times of occurrence of the observer is smaller than q, the corresponding observer is identified as an inert observer, and the inert observer is deleted from the observer set; residual k after removal of the inert observer _act The observer becomes an active observer and is used for representing a low-density model of the distribution characteristic of the training data set.

As a preferable scheme, the step of performing the condition judgment, and performing the sparse observer fusion if the condition is satisfied, and determining the abnormal voltage threshold includes:

the method comprises the steps of obtaining a safe mode observer set Saf_sample and a limit mode observer set Ext_sample, and representing the fluctuation characteristic of the power battery by using a fluctuation coefficient fluc_coeffcient, wherein the calculation expression is as follows:

fluc_coefficient＝min(Ext_sample)-max(Saf_sample)

taking fluc_threshold as a threshold value for determining the abnormality of the vehicle voltage, the calculation expression is as follows:

when the fluctuation feature quantity is larger than the fluc_threshold, the abnormal voltage is determined;

when sampling is performed, the fluc_threshold is ensured to be larger than the minimum value of safe_sample, if the condition is met, two observer sets are fused for detecting abnormal points of voltage fluctuation, and if the condition is not met, the sampling is required to be performed again.

As a preferable mode, the step of identifying the abnormal voltage based on the abnormal voltage threshold value includes: when judging the abnormality degree of the new object o, calculating the Euclidean distance between the new object o and each observer in the low-density model to form a model with the length of k _act The index values of x observers closest to the distance matrix N are recorded, and an observation array P is constructed; at the same time, the average distance between the new object o and the x observers closest to the new object o is calculated to represent the degree y of abnormality _o ：

Where M (-) is a median function and d (,) is the Euclidean distance.

In a second aspect, a multi-sparse observer fusion power battery abnormal voltage identification system is provided, including:

the battery cell voltage data extraction module is used for extracting and preprocessing battery cell voltage data from historical data uploaded by the electric automobile;

the voltage fluctuation characteristic quantity construction module is used for determining the length of a time window based on the preprocessed battery cell voltage data and constructing a voltage fluctuation characteristic quantity;

the battery fluctuation characteristic quantity data set dividing module is used for carrying out mode division on the voltage fluctuation characteristic quantity and respectively constructing battery fluctuation characteristic quantity data sets in different modes;

the abnormal voltage threshold determining module is used for carrying out sparse processing on the voltage fluctuation characteristic quantity data sets in different modes, constructing a sparse observer set, judging conditions, carrying out sparse observer fusion if the conditions are met, and determining an abnormal voltage threshold;

and the abnormal voltage identification module is used for identifying the abnormal voltage of the real-time data of the vehicle based on the abnormal voltage threshold value.

In a third aspect, there is provided an electronic device comprising:

a memory storing at least one instruction; and the processor executes the instructions stored in the memory to realize the multi-sparse observer fusion power battery abnormal voltage identification method.

In a fourth aspect, a computer readable storage medium is provided, where at least one instruction is stored, where the at least one instruction is executed by a processor in an electronic device to implement the multi-sparse observer fusion power cell abnormal voltage identification method.

Compared with the prior art, the invention has at least the following beneficial effects:

for an electric automobile in operation, different vehicles have differences in abnormal voltage definition ranges due to differences in battery types, driving habits of drivers, operation environments and the like. According to the invention, the normal fluctuation range of the battery is represented by using a sparse representation method, the abnormal voltage is identified by adopting multi-sparse observer fusion, and early warning can be carried out on a vehicle with thermal runaway and potential fault risks, so that the driving safety is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of an overall method for identifying abnormal voltages of a multi-sparse observer fusion power battery in an embodiment of the invention;

FIG. 2 is a graph illustrating voltage across each cell in a first set of vehicle cell data according to an embodiment of the present invention;

FIG. 3 is a Std_Standard graph of each battery cell in a first set of vehicle battery cell data according to an embodiment of the present invention;

FIG. 4 is a graph of anomaly coefficients for each battery cell in a first set of vehicle battery cell data according to an embodiment of the present invention;

FIG. 5 is a graph illustrating voltage across each cell in a second set of vehicle cell data according to an embodiment of the present invention;

fig. 6 is a graph showing anomaly coefficients of each battery cell in a second set of vehicle battery cell data according to an embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings.

Referring to fig. 1, the method for identifying abnormal voltages of a fusion power battery with a multi-sparse observer provided by the embodiment of the invention comprises the following steps:

(1) According to national standard GB/T32960 technical Specification of remote service and management System for electric vehicles, uploading historical data of electric vehicles, extracting voltage data of battery monomers and preprocessing;

(2) Determining the length of a time window based on the voltage data of the battery cells, and constructing a voltage fluctuation characteristic quantity;

(3) Carrying out mode division on the voltage fluctuation characteristic quantity, and respectively constructing battery fluctuation characteristic quantity data sets in different modes;

(4) Performing sparse processing on the voltage fluctuation characteristic quantity data sets in different modes, constructing a sparse observer set, performing condition judgment, and performing sparse observer fusion and determining an abnormal voltage threshold value if the conditions are met;

(5) And carrying out abnormal voltage identification on the real-time data of the vehicle based on the abnormal voltage threshold.

In one possible implementation manner, in the step (1), the data uploaded by the electric automobile in the national standard GB/T32960 technical Specification of remote service and management System of electric automobile is required to be preprocessed and analyzed, and the preprocessing principle is as follows:

1) Aiming at the problems of time disorder and data repetition in sample data, the data are ordered according to time and redundancy removing operation is carried out;

2) If the list of voltage values of the single battery has a null value, the frame data is not valid in all voltage values, so that the frame data is considered as invalid frame data;

3) If the highest monomer voltage value is greater than 6V, the frame data is determined to be invalid;

4) If the lowest cell voltage value is less than 0V, invalid frame data is determined.

In one possible embodiment, step (2) selects the standard deviation as the characteristic quantity for characterizing the voltage fluctuation of the single battery cell, so that in order to characterize the voltage fluctuation condition of the single battery cell in a specific time period, a viewing window technology is adopted. The calculation formula is as follows:

wherein x is _i Represents the voltage value at i time in the observation window, mu represents the average of the voltage values at all time in the observation window, n represents the number of samples in the observation window, std _i The standard deviation of the voltage of the i-th cell fixed time window is represented.

To characterize the lateral comparison between different cells over the same time period, the Pre-standard method was proposed. The Pre-standard method is improved based on the Z-score standardization method for early voltage fluctuation characteristics of the power battery at the time of failure. Whether the battery pack fluctuates slightly at an early stage or the battery cells change instantaneously at a Pre-failure stage, the abnormal battery cells often occur in a small number of the plurality of battery cells, so that in the Pre-standard method, the average value is replaced by the median of the standard deviation of the voltage of the battery cells to represent the normal state of the battery pack. Similarly, when calculating the standard deviation, the analysis of a plurality of thermal runaway fault vehicles shows that the standard deviation of abnormal monomers in the early stage of the fault is often more than 10 times of the standard deviation of normal state, so that the standard deviation is calculated after eliminating the monomers which are more than 10 times of the median of the standard deviation of the voltage of the battery, thereby ensuring that the monomers are not influenced by the abrupt change of the voltage in the early stage of the fault, and the calculation formula is as follows:

wherein Std is _median Representing the median, std, of standard deviations of all cell voltage fixed time windows _std The standard deviation of all the unit cells after the treatment by the Pre-standard method is shown. Std_Standard _i The abnormal voltage fluctuation score of the ith battery cell after the treatment by the Pre-standard method is shown.

In one possible implementation, step (3) proposes two modes of characterizing the range of fluctuation of the power battery voltage in a normal state, which can be used for sparse observers, based on the operating characteristics of the power battery. For power battery packs, small abnormal fluctuations at the early stages of failure are often manifested in a single or a small number of cells, and therefore, for the vast majority of cells, they can still be guaranteed to operate in a normal state in the presence of abnormal fluctuations of cells. The median of all the single battery fluctuation characteristic quantities at the same time is taken to represent the general fluctuation trend of the power battery pack at the same time, and the power battery pack is called a safety mode; and taking the maximum value of all the single battery fluctuation characteristic quantities at the same time to represent the limit fluctuation trend of the power battery pack at the time, and the limit mode is called. And independently storing the battery fluctuation characteristic quantity sets in different modes, and constructing battery fluctuation characteristic quantity data sets in different modes.

In one possible implementation manner, in the step (4), an abnormal voltage threshold determination method based on multi-sparse observer fusion is provided. The observer is first initialized and for a set of battery fluctuation feature amounts M in different modes, each object in the set may be referred to as an observer. Randomly extracting a plurality of objects in M to form an observer set, and obtaining the number of observers required for specifying a confidence interval and a sampling error according to a statistical sampling theory, wherein the calculation formula is as follows:

where k is the number of observers needed, M is the number of objects in M, σ is the standard deviation of M, ε is the limit error of non-resampling, ε=0.1σ is the confidence coefficient, and Z is the 95% confidence interval, so z=1.96.

Secondly, constructing an observation matrix, and calculating Euclidean distance D of each observer and each object in M _i,j Constructing a distance matrix D, i.e. d= { D _i,j I e (1, 2,., m), j e (1, 2,..k) }. The observation matrix P is constructed to store an index of the x most recent observers from each object in the dataset. x is generally 3-10, and does not cause large deviation in the identification of outliers, but it is related to the value of k and scene characteristics. Therefore, the comprehensive consideration should be taken into account in the selection process, so that the problems of model oversimplification and overfitting are avoided.

The inert observer is then removed, all index values in the observation matrix are counted, and the creation matrix U represents the number of occurrences of each observer, i.e. p= { P _j J e (1, 2,..k) }. The inertia threshold q is set according to the matrix U, and the calculation formula is as follows:

q＝Q _ρ (U)

wherein Q is _ρ (-) is a fractional number function according to the warpThe effect is better when the test ρ is taken to be 0.3.

When the number of observer occurrences is less than q, an inactive observer is identified and deleted from the observer set. The screening can ensure that the rest observers can represent the middle-high density area of the training data, and avoid the interference of abnormal points on the identification effect. Residual k after removal of the inert observer _act The observers become active observers, i.e. low density models that can be used to characterize the distribution characteristics of the training dataset.

And finally, respectively carrying out the operation on the fluctuation characteristic quantities under the two modes to obtain a safe mode observer set Saf_sample and a limit mode observer set Ext_sample. The fluctuation coefficient fluc_coeffient is used for representing the fluctuation characteristic of the power battery, and the calculation formula is as follows:

fluc_coefficient＝min(Ext_sample)-max(Saf_sample)

according to the formula, different vehicles and different training samples can generate different influences on the fluctuation coefficient, so that the uniqueness of voltage fluctuation of different vehicles under the influence of multiple factors is just reflected, and for the same vehicle, the fluctuation coefficient is supposed to show corresponding change along with the gradual aggravation of the aging degree of different service time of the battery. The larger the fluctuation coefficient is, the more severe the fluctuation of the battery in the environment is compared with other batteries, and the voltage fluctuation range in the normal running state is larger, so the judgment standard of abnormal voltage is also improved.

Taking fluc_threshold as a threshold value for judging the abnormal voltage of the vehicle, the calculation formula is as follows:

when the fluctuation feature amount is larger than fluc_threshold, it is determined as an abnormal voltage. Considering the distribution characteristics of the two modes and the systematic errors existing in the sparse representation, the problem of unbalanced sample distribution is avoided, the fact that the fluc_threshold is larger than the minimum value of safe_sample is ensured in the sampling process, the two observer sets are fused for detecting abnormal points of voltage fluctuation if the two observer sets are satisfied, and the sampling process is required to be carried out again if the two observer sets are not satisfied.

In a possible implementation manner, in step (5), when the new object o enters the device and needs to determine the abnormality degree, the Euclidean distance between the new object o and each observer in the low-density model should be calculated to form a length k _act The index values of the x observers nearest to the matrix N are recorded, and an observation array P is constructed. At the same time, the average distance between the new object o and the x observers nearest to the new object o is calculated and can be expressed as the degree of abnormality y _o The calculation formula is as follows:

wherein M (-) is a median function and d (,) is the Euclidean distance.

The multi-sparse observer fusion power battery abnormal voltage identification method surrounds the identification of the abnormal voltage of the power battery of the electric automobile, on the basis of real automobile operation big data uploaded by the electric automobile, the multi-sparse observer fusion power battery abnormal voltage self-adaptive identification method is provided aiming at the fluctuation range difference of the voltage caused by the difference of battery types, driving habits of drivers, operation environments and the like of different vehicles, the power battery voltage fluctuation is represented by adopting the characteristic scale based on the variance and Pre-standard method, the normal fluctuation range of the battery is represented by adopting the sparse representation method, the abnormal voltage is identified by adopting the multi-sparse observer fusion, and early warning can be carried out on vehicles with thermal runaway and potential fault risks, so that the driving safety is improved.

In order to verify the early warning effect of the method of the invention on a thermal runaway and a vehicle at risk of potential failure, the operational data of two actual operating vehicles are used for verification. The first set of vehicle battery cell data used for training with data from day 11 of 2019, 8, to day 16 of 2019, and data from day 29 of 2019 for testing. Fig. 2 shows the voltage change condition of each single battery in the test time segment, and as can be found from fig. 2, the number 28 single battery has changed the maximum and minimum voltage values for several times in a short time, and the voltage fluctuation condition is more severe than the rest single batteries, which clearly belongs to the abnormal fluctuation of the voltage. Fig. 3 shows the variation of the fluctuation feature std_standard, and it can be seen from fig. 3 that std_standard at this time is significantly higher than other times, so std_standard accurately characterizes this voltage anomaly. The fluctuation characteristic quantity of each battery cell is put into a model for testing, so as to obtain a cell voltage abnormality coefficient fluctuation diagram shown in fig. 4, wherein the diagram shows whether the voltage fluctuation at the current moment is abnormal compared with the vehicle history data, and the abnormality point is determined as the abnormality point because the abnormality coefficient of the voltage fluctuation at the moment is higher than an abnormality threshold value as can be seen from the diagram.

The second set of vehicle battery cell data uses data from 13/2020 to 19/2020 for training and data from 3/2020 for testing, with thermal runaway failure occurring in the vehicle at 2/3. Fig. 5 shows the voltage change of each cell in the test time period, and it can be seen from the graph that before 325 frames of data, the vehicle is in a charged state and is nearly full, the voltage of each cell is continuously increased and the voltage fluctuation always shows better consistency, at 325 frames of data, the voltage suddenly drops in the 23-27 cells, the voltage suddenly rises in the 28 cells to a minimum of 0.25V, the voltage reaches 4.998V, and the change is continued until the thermal runaway fault occurs. The fluctuation characteristic quantity of each battery monomer is put into a model for testing, so that a monomer voltage anomaly coefficient fluctuation diagram shown in fig. 6 is obtained, and at the moment that a fault monomer is suddenly changed, the No. 23-28 monomers are higher than an anomaly threshold value determined according to vehicle historical data, so that the method provided by the invention can timely identify voltage anomaly fluctuation, and plays a role in predicting faults to a certain extent.

For an electric automobile in operation, different vehicles have differences of abnormal voltage definition ranges due to differences of battery types, driving habits of drivers, operation environments and the like, so that the method is based on a multi-sparse observer fusion method, a sparse representation method is adopted to represent a normal fluctuation range of voltage, an adaptive threshold is generated, and voltage moments exceeding the threshold are identified and positioned. In addition, the method greatly reduces the calculated amount in the aspects of feature quantity construction, sparse representation, mode division and the like, ensures that the method provided by the invention is more suitable for monitoring real-vehicle online data, ensures the instantaneity of an abnormal voltage identification method, and provides theoretical support for real-time safety risk monitoring of running vehicles.

Another embodiment of the present invention further provides a system for identifying abnormal voltages of a fusion power battery with a multi-sparse observer, including:

Another embodiment of the present invention also proposes an electronic device including:

Another embodiment of the present invention also proposes a computer-readable storage medium having stored therein at least one instruction that is executed by a processor in an electronic device to implement the multi-sparse observer fusion power battery abnormal voltage identification method.

For example, the instructions stored in the memory may be partitioned into one or more modules/units that are stored in a computer-readable storage medium and executed by the processor to perform the multi-sparse observer fusion power cell abnormal voltage identification method of the present invention. The one or more modules/units may be a series of computer readable instruction segments capable of performing a specified function, which describes the execution of the computer program in a server.

The electronic equipment can be a smart phone, a notebook computer, a palm computer, a cloud server and other computing equipment. The electronic device may include, but is not limited to, a processor, a memory. Those skilled in the art will appreciate that the electronic device may also include more or fewer components, or may combine certain components, or different components, e.g., the electronic device may also include input and output devices, network access devices, buses, etc.

The processor may be a central processing unit (CentraL Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (DigitaL SignaL Processor, DSP), application specific integrated circuits (AppLication Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (fierld-ProgrammabLe Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory may be an internal storage unit of the server, such as a hard disk or a memory of the server. The memory may also be an external storage device of the server, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure DigitaL (SD) Card, a FLash Card (FLash Card) or the like, which are provided on the server. Further, the memory may also include both an internal storage unit and an external storage device of the server. The memory is used to store the computer readable instructions and other programs and data required by the server. The memory may also be used to temporarily store data that has been output or is to be output.

It should be noted that, because the content of information interaction and execution process between the above module units is based on the same concept as the method embodiment, specific functions and technical effects thereof may be referred to in the method embodiment section, and details thereof are not repeated herein.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application implements all or part of the flow of the method of the above embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, where the computer program, when executed by a processor, may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing device/terminal apparatus, recording medium, computer Memory, read-Only Memory (ROM), random access Memory (RAM, random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. Such as a U-disk, removable hard disk, magnetic or optical disk, etc.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.

The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.

Claims

1. The method for identifying the abnormal voltage of the fusion power battery of the multi-sparse observer is characterized by comprising the following steps of:

2. The method for identifying abnormal voltages of a multi-sparse observer fusion power battery according to claim 1, wherein the steps of extracting and preprocessing voltage data of a battery cell according to historical data uploaded by an electric vehicle in national standard GB/T32960 technical specification of an electric vehicle remote service and management system include:

3. The method for identifying abnormal voltages of a fusion power battery with a plurality of sparse observers according to claim 1, wherein the voltage fluctuation feature is a voltage fluctuation condition of a single battery cell in a specific time period, and the calculation expression is as follows:

4. The method for identifying abnormal voltages of a hybrid power battery with multiple sparse observers according to claim 1, wherein the step of pattern-dividing the voltage fluctuation feature quantity comprises: taking the median representation of the fluctuation characteristic quantity of all the battery monomers at the same time to represent the general fluctuation trend of the power battery pack at the corresponding time, and calling the general fluctuation trend as a safety mode; and taking the maximum value of all the battery monomer fluctuation characteristic quantities at the same time to represent the limit fluctuation trend of the power battery pack at the corresponding time, and calling the limit mode.

5. The method for identifying abnormal voltages of a hybrid power battery with multiple sparse observers according to claim 4, wherein the step of sparsely processing the voltage fluctuation feature data sets in different modes to construct a sparse observer set comprises:

calculating each object in each observer and battery fluctuation feature quantity set MEuclidean distance D of (2) _i,j Constructing a distance matrix D, i.e. d= { D _i,j I e (1, 2,., m), j e (1, 2,., k); constructing an observation matrix P for storing indexes of x observers nearest to each object in the data set;

q＝Q _ρ (U)

wherein Q is _ρ () is a quantile function;

6. The method for identifying abnormal voltages of a multi-sparse observer fusion power battery according to claim 5, wherein the step of determining the abnormal voltage threshold value includes the steps of:

fluc_coefficient＝min(Ext_sample)-max(Saf_sample)

7. The method for identifying abnormal voltages of a hybrid power battery with multiple sparse observers according to claim 6, wherein the step of identifying abnormal voltages of the real-time data of the vehicle based on the abnormal voltage threshold value comprises:

when judging the abnormality degree of the new object o, calculating the Euclidean distance between the new object o and each observer in the low-density model to form a model with the length of k _act The index values of x observers closest to the distance matrix N are recorded, and an observation array P is constructed; at the same time, the average distance between the new object o and the x observers closest to the new object o is calculated to represent the degree y of abnormality _o ：

Where M (-) is a median function and d (,) is the Euclidean distance.

8. The utility model provides a many sparse observers fuses power battery abnormal voltage identification system which characterized in that includes:

9. An electronic device, comprising:

a memory storing at least one instruction; and

A processor executing instructions stored in the memory to implement the multi-sparse observer fusion power cell abnormal voltage identification method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized by: the computer-readable storage medium has stored therein at least one instruction that is executed by a processor in an electronic device to implement the multi-sparse observer fusion power cell abnormal voltage identification method of any one of claims 1 to 7.