CN114919413A - Method and system for diagnosing battery abnormity of electric vehicle - Google Patents

Method and system for diagnosing battery abnormity of electric vehicle Download PDF

Info

Publication number
CN114919413A
CN114919413A CN202210579978.2A CN202210579978A CN114919413A CN 114919413 A CN114919413 A CN 114919413A CN 202210579978 A CN202210579978 A CN 202210579978A CN 114919413 A CN114919413 A CN 114919413A
Authority
CN
China
Prior art keywords
temperature
voltage
vehicle
battery
current
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210579978.2A
Other languages
Chinese (zh)
Inventor
章正柱
董江平
王正明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Geely Evun Technology Co ltd
Zhejiang Geely Holding Group Co Ltd
Original Assignee
Hangzhou Geely Evun Technology Co ltd
Zhejiang Geely Holding Group Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hangzhou Geely Evun Technology Co ltd, Zhejiang Geely Holding Group Co Ltd filed Critical Hangzhou Geely Evun Technology Co ltd
Priority to CN202210579978.2A priority Critical patent/CN114919413A/en
Publication of CN114919413A publication Critical patent/CN114919413A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L3/00—Electric devices on electrically-propelled vehicles for safety purposes; Monitoring operating variables, e.g. speed, deceleration or energy consumption
    • B60L3/0023—Detecting, eliminating, remedying or compensating for drive train abnormalities, e.g. failures within the drive train
    • B60L3/0046—Detecting, eliminating, remedying or compensating for drive train abnormalities, e.g. failures within the drive train relating to electric energy storage systems, e.g. batteries or capacitors
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
    • B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/60—Other road transportation technologies with climate change mitigation effect
    • Y02T10/70—Energy storage systems for electromobility, e.g. batteries

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Sustainable Development (AREA)
  • Sustainable Energy (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Secondary Cells (AREA)

Abstract

The invention provides a method and a system for diagnosing battery abnormity of an electric vehicle. The method comprises the following steps: collecting battery data of X vehicles in a charging stage, wherein the battery data comprises the voltage or temperature of a battery, the current of the battery and the date which are collected at preset time intervals; dividing the battery current into a plurality of current intervals, and grouping the voltage or the temperature of the battery according to three dimensions of a vehicle-date-current interval; constructing a statistical model according to the vehicle-date-current interval grouping data to count the abnormal voltage or temperature proportion of X vehicles; and (4) performing descending order arrangement on the voltage or temperature abnormal proportion of the X vehicles, and sending out a battery abnormal early warning signal to the vehicles arranged in the front n. The method can timely send out early warning to the vehicle owner with the potential failure of the battery.

Description

Method and system for diagnosing battery abnormity of electric vehicle
Technical Field
The invention relates to the technical field of electric vehicle battery safety, in particular to a method and a system for diagnosing battery abnormity of an electric vehicle.
Background
The electric automobile is self-ignited due to the battery failure, so that the life safety and the property safety of an automobile owner are endangered, huge economic loss is caused to the automobile owner, and serious events such as casualties and the like are caused to the society. Therefore, potential problems and faults of the electric vehicle battery are identified in advance by monitoring the battery performance index of the electric vehicle, early warning is given to a vehicle owner in time, the vehicle owner is reminded of entering the station for maintenance, and a series of social hazard events caused by battery abnormity of the electric vehicle can be avoided.
Generally, an electric vehicle is in two states, i.e., a charged state and a discharged state. The electric automobile converts electric energy into kinetic energy in a discharge state to provide power for the automobile. In the driving process of the automobile, the automobile is easily influenced by personal and road condition factors such as driving habits of a driver, environment around a driving road and the like, the performance index of the battery is monitored to identify potential faults of the battery under the condition, an analysis result is easily influenced by various external variables, and the accuracy and the stability of the analysis result are reduced.
The prior art exists the following methods: the performance index of the battery is predicted through the time series prediction model, and the predicted value is compared with a performance index abnormity threshold value set in advance, and if the abnormal value is exceeded, the battery is considered to be possibly failed. The method has certain defects, on one hand, the historical value of the performance index is modeled and predicted only through time series model prediction, and the difficulty is very high, because the performance index of the battery, such as voltage fluctuation, is large, and the wave crest and the wave trough are difficult to accurately fit only by the time series model periodically. On the other hand, the difficulty of setting an abnormal threshold for a reasonable performance index is still high.
Disclosure of Invention
One purpose of the invention is to identify the potential fault of the automobile battery in advance and send out early warning to the automobile owner in time.
It is a further object of the invention to improve the accuracy of identifying a latent fault.
Specifically, the present invention provides a method of diagnosing an abnormality of a battery of an electric vehicle, including the steps of:
collecting battery data of X vehicles in a charging stage, wherein the battery data comprises the voltage or temperature of a battery, the current of the battery and the date which are collected at preset time intervals;
dividing the battery into d current intervals, and grouping the voltage or the temperature of the battery according to three dimensions of a vehicle-date-current interval;
counting the abnormal voltage or temperature proportion of X vehicles according to the vehicle-date-current interval grouped data;
and (3) performing descending order arrangement on the voltage or temperature abnormal proportion ratios of the X vehicles, and sending out a battery abnormal early warning signal to the vehicles arranged in the front n.
Optionally, the step of counting the abnormal voltage or temperature ratios of the X vehicles includes the following steps:
calculating and obtaining the abnormal voltage or temperature ratio r of the vehicle i in the current interval j according to the following formula j Wherein 0 < i < X:
Figure BDA0003661952430000021
wherein q is j The abnormal voltage or temperature record number of the vehicle i in the current section j is represented, and the W represents the total voltage or temperature record number of the vehicle i in the d current sections;
according to abnormal ratio r of voltage or temperature j The calculation formula (c) calculates and obtains the abnormal voltage or temperature proportion in other current intervals, thereby obtaining all the abnormal voltage or temperature proportions r of the d current intervals 1 、r 2 、...、r d And further obtaining the voltage or temperature abnormal ratio of the vehicle i:
Figure BDA0003661952430000022
optionally, the voltage or temperature abnormal proportion r of the vehicle i in the current interval j is obtained through calculation according to the following formula j Wherein 0 < i < X:
Figure BDA0003661952430000023
in the step (2) of (a),
the voltage or temperature anomaly of the vehicle i at a certain date within the current interval j is determined as follows:
according to the quartile principle, the mean value or the quantile value of the voltage or the temperature of the vehicle i on a certain date is recorded as z, and if the following conditions are met, the mean value or the quantile value of the voltage or the temperature of the vehicle i on the certain date is considered to be abnormal:
z＞Z 0.75 +1.5*(Z 0.7 5-Z 0.25 )
wherein Z is a random variable, and in the current interval j, the mean or fractional value of the voltage or temperature of all vehicles on each day is a sample of the random variable Z 0.75 And Z 0.25 Is the quantile of the random variable Z;
and recording that the voltage or the temperature of the vehicle i has one abnormality.
Optionally, the voltage or temperature abnormal proportion r of the vehicle i in the current interval j is obtained through calculation according to the following formula j Wherein 0 < i < X:
Figure BDA0003661952430000031
in the step (2) of (a),
the voltage or temperature anomaly of the vehicle i at a certain date within the current interval j is determined as follows:
according to the 3 sigma principle, the mean value or the quantile value of the voltage or the temperature of the vehicle i on a certain date is recorded as z, and if the following conditions are met, the mean value or the quantile value of the voltage or the temperature of the vehicle i on the date is considered to be abnormal:
z＞Z β
wherein Z is a random variable, and in the current interval j, the mean or fractional value of the voltage or temperature of all vehicles on each day is a sample of the random variable Z β Is the quantile of the random variable Z, P (Z > Z β ) 1-beta, beta is more than or equal to 0.9, and P represents a probability distribution function;
and recording that the voltage or the temperature of the vehicle i has one abnormality.
Optionally, in the step of counting the abnormal voltage or temperature ratios of the X vehicles, the abnormal voltage or temperature ratios are directly identified by using an unsupervised abnormal detection algorithm in machine learning.
Optionally, the step of counting the abnormal voltage or temperature ratio of the X vehicles includes the following steps:
grouping the vehicles i according to the date, combining the average value of the voltage or the temperature of all current intervals and various quantile values into an h-dimensional vector Q (c) 1 ,c 2 ,...,c h ) And Q (c) 1 ,c 2 ,...,c h ) As a characteristic of vehicle i on a certain date;
training and learning all the daily samples of the X vehicles by using an unsupervised anomaly detection algorithm, and counting the daily number of each vehicle as g, wherein the total number of the samples is X × g Q vectors;
outputting the result of whether the voltage or the temperature of the vehicle i in all the days within g dates is abnormal to obtain the total recorded number t of the voltage or the temperature abnormality of the vehicle i;
and obtaining the abnormal voltage or temperature ratio of the vehicle i according to the formula ratio t/g.
Optionally, the acquiring battery data of X vehicles in a charging phase includes the following steps:
acquiring the cell voltage or cell temperature of each vehicle in X vehicles in a charging stage, wherein a battery of the vehicle consists of m cells, and the cell voltage or the cell temperature at the moment of k is recorded as U k1 、U k2 、...、U km And the highest voltage is recorded as U k,max Minimum voltage of U k,min The median of the voltage is U k,median ；
Obtaining the pressure or temperature difference by one of the following equations
Figure BDA0003661952430000032
And applying said pressure or temperature difference
Figure BDA0003661952430000033
As the voltage or temperature of the battery:
Figure BDA0003661952430000034
Figure BDA0003661952430000035
Figure BDA0003661952430000036
wherein, U k,a％ Represents P (U) k ＞U k,a％ )＝a％，U k,b％ Represents P (U) k ＞U k,b％ ) B%, 0 < a < 100, 0 < b < 100, and a > b, U k Is a random variable, and the voltage or temperature of m cells is a random variable U k Sample of (2), P (U) k ) Is a probability distribution function;
optionally, the pressure or temperature difference is obtained by calculation using one of the following equations
Figure BDA0003661952430000041
In the step (2), which formula is selected to calculate the pressure difference or the temperature difference is determined in the following manner
Figure BDA0003661952430000042
The pressure difference or the temperature difference obtained by the calculation of the three formulas
Figure BDA0003661952430000043
Substituting the obtained data into the subsequent steps to verify the effect of the battery abnormity diagnosis;
pressure or temperature differences obtained by the most effective formula
Figure BDA0003661952430000044
And takes it as the voltage or temperature of the battery.
Optionally, dividing the battery current into d current intervals, and grouping the voltage or temperature of the battery according to three dimensions of a vehicle-date-current interval, including the steps of:
grouping the collected voltage or temperature of the battery according to vehicles and dates, and filling the voltage or temperature which is not recorded by a certain vehicle at a certain moment into a null value;
deleting the voltage or temperature data of the battery with the small sample amount of the vehicle-date to obtain the corrected vehicle-date grouping data;
and further grouping the corrected vehicle-date grouped data according to the current interval in the charging phase, thereby obtaining the grouped data with three dimensions of the vehicle-date-current interval.
Optionally, the deleting the voltage or temperature data of the vehicle-date battery with the small sample size to obtain the modified vehicle-date grouping data includes the following steps:
let Cnt be the number of data records per vehicle counted by a certain date for X vehicles 1 ,Cnt 2 ,...,Cnt X Null records do not count, Cnt is a random variable;
setting the alpha quantile of Cnt to Cnt α I.e. Cnt α Denotes P (Cnt > Cnt α ) 1- α, where p (cnt) is a probability distribution function;
deleting fewer than Cnt data records α Vehicle-date data of (a);
optionally, in the step of dividing the battery into d current intervals, the current intervals are determined according to the following method:
setting a plurality of possible values for the interval of the current;
calculating the voltage or temperature of the vehicle-date in each interval
Figure BDA0003661952430000045
Standard deviation of (2)
Figure BDA0003661952430000046
Will meet the standard deviation
Figure BDA0003661952430000047
The possible values less than or equal to the first preset threshold and greater than or equal to the second preset threshold are taken as the interval of the current.
In particular, the present invention provides a system for diagnosing battery abnormality of an electric vehicle, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method as described above when executing the computer program.
According to the scheme of the invention, the acquired data is the battery data of the vehicle in the charging stage, and compared with the battery data of the vehicle in the discharging stage, the influence of personal and road condition factors such as the driving habit of a driver, the environment around a driving road and the like is avoided, so that the stability and the accuracy of an analysis result are lower. And the voltage or the temperature of the battery is grouped according to three dimensions of a vehicle-date-current interval, the voltage or temperature abnormal proportion of the X vehicles is counted, the voltage or temperature abnormal proportion ratio of the X vehicles is arranged in a descending order, and the vehicle arranged in the front n sends out a battery abnormal early warning signal, so that early warning can be sent out to vehicle owners with potential faults of the battery in time.
The invention provides a method for dividing the charging current of a battery by using machine learning or statistics, wherein different current division intervals are regarded as various working conditions of the battery.
In addition, the battery fault can be identified more accurately by the method for counting the abnormal voltage or temperature ratio of X vehicles in the embodiment of the invention.
The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments thereof taken in conjunction with the accompanying drawings.
Drawings
Some specific embodiments of the invention will be described in detail hereinafter by way of example and not by way of limitation with reference to the accompanying drawings. The same reference numbers in the drawings identify the same or similar elements or components. Those skilled in the art will appreciate that the drawings are not necessarily drawn to scale. In the drawings:
FIG. 1 shows a schematic flow diagram of a method of electric vehicle battery abnormality diagnosis according to one embodiment of the present invention;
FIG. 2 shows a schematic flow diagram of the acquisition of battery data for X vehicles in a charging phase as shown in FIG. 1;
fig. 3 shows a schematic flow chart of a method of grouping the voltage or temperature of the battery in three dimensions of the vehicle-date-current interval in step S200 shown in fig. 1;
FIG. 4 shows a schematic flow diagram of a method of determining an interval of currents according to one embodiment of the invention;
fig. 5 is a schematic flow chart illustrating a method of counting the abnormal voltage or temperature ratios of X vehicles in step S300 shown in fig. 1;
FIG. 6 shows a schematic flow chart of a method of determining a voltage or temperature anomaly of a vehicle i at a date within a current interval j according to one embodiment of the present invention;
FIG. 7 shows a schematic flow diagram of a method of determining a voltage or temperature anomaly of a vehicle i at a certain date within a current interval j according to another embodiment of the present invention;
FIG. 8 shows a ranking of the first 16 vehicles in descending order of voltage or temperature anomaly ratio ratios for X vehicles, according to one embodiment of the invention;
FIG. 9 shows a pressure differential fluctuation graph for 5 accident vehicles, according to one embodiment of the present invention;
FIG. 10 shows a schematic block diagram of a system for electric vehicle battery abnormality diagnosis according to one embodiment of the present invention;
FIG. 11 shows a schematic flow chart of statistics of voltage or temperature anomaly ratios for X vehicles according to another embodiment of the present invention;
in the figure: 100-a data acquisition module, 200-a message analysis module, 300-a battery index analysis module, 400-an anomaly detection algorithm module and 500-an early warning signal distribution module.
Detailed Description
Fig. 1 shows a schematic flowchart of a method of diagnosing abnormality of a battery of an electric vehicle according to an embodiment of the present invention. As shown in fig. 1, the method includes:
step S100, collecting battery data of X vehicles in a charging stage, wherein the battery data comprises the voltage or temperature of a battery, the current of the battery and the date which are collected at intervals of preset time;
step S200, dividing the battery into d current intervals, and grouping the voltage or the temperature of the battery according to three dimensions of a vehicle-date-current interval;
step S300, counting the abnormal voltage or temperature proportion of X vehicles according to the vehicle-date-current interval grouping data;
and S400, performing descending order arrangement on the voltage or temperature abnormal proportion ratios of the X vehicles, and sending out a battery abnormal early warning signal to the vehicles arranged in the front n.
According to the scheme of the invention, the acquired data is the battery data of the vehicle in the charging stage, and compared with the battery data of the vehicle in the discharging stage, the influence of personal and road condition factors such as the driving habit of a driver, the environment around a driving road and the like is avoided, so that the stability and the accuracy of an analysis result are lower. And the voltage or the temperature of the battery is grouped according to three dimensions of a vehicle-date-current interval, the voltage or temperature abnormal proportion of X vehicles is counted, the voltage or temperature abnormal proportion ratios of the X vehicles are arranged in a descending order, and the vehicles arranged in the front n send out battery abnormal early warning signals, so that early warning can be sent out to vehicle owners with potential faults of the battery in time.
The following detailed description is given by way of specific examples:
the first embodiment is as follows:
fig. 2 shows a schematic flow diagram of the acquisition of battery data of X vehicles in a charging phase as shown in fig. 1. As shown in fig. 2, the method includes:
step S110, collecting the cell voltage or cell temperature of each vehicle in X vehicles in a charging stage, wherein a battery of the vehicle consists of m cells, and the cell voltage or the cell temperature at the moment of k is recorded as U k1 、U k2 、...、U km And the highest voltage is recorded as U k,max Minimum voltage is U k,min The median of the voltage is U k,median ；
Step S120, calculating and obtaining the pressure difference or the temperature difference by using the following formula
Figure BDA0003661952430000071
And applying said pressure or temperature difference
Figure BDA0003661952430000072
As the voltage or temperature of the battery:
Figure BDA0003661952430000073
Figure BDA0003661952430000074
Figure BDA0003661952430000075
wherein, U k,a％ Represents P (U) k ＞U k,a％ )＝a％，U k,b％ Represents P (U) k ＞U k,b％ ) B%, 0 < a < 100, 0 < b < 100, and a > b, U k Is a random variable, and the voltage or temperature of m cells is a random variable U k Sample of (2), P (U) k ) Is a probability distribution function.
In step S120, it is determined which formula is selected to calculate the pressure difference or the temperature difference in the following manner
Figure BDA0003661952430000076
The pressure difference or the temperature difference obtained by the calculation of the three formulas
Figure BDA0003661952430000077
Substituting into the subsequent steps to verify the diagnosis effect of battery abnormality, and obtaining pressure difference or temperature difference by the formula with the best effect
Figure BDA0003661952430000078
And takes it as the voltage or temperature of the battery.
In step S110, for each electric vehicle, a pressure difference or a temperature difference may be used
Figure BDA0003661952430000079
As one state of its battery at time k.
In the step S120, the process is performed,
Figure BDA00036619524300000710
the formula is represented by using the difference value of the highest voltage and the lowest voltage at the moment k as the differential pressure or the differential temperature, and the calculation mode is simple, but has a problem that the differential pressure or the differential temperature is easily influenced by abnormal values of the voltage or the temperature of m cells. The uploading of voltage or temperature data by the T-box may cause a problem of error in voltage or temperature data of one or more cells due to various reasons, such as a T-box fault, an analysis error of the data acquisition system, and the like, for example, if the error data is exactly the lowest voltage, a voltage difference calculation deviation is large, and a subsequent analysis result is unreliable.
Figure BDA00036619524300000711
The formula takes the difference value of the maximum voltage and the median voltage as the representation of the voltage difference, and can avoid the influence of abnormal cell voltage or cell temperature data to a certain extent.
Figure BDA00036619524300000712
This formula is further defined by usingThe voltage difference values of different quantiles are used as the voltage difference representation at the moment k, so that the influence of abnormal cell voltage data can be completely avoided. In theory, the equation is the most reasonable differential pressure calculation method, and in the equation, a and b are parameters, and may be empirically set to be a 95 and b 5, or different values may be set according to the voltage data of the actual battery, and experiments may be performed to screen out a set of reasonable values.
Figure BDA00036619524300000713
Is actually
Figure BDA00036619524300000714
I.e. when a is 100 and b is 50.
In the implementation of the actual electric vehicle fault early warning scheme, the three formulas can be respectively calculated, and then which formula is used for evaluating which calculation mode is adopted for obtaining the better final effect of verification, wherein the evaluation mode is that the pressure difference or the temperature difference obtained by calculation according to the formula is adopted
Figure BDA0003661952430000081
Substituting into the subsequent steps, and finally verifying the effect of the battery abnormity diagnosis.
Fig. 3 shows a schematic flow chart of a method for grouping the voltage or temperature of the battery in three dimensions of the vehicle-date-current interval in step S200 shown in fig. 1. The method comprises the following steps:
step S210, grouping the acquired voltage or temperature of the battery according to vehicles and dates, and filling the voltage or temperature which is not recorded by a certain vehicle at a certain moment into a null value;
step S220, deleting the voltage or temperature data of the vehicle-date battery with less sample amount to obtain the corrected vehicle-date grouping data;
in step S230, the corrected vehicle-date grouping data are further grouped according to the current interval in the charging phase, so as to obtain grouping data of three dimensions of the vehicle-date-current interval.
In the step S210, the coal miningThe collected voltage or temperature of the battery is grouped according to vehicle and date, wherein the date is day-accurate. For the vehicle i, data is collected every 10 seconds, the time every 10 seconds is recorded as time K, and the total number of data collection times per day is K. Thus, for a certain day, the pressure difference or temperature difference, respectively, of the ith vehicle is
Figure BDA0003661952430000082
If no record is recorded at a certain time, the pressure difference filling at the time is a null value. The method takes the pressure difference or the temperature difference of a certain vehicle in one day as a random variable
Figure BDA0003661952430000083
The recorded number of all the pressure difference values or temperature difference values of X vehicles in a certain day is counted for ensuring
Figure BDA0003661952430000084
Is a random variable whose number of samples must reach a certain number, and if the number of samples is small, the distribution function formed by the random variable may not be accurately characterized
Figure BDA0003661952430000085
This random variable. Therefore, it is necessary to delete the vehicle-date differential pressure values whose sample size is small.
The method uses the finest data acquisition granularity, collects the battery data of the battery of the electric automobile once every 10 seconds, and then constructs the analysis index by a statistical method. The processing method can consider the microscopic characteristics of the analysis indexes and the macroscopic characteristics of the analysis indexes. The hidden value of the data is comprehensively mined from the macro and micro angles in all aspects.
In step S220, for X vehicles, the number of data records per vehicle counted on a certain date is Cnt 1 ，Cnt 2 ，……，Cnt X In the method, the null record is not counted, and Cnt is considered as a random variable, and p (Cnt) is a probability distribution function thereof. The vehicle-date data for which the number of data records is less than Cnt α, α may be empirically set to 0.1, 0.2, etc., is deleted.Alpha can also be set to a range, such as alpha epsilon 0.05,0.4]Then, a plurality of experimental analyses are carried out according to specific battery data, and a reasonable value is found in the range.
In step S230, the current of the battery of the electric vehicle may fluctuate greatly during the charging process, for example, it may fluctuate from-200 to 0, or it may fluctuate in other larger ranges. If the pressure difference data in the range of [ -200,0] is modeled, the change of the pressure difference can be influenced by the large fluctuation of the current. Therefore, the method divides the current into a plurality of intervals, for example, the current of [ -200,0] can be divided into current intervals of [ -200, -190], [ -190,180], … … and [ -10,0]20 small according to the interval of 10 amperes, each current interval is marked by the serial number 1, 2, … … and 20, and therefore vehicle-date data can be further grouped according to three dimensions of the vehicle-date-current intervals. Assuming that the current interval is finally divided into d current intervals, generally d > 5.
The current is divided into small current intervals, which can be empirically divided at certain intervals, such as 10 amperes directly, but can be divided by 5 amperes, 15 amperes or other values.
Fig. 4 shows a schematic flow diagram of a method of determining an interval of currents according to an embodiment of the invention. As shown in fig. 4, the method includes:
step S231, setting a plurality of possible values for the interval of the current;
step S232, calculating the voltage or temperature of the vehicle-date in each interval
Figure BDA0003661952430000093
Standard deviation of (2)
Figure BDA0003661952430000094
Step S233, standard deviation will be met
Figure BDA0003661952430000095
The interval of the current is set as a possible value which is less than or equal to a first preset threshold and greater than or equal to a second preset threshold.
In step S231, possible values may be, for example, 3, 5, 8, 10, 12, 15 amps, etc. In step S233, the first preset threshold may be, for example, 0.02, and the second preset threshold may be, for example, 0.01. For example, when 8 amperes is selected as the interval, the vehicle-date pressure difference or temperature difference variable is counted
Figure BDA0003661952430000096
The standard deviation of the current is in accordance with the condition, and the current interval is divided according to 8 amperes in subsequent data processing.
Fig. 5 shows a schematic flowchart of the method for counting the abnormal voltage or temperature ratio of X vehicles in step S300 shown in fig. 1. As shown in fig. 5, the method includes:
step S310, calculating and obtaining the abnormal voltage or temperature ratio r of the vehicle i in the current section j according to the following formula j Wherein i is more than 0 and less than X:
Figure BDA0003661952430000091
wherein q is j The abnormal voltage or temperature record number of the vehicle i in the current section j is represented, and the W represents the total voltage or temperature record number of the vehicle i in the d current sections;
step S320, according to the abnormal ratio r of voltage or temperature j The calculation formula calculates and obtains the abnormal proportion of the voltage or the temperature in other current intervals, thereby obtaining all the abnormal proportion r of the voltage or the temperature in d current intervals 1 、r 2 、...、r d And further obtaining the voltage or temperature abnormal ratio of the vehicle i:
Figure BDA0003661952430000092
before step S310, the method further includes the following steps:
1) for a certain current interval j, deleting the record of the pressure difference or temperature difference data of the vehicle-current interval j as a null value;
2) for a certain vehicle i, the total number of data in the current interval j is pj;
3) according to a certain statistical method, screening out the abnormal pressure difference or temperature difference record number q of the vehicle i in the current interval j j ；
4) Calculating the abnormal proportion q of the vehicle i in the current section j j /pj；
5) Multiplying the anomaly ratio by a weight coefficient w j Assuming that the total number of records of the vehicle i in the d current sections is W, W is j ＝p j /W。
In the step S310 of the present embodiment,
Figure BDA0003661952430000101
in the step S320, the process proceeds,
Figure BDA0003661952430000102
fig. 6 shows a schematic flow chart of a method of determining a voltage or temperature anomaly of a vehicle i at a certain date within a current interval j according to an embodiment of the invention. As shown in fig. 6, the voltage or temperature abnormality of the vehicle i at a certain date in the current interval j is determined as follows:
step S311, according to the quartile principle, the mean value or the quantile value of the voltage or the temperature of the vehicle i on a certain date is recorded as z, and if the following conditions are satisfied, the mean value or the quantile value of the voltage or the temperature of the vehicle i on the certain date is considered to be abnormal:
z＞Z 0.75 +1.5*(Z 0.7 5-Z 0.25 )
wherein Z is a random variable, and in the current interval j, the mean or fractional value of the voltage or temperature of all vehicles on each day is a sample of the random variable Z 0.75 And Z 0.25 Is the quantile of the random variable Z;
in step S312, it is recorded that the voltage or temperature of the vehicle i is abnormal once.
In step S311, the quantile value of the voltage or temperature of the vehicle i on a certain date may be, for example, a 10% quantile, a 20% quantile, a 30% quantile, a 40% quantile, a 50% quantile, a 60% quantile, a 70% quantile, an 80% quantile, or a 90% quantile.
Fig. 7 shows a schematic flow chart of a method of determining a voltage or temperature anomaly of a vehicle i at a certain date within a current interval j according to another embodiment of the present invention. As shown in fig. 7, the voltage or temperature abnormality of the vehicle i at a certain date in the current interval j is determined in the following manner:
step S313, in accordance with the 3 σ rule, the mean value or the quantile value of the voltage or the temperature of the vehicle i on a certain date is represented as z, and if the following conditions are satisfied, the mean value or the quantile value of the voltage or the temperature of the vehicle i on the certain date is considered to be abnormal:
z＞Z β
wherein Z is a random variable, and in the current interval j, the mean or fractional value of the voltage or temperature of all vehicles on each day is a sample of the random variable Z β Is the quantile of a random variable Z, P (Z > Z) β ) 1-beta, beta is more than or equal to 0.9, and P represents a probability distribution function;
in step S314, it is recorded that the voltage or temperature of the vehicle i is abnormal once.
In step S313, β generally takes a value of 0.95 or 0.99. The smaller the value of β is set, the higher the probability that the accident vehicle is identified, but the value set cannot be too small, and if too small, it causes the 3 σ rule to be violated, so β is generally 0.9 or more.
In step S314, for example, if the pressure difference or temperature difference value z of the current section 1 of the vehicle i in a certain day (e.g. 2020-10-25) satisfies the above formula in step S311 or step S313, it is considered that the pressure difference or temperature difference average value of the vehicle i in the current section 1 in the day 2020-10-25 is an abnormal value, and the battery pressure difference of the vehicle i is recorded to be abnormal once.
In step S400, n may be 10, 20, etc., or may be calculated as a ratio of the total number X of vehicles, such as ceil (X0.01), which is an upward integer function, such as ceil (1.2) 2.
In order to verify the accuracy of the method of the embodiment of the invention, the method comprises the following steps that Y batteries are contained in X vehiclesWhen the accident vehicle is analyzed and modeled by the method, the data of the accident occurrence date of the Y vehicle and all the data after the accident occurrence are deleted. In a specific embodiment, for example, X is 1000 and Y is 5. In the specific implementation, the specific application in step S100
Figure BDA0003661952430000111
In step S230, the current is divided into small current sections at intervals of 10 amperes. As shown in fig. 8, a ranking chart of the first 16 vehicles of the X vehicles of the embodiment of the present invention is shown, in which the voltage or temperature abnormal ratio of the vehicles is arranged in descending order, where VIN is the VIN number of the vehicle in VIN column, and ratio is the calculated abnormal pressure difference or temperature difference recording ratio. As can be seen from fig. 8, according to the method described herein, it can be considered that the front ceil (1000 x 0.01) ═ 10 vehicles are at risk of battery failure. As shown in the figure, the first 3 ranked vehicles in the first 10 vehicles are actual accident vehicles, and the other 7 vehicles are non-accident vehicles.
The change in differential pressure for 5 accident vehicles with potential battery failure is shown in fig. 9, in which the boxed partial graph includes the differential pressure fluctuation curve for the accident vehicle with vin equal to 4 at the top and the differential pressure fluctuation curve for the accident vehicle with vin equal to 5 at the bottom, and vin equal to 4 and vin equal to 5 are two accident vehicles that are not identified by the method set forth herein. From the graph of the differential pressure change, it can be seen that the fluctuation amplitude of the differential pressure data is smaller than that of the other 3 vehicles, and the differential pressure fluctuation of the two vehicles is basically smooth, so that the method cannot detect that the vehicle is an accident vehicle. As can be roughly estimated from the figure, vehicles with vin 4 and vin 5 are not identified because their battery faults are not reflected by a differential pressure fluctuation.
In particular, the invention also provides a system for diagnosing the battery abnormality of the electric vehicle, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of the method when executing the computer program.
The configuration of the system for diagnosing the abnormality of the battery of the electric vehicle will be briefly described below. As shown in fig. 10, the system further includes a data collection module 100, where the data collection module 100 collects the message data uploaded by the electric vehicle T-Box in real time. The processor comprises a message analysis module 200, a battery index analysis module 300, an anomaly detection algorithm module 400 and an early warning signal distribution module 500. The message parsing module 200 is connected to the data acquisition module 100, and is configured to parse the current of the battery, the real-time voltage and temperature data of all the battery cores, or other performance index data according to a message specification.
Because the message is not in accordance with the specification due to the accidental fault of the T-Box equipment and the transmission error of the uploaded message, and some indexes obtained by analyzing the message may be abnormal values or wrong values, a certain abnormal filtering rule needs to be set, abnormal data is corrected or directly deleted, and the acquired data is basically regular and orderly battery performance index data through data cleaning and data processing of the battery index analysis module 300.
The anomaly detection algorithm module 400 analyzes the battery performance indexes (such as differential pressure and temperature difference) in steps according to the method provided by the foregoing, and performs modeling analysis on the anomaly data by using the algorithm set forth above to obtain the ratio of the index anomaly data of each vehicle, and then arranges the index anomaly data in a descending order according to the ratio of the anomaly data to determine the high risk of battery failure of the vehicle n before ranking. The early warning signal distribution module 500 is used for distributing the risk early warning signal to the owner of the electric vehicle with the high-risk battery fault through a data distribution technology according to the analysis result of the anomaly detection algorithm module, and reminding the owner of checking the battery or entering the station for maintenance. Other features of the system correspond to the above-described method one-to-one, and are not described herein again.
The second embodiment:
the difference between the second embodiment and the first embodiment is the difference of step S300. In the second embodiment of the invention, the method for counting the abnormal voltage or temperature proportion of the X vehicles directly identifies the abnormal voltage or temperature proportion by using an unsupervised abnormal detection algorithm in machine learning. The unsupervised anomaly detection algorithm may be, for example, one-class SVM, ifoest, or the like. For example, fig. 11 shows a schematic flowchart of counting the abnormal ratio of the voltage or the temperature of X vehicles according to the embodiment of the present invention. As shown in fig. 11, the step S300 includes:
step S321, grouping the vehicles i according to the date, combining the average value of the voltage or the temperature of all the current intervals and various quantile values into an h-dimensional vector Q (c) 1 ,c 2 ,...,c h ) And Q (c) 1 ,c 2 ,...,c h ) As a characteristic of vehicle i on a certain date;
step S322, using unsupervised anomaly detection algorithm to train and learn samples of all days of X vehicles, and recording the number of days of each vehicle as g, wherein the total number of the samples is X g Q vectors;
step S323, outputting the result of whether the voltage or the temperature of the vehicle i in all the days within g dates is abnormal to obtain the total record number t of the voltage or the temperature abnormality of the vehicle i;
in step S324, the voltage or temperature abnormality ratio of the vehicle i is obtained according to the formula ratio t/g.
Other technical solutions are the same as those in the first embodiment, and are not described herein again. The invention obtains the funding of national key research and development plan (No. 2020YFB1711803).
Thus, it should be appreciated by those skilled in the art that while a number of exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications of the general principles of the invention may be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the invention. Accordingly, the scope of the invention should be understood and interpreted to cover all such other variations or modifications.

Claims (10)

1. A method of diagnosing an abnormality of a battery of an electric vehicle, characterized by comprising the steps of:
collecting battery data of X vehicles in a charging stage, wherein the battery data comprises the voltage or temperature of a battery, the current of the battery and the date which are collected at preset time intervals;
dividing the battery current into d current intervals, and grouping the voltage or the temperature of the battery according to three dimensions of a vehicle-date-current interval;
constructing a statistical model according to the vehicle-date-current interval grouping data to count the abnormal voltage or temperature proportion of X vehicles;
and (3) performing descending order arrangement on the voltage or temperature abnormal proportion ratios of the X vehicles, and sending out a battery abnormal early warning signal to the vehicles arranged in the front n.
2. The method according to claim 1, wherein the step of counting the abnormal voltage or temperature proportion of X vehicles comprises the following steps:
calculating and obtaining the abnormal voltage or temperature ratio r of the vehicle i in the current interval j according to the following formula j Wherein i is more than 0 and less than X:
Figure FDA0003661952420000011
wherein q is j The abnormal voltage or temperature record number of the vehicle i in the current section j is represented, and the W represents the total voltage or temperature record number of the vehicle i in the d current sections;
according to abnormal ratio r of voltage or temperature j The calculation formula (c) calculates and obtains the abnormal voltage or temperature proportion in other current intervals, thereby obtaining all the abnormal voltage or temperature proportions r of the d current intervals 1 、r 2 、...、r d And further obtaining the voltage or temperature abnormal ratio of the vehicle i:
Figure FDA0003661952420000012
3. method according to claim 2, characterized in that the voltage or temperature anomaly ratio r of the vehicle i in the current interval j is calculated according to the following formula j Wherein i is more than 0 and less than X:
Figure FDA0003661952420000013
in the step (2) of (a),
the voltage or temperature anomaly of the vehicle i at a certain date within the current interval j is determined as follows:
according to the quartile principle, the mean value or the quantile value of the voltage or the temperature of the vehicle i on a certain date is recorded as z, and if the following conditions are met, the mean value or the quantile value of the voltage or the temperature of the vehicle i on the date is considered to be abnormal:
z＞Z 0.75 +1.5*(Z 0.7 5-Z 0.25 )
wherein Z is a random variable, and in the current interval j, the mean or fractional value of the voltage or temperature of all vehicles on each day is a sample of the random variable Z 0.75 And Z 0.25 Is the quantile of the random variable Z;
and recording that the voltage or the temperature of the vehicle i has one abnormality.
4. Method according to claim 2, characterized in that the voltage or temperature anomaly ratio r of the vehicle i in the current interval j is calculated according to the following formula j Wherein 0 < i < X:
Figure FDA0003661952420000021
in the step (2) of (a),
the voltage or temperature anomaly of the vehicle i at a certain date within the current interval j is determined as follows:
according to the 3 sigma principle, the mean value or the quantile value of the voltage or the temperature of the vehicle i on a certain date is recorded as z, and if the following conditions are met, the mean value or the quantile value of the voltage or the temperature of the vehicle i on the certain date is considered to be abnormal:
z＞Z β
wherein Z is a random variable, and in the current interval j, the mean or fractional value of the voltage or temperature of all vehicles on each day is a sample of the random variable Z β Is the quantile of the random variable Z, P (Z > Z β ) 1-beta, beta is more than or equal to 0.9, and P represents a probability distribution function;
and recording the voltage or temperature of the vehicle i as abnormal once.
5. The method according to claim 1, wherein in the step of counting the abnormal ratio of the voltage or the temperature of the X vehicles, the abnormal ratio of the voltage or the temperature is directly identified by using an algorithm of unsupervised abnormality detection in machine learning.
6. The method according to claim 5, wherein the step of counting the abnormal voltage or temperature ratio of X vehicles according to the vehicle-date-current interval grouping data comprises the following steps:
grouping the vehicles i according to the date, combining the average value of the voltage or the temperature of all current intervals and various quantile values to form an h-dimensional vector Q (c) 1 ,c 2 ,...,c h ) And Q (c) 1 ,c 2 ,...,c h ) As a characteristic of vehicle i on a certain date;
training and learning samples of all days of the X vehicles by using an unsupervised anomaly detection algorithm, and counting the number of days of each vehicle as g, wherein the total number of the samples is X g Q vectors;
outputting the result of whether the voltage or the temperature of the vehicle i in all the days within g dates is abnormal to obtain the total recorded number t of the voltage or the temperature abnormality of the vehicle i;
and obtaining the abnormal voltage or temperature ratio of the vehicle i according to the formula ratio t/g.
7. The method according to any one of claims 1-6, wherein said collecting battery data of X vehicles in a charging phase comprises the steps of:
acquiring the cell voltage or cell temperature of each vehicle in X vehicles in a charging stage, wherein a battery of the vehicle consists of m cells, and the cell voltage or the cell temperature at the moment of k is recorded as U k1 、U k2 、...、U km And the highest voltage is recorded as U k,max Minimum voltage is U k,min The median of the voltage is U k,median ；
Obtaining the pressure or temperature difference by one of the following equations
Figure FDA0003661952420000034
And applying said pressure or temperature difference
Figure FDA0003661952420000035
As the voltage or temperature of the battery:
Figure FDA0003661952420000031
Figure FDA0003661952420000032
Figure FDA0003661952420000033
wherein, U k,a％ Represents P (U) k ＞U k,a％ )＝a％，U k,b％ Represents P (U) k ＞U k,b％ ) B%, 0 < a < 100, 0 < b < 100, and a > b, U k Is a random variable, and the voltage or temperature of m cells is a random variable U k Sample of (2), P (U) k ) Is a probability distribution function;
optionally, the pressure or temperature difference is obtained by calculation using one of the following equations
Figure FDA0003661952420000036
In the step (2), which formula is selected to calculate the differential pressure or the differential temperature is determined in the following manner
Figure FDA0003661952420000037
The pressure difference or the temperature difference obtained by calculation by using the three formulas
Figure FDA0003661952420000038
Substituting the obtained data into the subsequent steps to verify the effect of battery abnormity diagnosis;
pressure or temperature differences obtained by means of the most effective formula
Figure FDA0003661952420000039
And takes it as the voltage or temperature of the battery.
8. The method of claim 7, wherein dividing the battery current into d current intervals and grouping the voltage or temperature of the battery in three dimensions of a vehicle-date-current interval comprises the steps of:
grouping the collected voltage or temperature of the battery according to vehicles and dates, and filling the voltage or temperature which is not recorded by a certain vehicle at a certain moment into a null value;
deleting the voltage or temperature data of the battery with the small sample amount of the vehicle-date to obtain the corrected vehicle-date grouping data;
and further grouping the modified vehicle-date grouping data according to the current interval in the charging phase, thereby obtaining grouping data of three dimensions of the vehicle-date-current interval.
9. The method according to claim 8, wherein the deleting the voltage or temperature data of the battery of the vehicle-date having the small sample amount to obtain the corrected vehicle-date grouping data comprises the steps of:
let Cnt be the number of data records per vehicle counted by a certain date for X vehicles 1 ,Cnt 2 ,...,Cnt X Null records do not count, Cnt is a random variable;
setting alpha quantile of Cnt to Cnt α I.e. Cnt α Denotes P (Cnt > Cnt α ) 1- α, where p (cnt) is a probability distribution function;
deleting fewer data records than Cnt α Number of vehicle-datesAccordingly;
optionally, in the step of dividing the battery into d current intervals, the current intervals are determined according to the following method:
setting a plurality of possible values for the interval of the current;
calculating the voltage or temperature of the vehicle-date in each interval
Figure FDA0003661952420000041
Standard deviation of (2)
Figure FDA0003661952420000042
Will meet the standard deviation
Figure FDA0003661952420000043
The possible values less than or equal to the first preset threshold and greater than or equal to the second preset threshold are taken as the interval of the current.
10. A system for electric vehicle battery abnormality diagnosis, characterized by comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method according to any one of claims 1-9 when executing the computer program.
CN202210579978.2A 2022-05-25 2022-05-25 Method and system for diagnosing battery abnormity of electric vehicle Pending CN114919413A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210579978.2A CN114919413A (en) 2022-05-25 2022-05-25 Method and system for diagnosing battery abnormity of electric vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210579978.2A CN114919413A (en) 2022-05-25 2022-05-25 Method and system for diagnosing battery abnormity of electric vehicle

Publications (1)

Publication Number Publication Date
CN114919413A true CN114919413A (en) 2022-08-19

Family

ID=82811598

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210579978.2A Pending CN114919413A (en) 2022-05-25 2022-05-25 Method and system for diagnosing battery abnormity of electric vehicle

Country Status (1)

Country Link
CN (1) CN114919413A (en)

Similar Documents

Publication Publication Date Title
CN111241154B (en) Storage battery fault early warning method and system based on big data
CN112092675A (en) Battery thermal runaway early warning method, system and server
CN112946483B (en) Comprehensive evaluation method for battery health of electric vehicle and storage medium
CN109604186A (en) Power battery performance flexibility assesses method for separating
CN115366683B (en) Fault diagnosis strategy for multi-dimensional model fusion of power battery of new energy automobile
CN112255560B (en) Battery cell health degree prediction method
CN111762096A (en) New energy automobile safety early warning method and system based on artificial intelligence
CN110146817A (en) The diagnostic method of lithium battery failure
CN114829961A (en) Method and system for predicting battery health using machine learning model
CN114361616B (en) Battery thermal runaway early warning method, system and storage medium
CN116482536A (en) Power battery fault early warning and safety risk assessment method based on data driving
DE102019125375A1 (en) State value for rechargeable batteries
CN115372828A (en) Battery cell consistency evaluation method based on charging segment data and unsupervised algorithm
CN114397592A (en) Health degree prediction method and device for power battery of electric vehicle
Sun et al. An ensemble system to predict the spatiotemporal distribution of energy security weaknesses in transmission networks
CN114462259A (en) SOC-based battery remaining life prediction method and system, automobile and medium
DE102022120962A1 (en) BATTERY ERROR DETECTION
CN116482540A (en) Analysis and prediction method, device and system for battery voltage inconsistency
Li et al. A Novel Method for Lithium‐Ion Battery Fault Diagnosis of Electric Vehicle Based on Real‐Time Voltage
CN114036647A (en) Power battery safety risk assessment method based on real vehicle data
CN114631032A (en) Method and system for monitoring health of battery pack
CN117630672A (en) Method and device for predicting a device battery of a diagnostic device using a converter model
CN114919413A (en) Method and system for diagnosing battery abnormity of electric vehicle
CN208350965U (en) A kind of pure electric automobile battery state-of-health measures device
CN116901707A (en) Power battery pack fault early warning method, system and vehicle

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination