CN106526493A - Power battery external short circuit fault diagnosing and temperature rise prediction method and system based on BP neural networks - Google Patents
Power battery external short circuit fault diagnosing and temperature rise prediction method and system based on BP neural networks Download PDFInfo
- Publication number
- CN106526493A CN106526493A CN201610972484.5A CN201610972484A CN106526493A CN 106526493 A CN106526493 A CN 106526493A CN 201610972484 A CN201610972484 A CN 201610972484A CN 106526493 A CN106526493 A CN 106526493A
- Authority
- CN
- China
- Prior art keywords
- neural network
- battery
- short circuit
- external short
- host computer
- 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.)
- Granted
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 139
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000003745 diagnosis Methods 0.000 claims abstract description 19
- 238000012360 testing method Methods 0.000 claims abstract description 15
- 230000007246 mechanism Effects 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims description 26
- 230000005611 electricity Effects 0.000 claims description 25
- 230000008569 process Effects 0.000 claims description 10
- 238000012544 monitoring process Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 208000032953 Device battery issue Diseases 0.000 claims description 6
- 210000004218 nerve net Anatomy 0.000 claims description 5
- 230000007935 neutral effect Effects 0.000 claims description 4
- 230000000630 rising effect Effects 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims 1
- 239000007788 liquid Substances 0.000 abstract description 3
- 230000004888 barrier function Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 206010000369 Accident Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The invention provides a power battery external short circuit fault diagnosing and temperature rise prediction method and system based on BP neural networks. Through using two BP neural networks, firstly a sample frequency, a current threshold and a power threshold are initialized, when a real-time monitored current signal exceeds a current threshold, a fault diagnosis mechanism is started, the current signal is collected and stored and short circuit fault discharge power is calculated, through comparing the relation between the discharge power and a power threshold, the liquid leakage in a battery short circuit fault is diagnosed, according to a diagnosis result, a corresponding BP neural network which is established through a test in advance and is trained is selected, the battery discharge power after a fault is inputted into the network, thus the maximum temperature rise caused by a short circuit fault is predicted, and a control basis is provided for the early intervention of a thermal management system. The method is easy to realize and operate, the thermal management system can be assisted to reduce trigger probability of thermal runaway, and the safety protection performance of a power battery external short circuit fault is improved.
Description
Technical field
The present invention relates to the lithium-ion-power cell short circuit of electrokinetic cell security technology area, more particularly to electric automobile
Temperature prediction technology in the safety management of failure.
Background technology
Currently, with further genralrlization and the application of electric automobile, its some safety problem also gradually appears, such as certainly
The accidents such as combustion, blast, thermal runaway occur again and again.Data shows that a big chunk electric automobile fire accident is by onboard power
What battery failures were caused.Therefore, research and safety and reliability of the failure for raising electric automobile for predicting electrokinetic cell
Property tool be of great significance.
In the middle of numerous battery failures forms, external short-circuit of battery is a class failure the most serious.Because short-circuit mistake
In journey, inside battery chemism is destroyed rapidly, and in the short time, electric current is increased dramatically, and substantial amounts of heat is piled up in inside battery, battery
Temperature rises rapidly, after temperature is increased to a marginal value, will cause thermal runaway.Therefore, Accurate Diagnosis battery short circuit
Maximum temperature of the failure condition further to being likely to occur is predicted, and is security protection and the important content in heat management system.
How the maximum temperature that Accurate Prediction short circuit is likely to occur, is the problem of a urgent need to resolve.
The content of the invention
For technical problem present in above-mentioned this area, the invention provides a kind of power current based on BP neural network
Pond external short circuit fault diagnosis and temperature prediction method, specifically include following steps：
A is walked, and is opened host computer and is initialized sample frequency f, current threshold Is, power threshold Cs；
B is walked, and host computer passes through current sensor real-time monitoring current signal I, if I<Is, the host computer continues through
Current sensor real-time monitoring current signal, repeats b steps, if I >=Is, trigger external short-circuit of battery fault diagnosis and maximum temperature
Forecasting mechanism is risen, is walked into c；
C is walked, and the host computer is according to sample frequency f in tiMoment collection stored current signal Ii, calculate outer
The electricity C discharged by portion's short circuit, calculation relational expression are as follows：
Wherein, N is the number of samples after there is external short circuit.
D is walked, and diagnoses whether the external short circuit causes battery leakage, if C >=Cs, battery is diagnosed as not yet to be occurred to leak
Result is shown in the host computer interface, and is walked into e, if C by liquid<Cs, then it is diagnosed as leakage, result is shown
In the host computer interface, and walk into f；
E is walked, neural network 1 process, and the host computer is input into the c electricity C that calculate of step to pre-building
And in the BP neural network 1 for training, the output of the BP neural network 1 being drawn, the output is external short-circuit of battery event
The predictive value Δ T of barrier maximum temperature risemax。
F is walked, and neutral net 2 processes, and the host computer is input into the electricity C that c steps are calculated to pre-building
And in the BP neural network 2 for training, the output of the BP neural network 2 being drawn, the output is external short-circuit of battery event
The predictive value Δ T of barrier maximum temperature risemax。
Further, the foundation of the BP neural network 1 and BP neural network 2 and training process are specifically included：
(1). determine the training sample of the BP neural network 1 and BP neural network 2；
(2). set up the BP neural network 1 and BP neural network 2；
(3). respectively the BP neural network 1 and BP neural network 2 are trained；
(4). determine the optimal BP neural network 1 and BP neural network 2.
Wherein, in the training sample for determining the BP neural network 1 and BP neural network 2, also to varying environment temperature
Battery under the lower and different state-of-charges of degree carries out external short circuit test, specifically includes：To the battery under different state-of-charges
After initialization, respectively to each triggering external short circuit；The current sensor gathers the electricity at each moment based on the sample frequency
Stream, obtains the electricity C discharged by external short circuit according to the formula (1)；Meanwhile, by temperature sensor based on the sampling frequency
Rate gathers battery plus-negative plate surface temperature T at the i-th moment1iAnd T2i, and it is based on test environment temperature T0Obtain the battery positive and negative
The temperature rise Δ T of pole1i=T1i-T0, Δ T2i=T2i-T0, try to achieve the average temperature rising maximum Δ T of the battery plus-negative platemax=max
[(ΔT1i+ΔT2i)/2] as the maximum temperature rise of external short-circuit of battery；The battery tested is divided into according to test result and not sent out
Raw leakage and generation two class of leakage.
Further, gather j and the electricity C discharged by the battery of leakage does not occur1jWith corresponding maximum temperature rise Δ
Tmax_1j, and the electricity C discharged by the battery of k generation leakage2kWith corresponding maximum temperature rise Δ Tmax_2kIt is refreshing as the BP
The training sample of Jing networks 1 and BP neural network 2.
Further, the step (2) set up the BP neural network 1 and BP neural network 2 is specifically included：Institute is set
It is the three-decker being made up of input layer, hidden layer and output layer to state BP neural network 1 and BP neural network 2.Input layer
Number n1For 1, output layer nodes n3For 1, rule of thumb formula determines the node in hidden layer n of the BP neural network2, calculate
Relational expression is as follows：
Wherein, the integer of α desirable 1～10, for n2Numerical value, lower bound round up, and the upper bound rounds downwards.
Further, the step (3) specifically includes:According to the training sample determined by the step (1), take 1 to α～
In 10, the BP neural network 1 and BP neural network 2 of different node in hidden layer corresponding during different integers is trained respectively,
Obtain the BP neural network 1 and BP neural network 2 of each different node in hidden layer for having trained.
Further, the step (4) determine the optimal BP neural network 1 and BP neural network 2 is concrete wraps
Include：In the BP neural network 1 and BP neural network 2 of the different node in hidden layer from obtained by step (3), hidden layer section is selected
Points n2It is minimum and meet the BP neural network of training error requirement respectively as optimal BP neural network 1 and BP nerve net
Network 2.
Present invention also offers a kind of be used to perform above-mentioned electrokinetic cell external short circuit fault diagnosis and temperature prediction method
Electrokinetic cell external short circuit fault diagnosis and temperature prediction system, including：Host computer, neural network module, current sensor,
Temperature sensor, electrokinetic cell, actual motion load and CAN.
Electrokinetic cell external short circuit fault diagnosis provided by the present invention and temperature prediction method and system, realize and operate
It is convenient, short circuit and the leakage situation that battery can be diagnosed in the starting stage that short circuit occurs, before not yet there is high temperature can be played,
And Accurate Prediction short circuit is by the technique effect of the maximum temperature rise for causing, while can be the protection of electrokinetic cell external short circuit failure
And the good basis of offer is further provided.
Description of the drawings
Electrokinetic cell external short circuit fault diagnosis and temperature prediction system diagram of the accompanying drawing 1 for the present invention,
Electrokinetic cell external short circuit fault diagnosis and temperature prediction method flow diagram of the accompanying drawing 2 for the present invention,
Accompanying drawing 3 is the short-circuit test flow process in 2 training sample of determination BP neural network 1 and BP neural network of the present invention
Figure,
BP neural network 1 and BP neural network 2 structure chart of the accompanying drawing 4 for the present invention.
Specific embodiment
As shown in figure 1, electrokinetic cell external short circuit fault diagnosis provided by the present invention and temperature prediction system include：On
Position machine, neural network module, current sensor, temperature sensor (not shown), electrokinetic cell, actual motion load and CAN
Bus.
Fig. 2 show a kind of electrokinetic cell external short circuit fault diagnosis based on BP neural network provided by the present invention and
Temperature prediction method, specifically includes following steps：
A is walked, and is opened host computer and is initialized sample frequency f, current threshold Is, power threshold Cs；
B is walked, and host computer passes through current sensor real-time monitoring current signal I, if I<Is, the host computer continues through
Current sensor real-time monitoring current signal, repeats b steps, if I >=Is, trigger external short-circuit of battery fault diagnosis and maximum temperature
Forecasting mechanism is risen, is walked into c；
C is walked, and the host computer is according to sample frequency f in tiMoment collection stored current signal Ii, calculate outer
The electricity C discharged by portion's short circuit, calculation relational expression are as follows：
Wherein, N is the number of samples after there is external short circuit.
D is walked, and diagnoses whether the external short circuit causes battery leakage, if C >=Cs, battery is diagnosed as not yet to be occurred to leak
Result is shown in the host computer interface, and is walked into e, if C by liquid<Cs, then it is diagnosed as leakage, result is shown
In the host computer interface, and walk into f；
E is walked, neural network 1 process, and the host computer is input into the c electricity C that calculate of step to pre-building
And in the BP neural network 1 for training, the output of the BP neural network 1 being drawn, the output is external short-circuit of battery event
The predictive value Δ T of barrier maximum temperature risemax。
F is walked, and neutral net 2 processes, and the host computer is input into the electricity C that c steps are calculated to pre-building
And in the BP neural network 2 for training, the output of the BP neural network 2 being drawn, the output is external short-circuit of battery event
The predictive value Δ T of barrier maximum temperature risemax。
In a preferred embodiment of the application, the foundation of the BP neural network 1 and BP neural network 2 and trained
Journey is specifically included：
(1). determine the training sample of the BP neural network 1 and BP neural network 2；
(2). set up the BP neural network 1 and BP neural network 2；
(3). respectively the BP neural network 1 and BP neural network 2 are trained；
(4). determine the optimal BP neural network 1 and BP neural network 2.
In a preferred embodiment of the application, the training for determining the BP neural network 1 and BP neural network 2
In sample, external short circuit test is carried out to the battery at a temperature of varying environment and under different state-of-charges also, is specifically included：Such as
Shown in Fig. 3, after initializing to the battery under different state-of-charges, respectively to each triggering external short circuit；The current sensor base
The electric current at each moment is gathered in the sample frequency, the electricity C discharged by external short circuit is obtained according to the formula (1)；Together
When, battery plus-negative plate surface temperature T at the i-th moment is gathered by temperature sensor based on the sample frequency1iAnd T2i, and base
In test environment temperature T0Obtain the temperature rise Δ T of the battery plus-negative plate1i=T1i-T0, Δ T2i=T2i-T0, try to achieve the battery
The average temperature rising maximum Δ T of both positive and negative polaritymax=max [(Δ T1i+ΔT2i)/2] as the maximum temperature rise of external short-circuit of battery；Root
The electricity C discharged according to the external short circuit and maximum temperature rise Δ TmaxThe external short-circuit of battery failure is divided into and is not sent out
Raw leakage and generation two class of leakage.
In a preferred embodiment of the application, there is no the electricity C discharged by the battery of leakage in collection j1jWith it is right
The maximum temperature rise Δ T for answeringmax_1j, and the electricity C discharged by the battery of k generation leakage2kWith corresponding maximum temperature rise Δ
Tmax_2kAs the BP neural network 1 and the training sample of BP neural network 2.
In a preferred embodiment of the application, the step (2) sets up the BP neural network 1 and BP nerve net
Network 2 is specifically included：It is three be made up of input layer, hidden layer and output layer to arrange the BP neural network 1 and BP neural network 2
Rotating fields.Input layer number n1For 1, output layer nodes n3For 1, rule of thumb formula determines the hidden of the BP neural network
The n of number containing node layer2, calculation relational expression is as follows：
Wherein, the integer of α desirable 1～10, for n2Numerical value, lower bound round up, and the upper bound rounds downwards.
In a preferred embodiment of the application, the step (3) specifically includes:According to by the step (1) determination
Training sample, the BP neural network 1 and BP that different node in hidden layer corresponding during different integers are taken in 1～10 to α be refreshing
Jing networks 2 are trained respectively, obtain the BP neural network 1 and BP nerve net of each different node in hidden layer for having trained
Network 2.
In a preferred embodiment of the application, the step (4) determines optimal 1 He of the BP neural network
BP neural network 2 is specifically included：The BP neural network 1 and BP nerve net of the different node in hidden layer from obtained by step (3)
In network 2, node in hidden layer n is selected2It is minimum and meet the BP neural network of training error requirement respectively as optimal BP
Neural network 1 and BP neural network 2.
The present invention is described in detail with reference to embodiments, and the scope of technical scheme is not limited to following
Specific embodiment.
Step one, short-circuit test determine the training sample of BP neural network 1 and BP neural network 2：By in varying environment
At a temperature of respectively the battery of different state-of-charges (SoC) is carried out described in Fig. 3 battery short circuit test, so that it is determined that training sample
This.2 Battery packs are selected to carry out short-circuit test in the present embodiment, each group includes 10 pieces of tested batteries, according to battery charge state
And different ambient temperature arranges that to give every piece of different numbering of battery as shown in table 1：
1 battery of table is numbered
The result for testing battery is carried out reclassifying numbering according to the leakage situation of short-circuit test, as shown in table 2：
Table 2 reclassifies numbering based on short-circuit test result to battery
Step 2, sets up BP neural network 1 and BP neural network 2：The BP neural network set up is three-decker, defeated
Enter node layer number n1For 1, output layer nodes n3For 1, the node in hidden layer of the BP neural network is determined according to formula (2)
n2, n2Desirable 3～11 integer；
Step 3, is respectively trained BP neural network 1 and BP neural network 2：With the C obtained in step oneN1～CN15And CL1
～CL5Respectively as the input of BP neural network 1 and BP neural network 2, and the Δ T to obtain in step onemax_N1～Δ
Tmax_N15With Δ Tmax_L1～Δ Tmax_L5Respectively as the output of BP neural network 1 and BP neural network 2, training sample is built,
To n2When taking different integers in 3～11, corresponding BP neural network 1 and BP neural network 2 are trained respectively, obtain each
The BP neural network 1 and BP neural network 2 of the different node in hidden layer for having trained, the present embodiment adopts levenberg-
Marquardt learning algorithms, Hessian matrixes with approximate representation can be：
H=JTJ (7)
The calculation expression of gradient is：
G=JTe (8)
Wherein, H is the Jacobian matrix comprising network error function pair weights and threshold value first derivative, and J is Jacobi square
Battle array, e is the error vector of network.
The iterative calculation formula of training is as follows：
X (k+1)=x (k)-[JTJ+μI]-1JTe (9)
Wherein, k is iterationses, and x (k) is the connection weight vector or threshold vector between the kth time each layer of iteration, and I is single
Bit matrix, coefficient μ (error performance reduction) after successful iteration each time are reduced, error after tentative iteration is carried out
Can increase in the case of increasing；
Step 4, determines optimal BP neural network 1 and BP neural network 2 respectively：Difference from obtained by step 3
In the BP neural network 1 and BP neural network 2 of node in hidden layer, node in hidden layer n is selected2It is minimum and meet training and miss
The BP neural network that difference is required is respectively as optimal BP neural network 1 and BP neural network 2, for the present embodiment, optimal
N in BP neural network 129 are taken, the n in optimal BP neural network 22Take 6；
Step 5, opens host computer initiation parameter, and in the present embodiment, sample frequency f takes 20Hz, according to step one
Short-circuit test, current threshold IsTake 50A, power threshold CsTake 0.12Ah；
Step 6, host computer pass through current sensor real-time monitoring current signal I, if I<Is, host computer continues through electricity
Flow sensor real-time monitoring current signal, repeat step six, if I >=Is, trigger external short-circuit of battery fault diagnosis and maximum temperature
Forecasting mechanism is risen, into step 7；
Step 7, host computer is according to sample frequency f in tiMoment collection stored current signal Ii, counted according to formula (1)
Calculate the electricity C discharged by short circuit；
Whether step 8, diagnosis short circuit cause battery leakage：If C >=Cs, battery is diagnosed as not yet leakage, and incites somebody to action
As a result host computer interface is shown in, into step 9, if C<Cs, battery is diagnosed as leakage, and result is shown in
Position machine interface, into step 10；
Step 9, neural network 1 process, host computer are input into the electricity C that step 7 calculates gained to pre-building and instruct
In the BP neural network 1 perfected, the output of BP neural network 1 is drawn, the output is the battery short circuit failure maximum temperature rise
Predictive value Δ Tmax。
Step 10, neutral net 2 are processed, and host computer is input into the electricity C that step 7 calculates gained to pre-building and instruct
In the BP neural network 2 perfected, the output of BP neural network 2 is drawn, the output is the battery short circuit failure maximum temperature rise
Predictive value Δ Tmax。
Although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
Understanding can carry out various changes, modification, replacement to these embodiments without departing from the principles and spirit of the present invention
And modification, the scope of the present invention be defined by the appended.
Claims (8)
1. a kind of electrokinetic cell external short circuit fault diagnosis and temperature prediction method based on BP neural network, it is characterised in that：
Specifically include following steps：
A is walked, and is opened host computer and is initialized sample frequency f, current threshold Is, power threshold Cs；
B is walked, and host computer passes through current sensor real-time monitoring current signal I, if I<Is, the host computer continues through electric current
Sensor real-time monitoring current signal, repeats b steps, if I >=Is, trigger external short-circuit of battery fault diagnosis and maximum temperature rise be pre-
Survey mechanism, walks into c；
C is walked, and the host computer is according to sample frequency f in tiMoment collection stored current signal Ii, calculate outside short
The electricity C discharged by road, calculation relational expression are as follows：
Wherein, N is the number of samples after there is external short circuit.
D is walked, and diagnoses whether the external short circuit causes battery leakage, if C >=Cs, battery is diagnosed as not yet leakage, will
As a result the host computer interface is shown in, and is walked into e, if C<Cs, then it is diagnosed as leakage, result is shown in described
Host computer interface, and walk into f；
E is walked, neural network 1 process, and the host computer is input into the c electricity C that calculate of step to pre-building and instruct
In the BP neural network 1 perfected, the output of the BP neural network 1 is drawn, the output is the external short-circuit of battery failure most
The predictive value Δ T of big temperature risemax。
F is walked, and neutral net 2 processes, and the host computer is input into the electricity C that c steps are calculated to pre-building and instruct
In the BP neural network 2 perfected, the output of the BP neural network 2 is drawn, the output is the external short-circuit of battery failure most
The predictive value Δ T of big temperature risemax。
2. the method for claim 1, it is characterised in that：The foundation of the BP neural network 1 and BP neural network 2 and instruction
Practice process to specifically include：
(1). determine the training sample of the BP neural network 1 and BP neural network 2；
(2). set up the BP neural network 1 and BP neural network 2；
(3). respectively the BP neural network 1 and BP neural network 2 are trained；
(4). determine the optimal BP neural network 1 and BP neural network 2.
3. method as claimed in claim 2, it is characterised in that：Described in the determination of the step (1), BP neural network 1 and BP are refreshing
In the training sample of Jing networks 2, external short circuit survey is carried out to the battery at a temperature of varying environment and under different state-of-charges also
Examination, specifically includes：After to the battery initialization under different state-of-charges, respectively to each triggering external short circuit；The current sense
Device gathers the electric current at each moment based on the sample frequency, obtains the electricity discharged by external short circuit according to the formula (1)
C；Meanwhile, battery plus-negative plate surface temperature T at the i-th moment is gathered by temperature sensor based on the sample frequency1iAnd T2i,
And it is based on test environment temperature T0Obtain the temperature rise Δ T of the battery plus-negative plate1i=T1i-T0, Δ T2i=T2i-T0, try to achieve described
The average temperature rising maximum Δ T of battery plus-negative platemax=max [(Δ T1i+ΔT2i)/2] as the maximum temperature of external short-circuit of battery
Rise；The electricity C discharged according to the external short circuit and maximum temperature rise Δ TmaxThe external short-circuit of battery failure is divided into
Generation leakage and generation two class of leakage.
4. the method as described in any one of Claims 2 or 3, it is characterised in that：J battery that leakage does not occur of collection is discharged
Electricity C1jWith corresponding maximum temperature rise Δ Tmax_1j, and the electricity C discharged by the battery of k generation leakage2kWith it is corresponding
Maximum temperature rise Δ Tmax_2k, as the BP neural network 1 and the training sample of BP neural network 2.
5. method as claimed in claim 4, it is characterised in that：The step (2) set up the BP neural network 1 and BP is refreshing
Jing networks 2 are specifically included：It is to be made up of input layer, hidden layer and output layer to arrange the BP neural network 1 and BP neural network 2
Three-decker.Input layer number n1For 1, output layer nodes n3For 1, rule of thumb formula determines the BP neural network
Node in hidden layer n2, calculation relational expression is as follows：
Wherein, α takes 1～10 integer, for n2Numerical value, lower bound round up, and the upper bound rounds downwards.
6. method as claimed in claim 5, it is characterised in that：The step (3) specifically includes:According to by the step (1)
It is determined that training sample, 1 He of BP neural network of different node in hidden layer corresponding during different integers is taken in 1～10 to α
BP neural network 2 is trained respectively, and the BP neural network 1 and BP for obtaining each different node in hidden layer for having trained is refreshing
Jing networks 2.
7. method as claimed in claim 6, it is characterised in that：The step (4) determines the optimal BP nerve net
Network 1 and BP neural network 2 are specifically included：The BP neural network 1 and BP of the different node in hidden layer obtained from step (3) is neural
In network 2, node in hidden layer n is selected2BP neural network that is minimum and meeting training error requirement, respectively as optimal
BP neural network 1 and BP neural network 2.
8. a kind of electrokinetic cell external short circuit fault diagnosis and temperature for performing method any one of aforementioned claim
Rise prognoses system, it is characterised in that：Including：Host computer, neural network module, current sensor, temperature sensor, power current
Pond, actual motion load and CAN.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610972484.5A CN106526493B (en) | 2016-11-01 | 2016-11-01 | Power battery external short circuit fault diagnosis and temperature prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610972484.5A CN106526493B (en) | 2016-11-01 | 2016-11-01 | Power battery external short circuit fault diagnosis and temperature prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106526493A true CN106526493A (en) | 2017-03-22 |
| CN106526493B CN106526493B (en) | 2018-11-23 |
Family
ID=58349757
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610972484.5A Active CN106526493B (en) | 2016-11-01 | 2016-11-01 | Power battery external short circuit fault diagnosis and temperature prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106526493B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107192956A (en) * | 2017-05-19 | 2017-09-22 | 北京理工大学 | A kind of battery short circuit leakage on-line monitoring method and device |
| CN107490441A (en) * | 2017-10-16 | 2017-12-19 | 广州贯行电能技术有限公司 | The determination methods and device of the electric reasonable temperature rise of cabinet caused by a kind of electric current |
| CN108363016A (en) * | 2018-02-22 | 2018-08-03 | 上海理工大学 | Battery micro-short circuit quantitative Diagnosis method based on artificial neural network |
| CN108693478A (en) * | 2018-04-17 | 2018-10-23 | 北京理工大学 | A kind of method for detecting leakage of lithium-ion-power cell |
| CN108919133A (en) * | 2018-07-27 | 2018-11-30 | 东北大学 | Electric automobile power battery group external short circuit electronic-hydraulic complex control testing stand and method |
| CN109738811A (en) * | 2019-01-28 | 2019-05-10 | 东北大学 | Li-ion batteries piles external short circuit method for diagnosing faults based on twin-stage model prediction |
| CN109755683A (en) * | 2018-12-04 | 2019-05-14 | 厦门大学 | A kind of battery pack internal temperature method of real-time based on compressive sensing theory |
| CN110308397A (en) * | 2019-07-30 | 2019-10-08 | 重庆邮电大学 | A kind of lithium battery multiclass fault diagnosis modeling method of mixing convolutional neural networks driving |
| CN110426637A (en) * | 2019-07-04 | 2019-11-08 | 佛山科学技术学院 | A kind of battery failures diagnostic method neural network based and device |
| CN111307480A (en) * | 2020-02-20 | 2020-06-19 | 吉林建筑大学 | Embedded heat pipe-based heat transfer management system, method and storage medium |
| US10948544B2 (en) | 2018-03-08 | 2021-03-16 | Industrial Technology Research Institute | Battery safety identifying method and method for setting hazard levels of battery internal short circuit and warning system using the same |
| CN113343633A (en) * | 2021-06-10 | 2021-09-03 | 上海交通大学 | Thermal runaway fault classification and risk prediction method and system for power lithium battery |
| CN113659246A (en) * | 2021-10-20 | 2021-11-16 | 中国气象科学研究院 | Battery system suitable for polar region ultralow temperature environment and temperature control method thereof |
| CN116429339A (en) * | 2023-06-07 | 2023-07-14 | 海斯坦普汽车组件(北京)有限公司 | Leakage detection method and system for new energy battery box |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010249802A (en) * | 2009-03-26 | 2010-11-04 | Daikin Ind Ltd | Opening/short circuit inspection method of external terminal in integrated circuit, and opening/short circuit inspection device of external terminal in integrated circuit |
| CN102494778A (en) * | 2011-11-14 | 2012-06-13 | 北京理工大学 | Artificial neural network-based highest surface temperature prediction method of secondary battery |
| JP2013145175A (en) * | 2012-01-13 | 2013-07-25 | Toyota Motor Corp | Battery system and short circuit detection method |
| JP2014022282A (en) * | 2012-07-20 | 2014-02-03 | Sharp Corp | Secondary battery abnormality detector, secondary battery, and method for detecting secondary battery abnormality |
| JP2015169483A (en) * | 2014-03-05 | 2015-09-28 | 三菱自動車工業株式会社 | Secondary battery abnormality determination device |
| CN105954685A (en) * | 2016-05-30 | 2016-09-21 | 东北大学 | Electric vehicle battery short circuit test bench with remote control function, and method |
-
2016
- 2016-11-01 CN CN201610972484.5A patent/CN106526493B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010249802A (en) * | 2009-03-26 | 2010-11-04 | Daikin Ind Ltd | Opening/short circuit inspection method of external terminal in integrated circuit, and opening/short circuit inspection device of external terminal in integrated circuit |
| CN102494778A (en) * | 2011-11-14 | 2012-06-13 | 北京理工大学 | Artificial neural network-based highest surface temperature prediction method of secondary battery |
| JP2013145175A (en) * | 2012-01-13 | 2013-07-25 | Toyota Motor Corp | Battery system and short circuit detection method |
| JP2014022282A (en) * | 2012-07-20 | 2014-02-03 | Sharp Corp | Secondary battery abnormality detector, secondary battery, and method for detecting secondary battery abnormality |
| JP2015169483A (en) * | 2014-03-05 | 2015-09-28 | 三菱自動車工業株式会社 | Secondary battery abnormality determination device |
| CN105954685A (en) * | 2016-05-30 | 2016-09-21 | 东北大学 | Electric vehicle battery short circuit test bench with remote control function, and method |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107192956A (en) * | 2017-05-19 | 2017-09-22 | 北京理工大学 | A kind of battery short circuit leakage on-line monitoring method and device |
| CN107490441A (en) * | 2017-10-16 | 2017-12-19 | 广州贯行电能技术有限公司 | The determination methods and device of the electric reasonable temperature rise of cabinet caused by a kind of electric current |
| CN107490441B (en) * | 2017-10-16 | 2020-12-04 | 广州贯行电能技术有限公司 | Method and device for judging reasonable temperature rise of electric cabinet caused by current |
| CN108363016A (en) * | 2018-02-22 | 2018-08-03 | 上海理工大学 | Battery micro-short circuit quantitative Diagnosis method based on artificial neural network |
| US10948544B2 (en) | 2018-03-08 | 2021-03-16 | Industrial Technology Research Institute | Battery safety identifying method and method for setting hazard levels of battery internal short circuit and warning system using the same |
| CN108693478A (en) * | 2018-04-17 | 2018-10-23 | 北京理工大学 | A kind of method for detecting leakage of lithium-ion-power cell |
| WO2020019707A1 (en) * | 2018-07-27 | 2020-01-30 | 东北大学 | Electric automobile power battery pack external short-circuit electro-hydraulic composite control test platform and method |
| CN108919133A (en) * | 2018-07-27 | 2018-11-30 | 东北大学 | Electric automobile power battery group external short circuit electronic-hydraulic complex control testing stand and method |
| CN109755683A (en) * | 2018-12-04 | 2019-05-14 | 厦门大学 | A kind of battery pack internal temperature method of real-time based on compressive sensing theory |
| CN109738811A (en) * | 2019-01-28 | 2019-05-10 | 东北大学 | Li-ion batteries piles external short circuit method for diagnosing faults based on twin-stage model prediction |
| WO2020155233A1 (en) * | 2019-01-28 | 2020-08-06 | 东北大学 | Method for diagnosing external short circuit fault of lithium-ion battery pack on basis of two-stage model prediction |
| CN109738811B (en) * | 2019-01-28 | 2020-12-01 | 东北大学 | External short circuit fault diagnosis method of lithium ion battery pack based on two-stage model prediction |
| CN110426637A (en) * | 2019-07-04 | 2019-11-08 | 佛山科学技术学院 | A kind of battery failures diagnostic method neural network based and device |
| CN110308397B (en) * | 2019-07-30 | 2021-04-02 | 重庆邮电大学 | Lithium battery multi-class fault diagnosis modeling method driven by hybrid convolutional neural network |
| CN110308397A (en) * | 2019-07-30 | 2019-10-08 | 重庆邮电大学 | A kind of lithium battery multiclass fault diagnosis modeling method of mixing convolutional neural networks driving |
| CN111307480A (en) * | 2020-02-20 | 2020-06-19 | 吉林建筑大学 | Embedded heat pipe-based heat transfer management system, method and storage medium |
| CN111307480B (en) * | 2020-02-20 | 2022-03-04 | 吉林建筑大学 | Embedded heat pipe-based heat transfer management system, method and storage medium |
| CN113343633A (en) * | 2021-06-10 | 2021-09-03 | 上海交通大学 | Thermal runaway fault classification and risk prediction method and system for power lithium battery |
| CN113343633B (en) * | 2021-06-10 | 2022-04-26 | 上海交通大学 | Thermal runaway fault classification and risk prediction method and system for power lithium battery |
| CN113659246A (en) * | 2021-10-20 | 2021-11-16 | 中国气象科学研究院 | Battery system suitable for polar region ultralow temperature environment and temperature control method thereof |
| CN116429339A (en) * | 2023-06-07 | 2023-07-14 | 海斯坦普汽车组件(北京)有限公司 | Leakage detection method and system for new energy battery box |
| CN116429339B (en) * | 2023-06-07 | 2023-09-01 | 海斯坦普汽车组件(北京)有限公司 | Leakage detection method and system for new energy battery box |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106526493B (en) | 2018-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106526493A (en) | Power battery external short circuit fault diagnosing and temperature rise prediction method and system based on BP neural networks | |
| Yang et al. | Characterization of external short circuit faults in electric vehicle Li-ion battery packs and prediction using artificial neural networks | |
| Finegan et al. | The application of data-driven methods and physics-based learning for improving battery safety | |
| Chen et al. | Model-based fault diagnosis approach on external short circuit of lithium-ion battery used in electric vehicles | |
| Yu et al. | Challenges and outlook for lithium-ion battery fault diagnosis methods from the laboratory to real world applications | |
| KR102439041B1 (en) | Method and apparatus for diagnosing defect of battery cell based on neural network | |
| CN107192956B (en) | A kind of battery short circuit leakage on-line monitoring method and device | |
| Xu et al. | Multi-scale short circuit resistance estimation method for series connected battery strings | |
| CN107843853A (en) | A kind of power battery pack is connected in series method for diagnosing faults | |
| CN106353690A (en) | Method for diagnosing lithium battery faults by Petri net | |
| CN108037463A (en) | A kind of lithium ion battery life-span prediction method | |
| CN107367698A (en) | The health status Forecasting Methodology of electric automobile lithium battery group | |
| CN115032542A (en) | Hybrid model-based battery thermal runaway pre-judgment method for energy storage system | |
| CN108490357A (en) | Lithium battery residual capacity prediction technique based on mechanism-data-driven model | |
| Zhang et al. | A multi-fault diagnosis method for lithium-ion battery pack using curvilinear Manhattan distance evaluation and voltage difference analysis | |
| CN113013514B (en) | Thermal runaway gas-sensitive alarm device of vehicle-mounted lithium ion power battery and detection method thereof | |
| Rui-hao et al. | Battery state of charge estimation for electric vehicle based on neural network | |
| CN108363016A (en) | Battery micro-short circuit quantitative Diagnosis method based on artificial neural network | |
| WO2022242058A1 (en) | Battery state of health estimation method for real new energy vehicle | |
| Yu et al. | Multi-Fault Diagnosis of Lithium-Ion battery systems based on correlation Coefficient and similarity Approaches | |
| Li et al. | Online lithium battery fault diagnosis based on least square support vector machine optimized by ant lion algorithm | |
| CN107329089A (en) | A kind of rigid package lithium ion battery explosion propagation effect appraisal procedure and device | |
| CN106291371A (en) | A kind of lithium battery lancing test emulation mode and device | |
| Yang et al. | Internal short circuit fault diagnosis for lithiumion battery based on voltage and temperature | |
| Szürke et al. | Statistical approach for designing generic 18650 battery model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |