CN108196190B

CN108196190B - Online fault diagnosis method for battery pack

Info

Publication number: CN108196190B
Application number: CN201711160325.6A
Authority: CN
Inventors: 郑岳久; 金昌勇; 高文凯; 来鑫; 黄鹏; 秦超
Original assignee: University of Shanghai for Science and Technology
Current assignee: University of Shanghai for Science and Technology
Priority date: 2017-11-20
Filing date: 2017-11-20
Publication date: 2020-02-18
Anticipated expiration: 2037-11-20
Also published as: CN108196190A

Abstract

The invention relates to an online fault diagnosis method for a battery pack, which is used for judging the reason of large difference between the single charge state of a battery and the average charge state of the battery pack and comprises the following steps: s1, intercepting a section of working condition of the battery pack, and identifying the type of the working condition based on the working condition; s2, acquiring the state of charge difference delta SOC between the state of charge of all battery monomers in the battery pack and the average state of charge of the battery pack under the working condition, and acquiring difference change characteristics; and S3, determining the reason that the difference between the battery cell charge state and the average charge state of the battery pack is large based on the working condition type obtained in the step S1 and the difference change characteristic obtained in the step S2 to be the existence of the initial charge State (SOC) difference or the occurrence of micro short circuit or the existence of the cell average capacity difference. The method has the advantage that the fault reason can be accurately found.

Description

Online fault diagnosis method for battery pack

Technical Field

The invention belongs to the technical field of on-line diagnosis of battery management system faults, particularly relates to an on-line fault diagnosis method for a battery pack, and particularly relates to a determination method for on-line diagnosis of whether a monomer in a series battery pack has an initial SOC difference or a micro short circuit or a monomer average capacity difference according to a monomer delta SOC change characteristic.

Background

In the normal use process of the electric automobile, most of thermal runaway of the electric automobile is caused by spontaneous combustion, and the main reason for causing the spontaneous combustion of the electric automobile is internal short circuit of a battery. Diagnosis of internal short circuitFor example, the method disclosed in chinese patent application CN106802396A identifies the difference in the state of charge SOC of the battery cells in the battery pack by using a frequency division model in combination with EKF, and diagnoses the approximate short-circuit resistance value of the internal short-circuit battery cell by using the difference state of charge Δ SOC obtained by the method. In the initial stage of the internal short circuit of the battery, if effective measures can be found and taken in time, the use safety and reliability of the electric automobile can be greatly improved. The phenomena of the initial short circuit in the battery and the short circuit outside the battery have very similar characteristics, for example, the two fault conditions can cause the state of charge (SOC) of a short-circuit single body in a period of time_Sheet) And battery average SOC (SOC)_mean) The SOC difference (Δ SOC) between them becomes larger, so we collectively refer to these two short-circuit failures as micro-short circuits.

However, when the absolute value of Δ SOC of a certain cell in the series-connected battery pack is increased over a period of time, a phenomenon is similarly shown in which the difference is increased not only when a micro short circuit occurs in the battery cell but also when the capacity of the certain cell is greatly different from the average capacity of the battery pack. For example, in a series battery pack, the average capacity of the battery pack is 100Ah, and the capacity of a certain battery cell A is 90Ah, i.e. there is a cell average capacity difference. It is assumed that all cells in the battery pack are at the same temperature and are fully charged, i.e., all cells have a current SOC of 100%. When the discharge cutoff state is reached, i.e., cell a state of charge is 0%, the average state of charge of the battery pack is 10% and the SOC variation of cell a changes from the initial 0% to 10%. During this discharge, the SOC difference of the cell a shows a phenomenon that becomes larger and larger, and the behavior of this phenomenon is similar to that of the cell in which a micro short circuit occurs.

If the reason for the phenomenon is not distinguished when the battery cell is subjected to the micro short circuit according to the phenomenon that the cell delta SOC is continuously increased in a certain period of time, the situation of misjudgment is definitely caused, but no related technology is available for effectively distinguishing two fault reasons.

Disclosure of Invention

The present invention is directed to provide an online battery pack fault diagnosis method for reducing erroneous determination, which overcomes the above-mentioned drawbacks of the prior art.

The purpose of the invention can be realized by the following technical scheme:

the online fault diagnosis method for the battery pack is used for judging the reason of large difference between the single battery charge state and the average battery charge state, and comprises the following steps of:

s1, intercepting a section of working condition of the battery pack, and identifying the type of the working condition based on the working condition;

s2, acquiring the state of charge difference delta SOC between the state of charge of all battery monomers in the battery pack and the average state of charge of the battery pack under the working condition, and acquiring difference change characteristics;

and S3, determining the reason that the difference between the battery cell charge state and the average charge state of the battery pack is large because of the initial SOC difference, the micro short circuit or the cell average capacity difference based on the working condition type obtained in the step S1 and the difference change characteristics obtained in the step S2.

Further, in step S1, the operating condition is determined by a time-dependent battery pack average state of charge (SOC)_mean-t curve characterization.

Further, in step S1, the operating condition types include:

SOC_mean-a first condition in which the t-curve undulation is less than a threshold x 1;

SOC_mean-a second condition in which the t-curve fluctuates unidirectionally above a threshold x 2;

SOC_mean-a third condition in which the t-curve comprises bi-directional undulations and both unidirectional segment undulations are greater than a threshold x 3;

SOC_mean-a fourth operating condition characterized by a t-curve having both the first operating condition and the second operating condition.

Further, in step S2, the difference change characteristics include a time-varying trend of Δ SOC and a difference change value λ, and the difference change value λ is calculated by:

λ＝[ΔSOC]max－[ΔSOC]min

wherein [ Δ SOC ] max and [ Δ SOC ] min are the maximum state of charge difference and the minimum state of charge difference respectively under the duration of the working condition.

Further, in step S2, the state of charge difference is calculated by the following formula:

ΔSOC＝SOC_mean-SOC

wherein, Δ SOC is the difference of the state of charge of a certain battery cell in the battery pack, SOC is the state of charge of the battery cell_meanIs the average state of charge of the battery.

Further, the step S3 is specifically:

if the type of the working condition is identified as a first working condition, judging the time-varying trend of the delta SOC, if the time-varying difference of the states of charge changes and the difference change value lambda is smaller than a threshold lambda 1, quitting the diagnosis, otherwise, judging that micro short circuit occurs;

if the type of the working condition is identified as a second working condition, comparing the difference change value lambda with a threshold lambda 1, if the difference change value lambda is smaller than the threshold lambda 1, judging whether the maximum charge state difference under the current working condition is smaller than a threshold theta, if so, judging that the single body is normal, and if not, judging the initial SOC difference; if the difference change value lambda is larger than or equal to the threshold value lambda 1, judging that the monomer is abnormal;

if the working condition type is identified as a third working condition or a fourth working condition, comparing the difference change value lambda with a threshold lambda 1, if the difference change value lambda is smaller than the threshold lambda 1, judging whether the maximum charge state difference under the current working condition is smaller than a threshold theta, if so, judging that the single body is normal, and if not, judging that the initial SOC difference is large; and if the difference change value lambda is larger than or equal to a threshold value lambda 1, judging whether the delta SOC changes along with the time, if so, judging that micro short circuit occurs, and if not, judging that the average capacity difference of the monomers exists.

Further, the difference in the average cell capacity in step S3 means that there is a difference between the average cell capacity and the average cell pack capacity.

Compared with the prior art, the method and the device have the advantages that the single battery faults are judged through the charge state difference between the charge state of the single battery and the average charge state of the battery pack based on different working conditions, the reason causing the phenomenon that the SOC difference of the single battery is large is further distinguished, the fault reasons such as capacity errors and the like are prevented from being mistakenly judged as micro short circuits, and the fault reasons are accurately found out.

Drawings

FIG. 1 is a flow chart of a battery pack online fault diagnosis method of the present invention;

FIG. 2 shows a first operating condition SOC_mean-a schematic diagram of the t-curve, wherein (2a) the battery is subjected to a micro-charge and micro-discharge cycle and (2b) the battery is in a resting state;

FIG. 3 is a second operating condition SOC_meanA schematic diagram of t curves, wherein (3a) is a curve of a battery pack during unidirectional discharge, (3b) is a curve of a battery pack during unidirectional charge, and (3c) is a curve of unidirectional discharge of other types of battery packs;

FIG. 4 is a third operating condition SOC_meanA schematic diagram of curves, wherein (4a) is a curve of a battery set when the battery set is charged first and then discharged, and (4b) is a curve of the battery set when the battery set is discharged first and then charged;

FIG. 5 is a fourth operating condition SOC_mean-t-curve diagram, wherein (5a) and (5b) represent two curves comprising both a first operating condition and a second operating condition;

FIG. 6 is a schematic view of an embodiment of a micro-short-circuit cell under a first operating condition;

FIG. 7 is a schematic diagram of an embodiment of a micro-short circuit and capacity difference monomer under a second condition;

FIG. 8 is a schematic diagram of an embodiment of a micro-short circuit and capacity difference monomer under a third condition;

FIG. 9 is a schematic diagram of an embodiment of a micro short circuit, capacity variation, and initial SOC variation unit under a fourth condition.

Detailed Description

The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.

The core content of the invention is that the abnormal monomer is diagnosed according to the difference of the normal monomer delta SOC and the abnormal monomer delta SOC change characteristics under different working conditions, and the reason that the delta SOC change abnormality of the monomer in the series battery pack is the initial SOC difference (namely the initial SOC of the monomer is greatly different from the average SOC of the battery pack), the micro short circuit is generated or the average capacity difference (namely the monomer capacity is greatly different from the average capacity of the battery pack) is diagnosed. As shown in fig. 1, the method specifically comprises the following steps:

and S1, intercepting a working condition of the battery pack by a Battery Management System (BMS) and identifying the type of the input working condition.

S2, acquiring the state of charge (SOC) of all the cells in the battery pack and the average SOC (SOC) of the battery pack under the working condition_mean) The state of charge difference Δ SOC therebetween.

S3, diagnosing according to the change characteristics of the normal monomer delta SOC and the abnormal monomer delta SOC under different working conditions, and diagnosing whether the reason of the abnormal change of the monomer delta SOC in the series battery pack is the initial SOC difference, or the micro short circuit, or the larger difference of the average capacities of the monomer and the battery pack.

In step S1, the manner of intercepting a section of the operating condition of the battery pack by the battery management system is as follows: a BMS begins to record the parameters of all the single batteries and the average parameter of the battery pack at a certain time to form SOC_mean-t-relation curve, SOC for this condition_mean-t relation curve is identified and classified according to SOC_meanAnd judging whether the intercepted working condition can be used as a working condition for effective diagnosis or not by the t relation curve, if not, finishing the diagnosis, and if so, continuing.

The operating conditions that can be effectively utilized are divided into 4, as shown in fig. 2-5, the first operating condition: SOC_meanT-curve fluctuation is less than threshold x1, which is 10% for this embodiment of threshold x 1; the second working condition is as follows: SOC_meanThe one-way fluctuation of the t curve is larger than a threshold value x2, and 40% is selected as the threshold value x2 in the embodiment; the third working condition is as follows: SOC_meanThe t-curve contains bi-directional undulations, wherein the two unidirectional segment undulations are greater than a threshold x3, which is selected to be 50% for this embodiment x 3; the fourth working condition: while including a combined condition of the first operating condition and the second operating condition.

Step S2 is to obtain all the cells in the battery packState of charge (SOC) and battery average SOC (SOC)_mean) The relation between delta SOC and t of each monomer is obtained in the whole working condition duration time, and the delta SOC change value lambda of each monomer under the working condition duration time is obtained (lambda is [ delta SOC ]]max－[ΔSOC]min), Δ SOC and SOC are also obtained_meanThe relationship (2) of (c).

The core of the invention is to diagnose the abnormal monomer according to the difference of the change characteristics of the normal monomer delta SOC and the abnormal monomer delta SOC under different working conditions.

The step S3 diagnosis method includes the following four methods:

1) in the first case, in which the battery is in a state of rest, or undergoes a very small degree of charge-discharge cycling, its SOC_mean-t curve fluctuation is less than 10%. If the delta SOC of a certain monomer changes along with time and the difference change value lambda is smaller than the threshold lambda 1, the diagnosis is quitted, otherwise, the monomer is judged to have a micro short circuit. The present embodiment selects the threshold λ 1 to be 4%.

In one embodiment of the present invention, as shown in fig. 6, in a series battery pack, in which a cell a has a micro short circuit, the capacity of the cell a is 100Ah, and the capacity of the other cells in the battery pack is also 100 Ah. SOC and SOC of initial state cell A_meanThe SOC of the single A is 90%, and the SOC of other single A without micro short circuit in the battery pack does not change along with the lapse of t, so that the SOC is not changed, and the battery pack has the advantages that the SOC is 100%, the single A with micro short circuit always consumes extra electric quantity in the standing process of the battery pack, 10Ah electric quantity is lost in the duration time of working conditions, and the SOC of the single A is 90%, and the SOC of other single A without micro short circuit in the battery pack_meanAbout 100%, the Δ SOC of the cell a changes from 0 in the initial state to 10%, the Δ SOC increases with time, and the Δ SOC change value λ of the cell increases (within a certain range) as the duration of the operating condition increases.

2) In the second condition, i.e. the battery is in the process of unidirectional charging or discharging, its SOC_mean-t curve undulation is greater than 40%. If the change value lambda of the delta SOC of a certain monomer is larger than the threshold value lambda 1, the monomer is indicated to have a micro short circuit or the monomer capacity is greatly different from the average capacity of the battery pack, but the judgment condition is insufficient, and only the monomer can be judged to have abnormality; if the variation value lambda of the single delta SOC is less than the threshold value lambda 1, and the delta SOC]max is smallAnd judging whether the single body is normal or not at the threshold value theta, and otherwise, judging that the initial SOC of the single body is greatly different from the average SOC of the battery pack. The threshold λ 1 is selected to be 4.5% in this embodiment, and the threshold θ is selected to be 3.5% in this embodiment.

In one embodiment of the present invention, as shown in fig. 7, in a series battery in which cell B has a micro short circuit, in a discharge process (second operation condition), the average capacity of the battery in the initial state and the capacity of cell B are both 100Ah, SOC_meanWhen the SOC of the monomer B is 0%, the SOC of the monomer B is 100%, theoretically 100Ah electricity is discharged, actually 90Ah is discharged, and 10Ah electricity is additionally consumed; in the same discharge time, since the other cells in the battery pack actually discharged 90Ah of electricity and remained 10Ah of electricity, the SOC was about 10%, Δ SOC of cell B was changed from 0 at the start of the discharge process to 10%, and Δ SOC increased as t increased.

In another embodiment of the present invention, as shown in FIG. 7, in a series-connected battery pack, the average capacity of the battery pack is 100Ah, wherein the capacity of the cell C is 90Ah smaller than the average capacity of the battery pack by 10Ah, and the SOC and SOC of the cell C in the initial state are determined during a discharging process (second operating condition)_meanEqually, in the same discharge time, the cell C and the other cells in the battery pack both discharged 90Ah of electricity, but because the cell C has a small capacity, the ratio of discharged electricity to total capacity is larger, the ratio of remaining electricity to total capacity is smaller, that is, the SOC of the cell C decreases more rapidly, at which time the SOC of the cell C is 0%, the SOC_meanApproximately 10%, Δ SOC changes from 0 at the beginning of the discharge process to 10%. It can be seen that Δ SOC does not change with the passage of t, but with SOC_meanGet smaller and bigger, SOC within duration of operating condition_meanThe larger the decrease, the larger (within a certain range) the Δ SOC variation λ of the monomer.

In summary, in the two embodiments, Δ SOC of the micro-short-circuit cell B and the capacity difference cell C is increased in one discharging process, and from the viewpoint of diagnosis, when a condition that a Δ SOC variation value λ of a certain cell is larger than a threshold value is found in the second working condition, it can only be determined that the cell is abnormal, and the cause of the abnormality cannot be specifically diagnosed as the micro-short-circuit or the capacity difference.

3) In a third operating mode, i.e. both the discharge and the charge of the battery pack, the SOC_mean-t-curve charge and discharge process fluctuation is more than 50%. If the variation value lambda of delta SOC of a certain monomer is larger than or equal to the threshold value lambda 1, and the delta SOC is along with the SOC_meanIf the battery pack capacity is changed, judging that the battery pack average capacity is different from the monomer capacity; otherwise, judging that the monomer has micro short circuit. If the variation value lambda of the single delta SOC is less than the threshold value lambda 1, and the delta SOC]And if max is smaller than the threshold value theta, judging that the single body is normal, otherwise, judging that the initial SOC of the single body is greatly different from the average SOC of the battery pack. In the present embodiment, the threshold λ 1 is selected to be 4.5%, and the threshold θ is selected to be 3.5%.

In one embodiment of the present invention, as shown in fig. 8, in a series battery in which the capacity of the cell D is 90Ah which is 10Ah smaller than the average cell capacity of the battery, the SOC and SOC of the cell D in the initial state_meanAre equal. Under the third working condition, in the discharging stage and in the same discharging time, the monomer D and other monomers in the battery pack both discharge 90Ah of electricity, but because the monomer D has small capacity, the ratio of the discharged electricity to the total capacity is larger, the ratio of the residual electricity to the total capacity is smaller, namely the SOC of the monomer D is reduced more quickly, the SOC of the monomer D is 0%, and the SOC is faster_meanAbout 10%, the Δ SOC variation λ of the monomer D is changed from 0 at the start of the discharge process to 10%; in the charging stage, under the same charging time, the monomer D and other monomers in the battery pack are charged with 90Ah electric quantity, and because the monomer D is small in capacity, the ratio of the charged electric quantity to the total capacity is larger than that of the other monomers in the battery pack, namely the SOC of the monomer D rises faster, at the moment, the SOC of the monomer D is 100 percent, and the SOC is 100 percent_meanAt 100%, the Δ SOC change value λ of the monomer D was changed from 10% to 0, and the initial state was restored. The delta SOC change value lambda is equal when the charging and discharging phases are finished. Under a third working condition comprising charging and discharging processes, the delta SOC of the monomer with capacity difference does not change along with the lapse of t, but follows the SOC_meanBecomes smaller and larger, and follows the SOC_meanBecoming larger and smaller.

In another embodiment of the present invention, as shown in fig. 8, a series battery pack in which a cell E has a micro short circuit, a cell E and other cells in the batteryThe capacities are all 100Ah, SOC and SOC of the initial state cell E_meanAll are 100%. Under the third working condition, in the discharging stage and in the same discharging time, the monomer E actually discharges 100Ah electric quantity, other monomers in the battery pack actually discharge 90Ah electric quantity, and because the monomer E has a micro short circuit, 10Ah electric quantity is additionally consumed, the SOC of the monomer E is 0%, and the SOC is 0%_meanAbout 10%, the delta SOC change value lambda of the single-phase E is changed from 0 in the initial state to 10%; in the charging stage, under the same charging time, other cells in the battery pack actually charge 90Ah electricity, theoretically, the electricity charged by the cell E is also 90Ah, but due to the fact that the cell E is slightly short-circuited, 10Ah electricity is additionally consumed, and 80Ah electricity is actually charged. At this time, the SOC of the monomer E was 80%, SOC_meanAbout 100%. In this process, the Δ SOC change λ of the monomer E changes from 10% to 20% in the process. The Δ SOC of a cell with a micro short circuit increases (within a certain range) with the lapse of t.

4) Under a fourth operating condition, i.e. a combined operating condition of the first operating condition and the second operating condition, the SOC thereof_meanThe t-curve includes both a battery rest phase with less than 10% fluctuation and a unidirectional charge or discharge phase with more than 40% fluctuation. If the change value lambda of the delta SOC of a certain monomer is larger than or equal to the threshold lambda 1 and the delta SOC changes along with t, judging that the monomer is slightly short-circuited; otherwise, judging that the monomer capacity is greatly different from the average capacity of the battery pack. If the variation value lambda of the single delta SOC is less than the threshold value lambda 1, and the delta SOC]And if max is smaller than the threshold value theta, judging that the single body is normal, otherwise, judging that the initial capacity of the single body is greatly different from the average capacity of the battery pack. In the present embodiment, the threshold λ 1 is selected to be 4.5%, and the threshold θ is selected to be 3.5%.

In one embodiment of the present invention, as shown in fig. 9, in a series battery pack, under a fourth operating condition, the capacity of the cell F is 90Ah, and the capacity of the other cells in the battery pack is 100 Ah. SOC and SOC of initial state cell F_meanAre all 100%. In the battery pack standing stage, the delta SOC of the monomer F is unchanged, and the delta SOC change value lambda is 0; in the discharging stage, due to the large capacity difference, the delta SOC of the monomer F follows the SOC_meanBecomes smaller and larger, and in the discharge stage, 80Ah is theoretically discharged, and the discharge stage is endedSOC of the body F was 11%, SOC_mean20% and the Δ SOC variation λ of the monomer F was 9%. The fourth operating condition can be one of the conditions that are effectively diagnosed because both the first and second operating conditions are included.

In one embodiment of the present invention, as shown in fig. 9, in a series battery pack, the capacity of a cell H is equal to the capacity of other cells in the battery pack, and is 100 Ah. SOC of initial State monomer H was 95%, SOC _mean100%, the initial delta SOC is 5%, and the delta SOC is basically unchanged in the whole working condition duration.

In an embodiment of the present invention, as shown in fig. 9, in a series battery pack, under a fourth operating condition, a micro short circuit exists in a certain cell I, and the capacity of the cell I is equal to that of other cells in the battery pack, and is 100 Ah. In the standing stage, the monomer I with the micro short circuit always has extra electric quantity consumption, 10Ah electric quantity is lost in the standing working condition duration, the SOC of the monomer A is 90%, and the SOC of other monomers without the micro short circuit in the battery pack does not change along with the lapse of t, so the SOC_meanAbout 100%, Δ SOC of the monomer I becomes 10% from 0 at the initial state; in the discharging stage, theoretically 90Ah of electricity is discharged, actually 80Ah is discharged, 10Ah of electricity is additionally consumed, and the SOC of the monomer I is 0%. In the same discharge time, the other cells in the battery pack actually discharged 80Ah of electricity, and 20Ah of electricity remained, so the SOC was about 20%, and the Δ SOC of cell I was changed from 10% to 20% at the start of the discharge process. The delta SOC change value lambda of the whole working condition monomer I is 20 percent.

According to the above process, the present invention can effectively diagnose whether the cells with abnormal Δ SOC variation in the series battery pack have initial state of charge (SOC) variation or have micro short circuit or have cell average capacity variation without changing the prior art.

Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made in the above embodiments by those of ordinary skill in the art without departing from the principle and spirit of the present invention.

Claims

1. The online fault diagnosis method for the battery pack is used for judging the reason of large difference between the single battery charge state and the average battery charge state, and comprises the following steps of:

s1, intercepting a section of working condition of the battery pack, and identifying the type of the working condition based on the working condition, wherein the working condition is represented by a curve of the average state of charge of the battery pack changing along with time, namely SOC_mean-t-curve characterization, the operating condition types including:

SOC_mean-a fourth condition characterized by a t-curve having both the first condition and the second condition;

s2, acquiring the state of charge difference delta SOC between the state of charge of all battery monomers in the battery pack and the average state of charge of the battery pack under the working condition, and acquiring difference change characteristics including the time-varying trend of the delta SOC and a difference change value lambda;

s3, determining the reason that the difference between the battery cell charge state and the average charge state of the battery pack is large because of the initial SOC difference, the micro short circuit or the cell average capacity difference based on the working condition type obtained in the step S1 and the difference change characteristics obtained in the step S2, specifically:

2. The online battery pack fault diagnosis method according to claim 1, wherein in step S2, the difference change value λ is calculated by the formula:

λ＝[ΔSOC]max－[ΔSOC]min

3. The online battery pack fault diagnosis method according to claim 2, wherein in step S2, the state of charge difference is calculated by the following formula:

ΔSOC＝SOC_mean-SOC

4. The online fault diagnosis method for a battery pack according to claim 1, wherein the difference in the average cell capacity in step S3 is a difference between the average cell capacity and the average battery pack capacity.