CN108226809A - A kind of multi-model and battery SOC evaluation method - Google Patents
A kind of multi-model and battery SOC evaluation method Download PDFInfo
- Publication number
- CN108226809A CN108226809A CN201810337096.9A CN201810337096A CN108226809A CN 108226809 A CN108226809 A CN 108226809A CN 201810337096 A CN201810337096 A CN 201810337096A CN 108226809 A CN108226809 A CN 108226809A
- Authority
- CN
- China
- Prior art keywords
- soc
- battery
- model
- voltage
- value
- 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
Links
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
-
- 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]
- G01R31/385—Arrangements for measuring battery or accumulator variables
- G01R31/387—Determining ampere-hour charge capacity or SoC
- G01R31/388—Determining ampere-hour charge capacity or SoC involving voltage measurements
Abstract
The invention discloses a kind of multi-model and battery SOC evaluation method, include the following steps:1)Obtain battery status parameter;2)Battery status parameter is normalized respectively;3)Battery status parameter after normalized is substituted into genetic algorithm optimization BP neural network estimation model, Adaptive Neuro-fuzzy Inference model and OS ELM neural network models respectively, obtains three kinds of SOC estimation results;4)According to SOC OCV relationships, step 3 is obtained)The corresponding preresearch estimates voltage value of three kinds of SOC estimation results institutes of gained;5)The deviation between preresearch estimates voltage value and voltage is calculated respectively;Deviation is normalized to obtain the weighting coefficient of preresearch estimates voltage value;6)The final estimated value of SOC is calculated according to weighting coefficient.It is parallel using three kinds of models using three important parameters of battery, nonlinear system with multiple linear system estimated results is described, then weighted superposition estimation SOC value of battery, effectively improve estimation precision.
Description
Technical field
The present invention relates to SOC estimating techniques fields, and in particular to a kind of multi-model and battery SOC evaluation method.
Background technology
Under energy saving and environmentally friendly overall situation, electric vehicle will become auto industry and realize that the preferred of long term growth produces
Product.At present, electric vehicle development at full speed, but development receives the restriction of power battery service life.As electronic
The power resources of automobile, battery are the bottlenecks for restricting Development of Electric Vehicles, and realize that the accurate estimation of the SOC of battery is to ensure vapour
The premise of vehicle reliability service.
The evaluation method of common SOC mainly has:Current integration method, open circuit voltage method, internal resistance method, Kalman filtering method,
Neural network etc..Current integration method is to obtain battery pack to the integration of time by calculating current in the past period to be disappeared
The electricity of consumption, so as to obtain remaining capacity, but due to current detecting, there are deviations, and SOC will be caused to calculate error, electric current is tired out
Meter integration causes error increasing;The theoretical foundation of open circuit voltage method is open-circuit voltage and the relatively-stationary functional relations of SOC,
But the premise of application is that the requirement battery standing time is longer, and this method is used alone and is only applicable to parking electric automobile state;It is interior
Resistance method is according to the relationship between the internal resistance of cell and SOC, determines SOC by measuring internal resistance, but the internal resistance of cell includes Europe
Nurse internal resistance and polarization resistance, in vehicle travel process, curent change is very fast, internal resistance is caused to calculate extremely complex, and discharge just
Phase internal resistance change rate is small, and it is difficult that this method independently carries out SOC estimations;Kalman filtering method is described dependent on accurate battery status
Model and accurate measured value, this is not easy to realize in practical applications;Neural network has self-organized learning ability, as long as
There are enough sample datas, you can to obtain comparatively ideal estimation effect, shortcoming is the methods of training data and training
It largely can all influence the estimated accuracy of SOC.Batteries of electric automobile SOC be one it is nonlinear, delay, multivariable
Coupling and complicated real-time system, requirement of real-time is very high, and conventional, single control method is difficult to obtain ideal effect,
According to the shortcomings that traditional batteries of electric automobile SOC estimation method, a kind of multi-model of Patent design of the present invention and battery SOC estimate
Calculation method, while consider the complex effects of the factors such as battery temperature, self-discharge rate and cell degradation degree, it realizes to electric vehicle
SOC value of battery is estimated.
Invention content
The technical problem to be solved in the present invention is to provide a kind of multi-model and battery SOC evaluation method, can solve
The problem of existing battery SOC evaluation method precision is not high.
The invention is realized by the following technical scheme:
A kind of multi-model and battery SOC evaluation method, include the following steps:
1) battery status parameter, voltage U, electric current I and temperature T including battery are obtained;
2) battery status parameter is normalized respectively;
3) by the battery status parameter after normalized substitute into respectively genetic algorithm optimization BP neural network estimation model,
Adaptive neural network-Fuzzy inference system model and OS-ELM neural network models obtain the corresponding SOC estimation results of each model;
4) according to SOC-OCV relationships, three kinds of SOC estimation result preresearch estimates corresponding voltage values U obtained by step 3) are utilizedm, m
=1,2,3;
5) preresearch estimates voltage value U is calculated respectivelymDeviation e between voltage Um, m=1,2,3;By deviation emReturn
One change obtains preresearch estimates voltage value UmWeighting coefficient;
6) the final estimated value of SOC is calculated according to weighting coefficient.
The further scheme of the present invention is that step 4) is combined using current integration method and open circuit voltage method and obtains SOC-OCV
Relationship.
The present invention further scheme be, it is contemplated that discharge-rate, temperature, cell degradation factor are to current integration method
It influences, using corresponding compensating approach current integration method.
The further scheme of the present invention is, step 4) traditional open circuit voltage method is carried out hysteretic characteristic and it is arbitrary parked when
Between two aspect correct.
The advantages of the present invention over the prior art are that:
Patent of the present invention propose a kind of multi-model and SOC estimation method, compared with routine, single evaluation method
Advantage is, which carries out multiple neural network models by a complicated nonlinear system and estimate parallel, obtains pair
Corresponding U is obtained then in conjunction with revised current integration method and open circuit voltage method in the SOC answeredm, calculate preresearch estimates voltage value
UmDeviation e between voltage Um, m=1,2,3;By deviation emNormalization obtains preresearch estimates voltage value UmWeighting system
Number, then the estimation result weighted superposition of three kinds of models is estimated into SOC value of battery, effectively improve estimation precision.
Description of the drawings
Fig. 1 is the SOC estimation method flow chart of the present invention.
Fig. 2 is genetic algorithm flow chart.
Fig. 3 is BP neural network structure chart.
Fig. 4 is fuzzy rule schematic diagram.
Fig. 5 is adaptive neural network-Fuzzy inference system model structure chart.
Fig. 6 is OS-ELM neural network model schematic diagrames.
Specific embodiment
A kind of as shown in Figure 1 multi-model and battery SOC evaluation method, including training stage and estimating stage,
The middle training stage includes the following steps:
1) battery status parameter detection is carried out by the detection device (LECU) in battery case, wherein utilizing LTC6803 chips
Cell voltage U is acquired, battery current I is detected using Hall current sensor, using NTC warmings resistance into the detection of trip temperature T;
The battery status parameter of n timing node in a period is acquired, the 1st timing node battery status is denoted as:[U1, I1, T1, t1],
2nd timing node battery status is denoted as:[U2, I2, T2, t2] ... the n-th timing node battery status is denoted as:[Un, In, Tn, tn]。
2) normalizing equation is utilized:By the battery status [U of i-th of timing nodei, Ii, Ti,
ti], i=1,2 ... ..., n are normalized to:[U'i, I'i, Ti', ti], i=1,2 ... ..., n;With voltage UiNormalization for,
Its normalizing equation is:Wherein Umax、UminThe maximum in n timing node voltage parameter is represented respectively
Value and minimum value.
3) by the battery status parameter [U' after normalizedi, I'i, Ti', ti] genetic algorithm optimization BP god is substituted into respectively
Through network estimation model, adaptive neural network-Fuzzy inference system model and OS-ELM neural network models, obtain three kinds of SOC and estimate
Calculate result:
Wherein, battery status parameter [U'i, I'i, Ti'] substitute into the process that genetic algorithm optimization BP neural network estimates model
It is as follows:
First to battery status parameter [U'i, I'i, Ti'] genetic algorithm optimization processing as shown in Figure 2 is carried out, including:
A. initialization of population:Individual UVR exposure method is real coding, each individual is a real number string, by input layer
It is formed with hidden layer connection weight, hidden layer threshold value, hidden layer and output layer connection weight and output layer threshold value, individual contains
Neural network whole weights and threshold value.
B. fitness calculates:The initial weight and threshold value of BP neural network are obtained according to individual, with training BP neural network
Forecasting system exports afterwards, using the Error Absolute Value between prediction output and desired output and as ideal adaptation angle value F.
Wherein k be coefficient, n be network output node, yiFor the desired output SOC value of i-th of timing node, oiIt is i-th
The prediction output SOC value of timing node;
C. selection operation:Select the SOC value of roulette method prediction battery, the select probability p of each individual ii,
Wherein FiFor the fitness value of individual i, N is population at individual number;
D. crossover operation:Using real number interior extrapolation method, k-th of chromosome akWith l-th of chromosome al
In j crossover operations,
akj=akj(1-b)+aljb
alj=alj(1-b)+akjb
B is [0,1] random number in formula;
D. mutation operation:Choose j-th of gene a of i-th of individualijInto row variation, mutation operation is as follows:
amaxAnd aminIt is aijBound, whereinr2For a random number, g is current changes
Algebraically;GmaxIt is maximum evolution number, r is the random number of [0,1].
Obtained best weight value and threshold value are sent in BP neural network estimation model as shown in Figure 3 and estimated,
Obtain the first SOC result of calculation SOC1。
Battery status parameter [U'i, I'i, Ti'] substitute into adaptive neural network-Fuzzy inference system model process it is as follows:
The thought of fuzzy reasoning is the static non linear mapping between outputting and inputting;It is assumed that system is defeated
Enter for ui∈Ui(i=1,2 ...), it exports as yi∈Yi(i=1,2 ...), it is " accurate " rather than mould to output and input
Paste set.Blurring is that accurate input is converted into fuzzy set, in rule base reasoning generate fuzzy knot using fuzzy rule
By ambiguity solution module conclusion is converted to accurate output.
Adaptive neural network-Fuzzy inference system model is divided into five layers, as shown in Figure 5:
First layer:By U'i, I'i, Ti' according to load voltage value, load current value and rated temperature value, it is divided into more than volume
Definite value, equal to rated value, less than three fuzzy sets of rated value, the high, medium and low membership function value of corresponding generation:
O1,j=μAi(x) (i=1,2,3)
O1,j=μB(i-3)(y) (i=4,5,6)
O1,j=μC(i-6)(z) (i=7,8,9)
X, y, z is the input of node in formula, Ai、B(i-3)、C(i-6)Be with the relevant fuzzy set of this node, first layer
Output is the degree of membership of condition part, and Ai、B(i-3)、C(i-6)Membership function can be that any appropriate parametrization is subordinate to letter
Number;
The second layer:Output node is the product of input signal membership function, the intensity of activation of delegate rules:
O2,i=ωi=μAl(x)μBm(y)μCn(z), (i=1,2,3 ..., 27, l=1,2,3;M=4,5,6;N=7,8,
9);
Third layer:Intensity of activation is normalized:
4th layer:The transmission function of each node is linear function, represents local linear model, each adaptive node
I is exported:
Wherein x, y, z is input variable, pi、qi、ri、wiIt is parameter sets, fiFor fuzzy rule, such as:
Rule 1:If x is A1, y is B1, then f1=p1x+q1y+r1;
Rule 2:If x is A2, y is B2, then f2=p2x+q2y+r2;
Layer 5:Calculate summation of total output as all input signals, that is, second of SOC estimation results SOC2
(note:Following formula O5,1Represent SOC2):
Battery status parameter [U'i, I'i, Ti', ti] substitute into OS-ELM neural network models process it is as follows:
The algorithm principle of OS-ELM neural network models is as shown in Figure 6:Hidden layer neurode parameter (the i.e. threshold of hidden node
Value) once initial give just no longer adjusts, only adjust output layer weighting parameter;Hidden layer output function G (ai,bi, x), wherein ai,
biThe node parameter being randomly generated, x are input variables, number of nodes L.
Initial phase:The battery status parameter of n timing node respectively forms training set, and one is chosen from training set
Data blockFor initializing, wherein R represents real number, N0Represent for the data block chosen
Number, L are node in hidden layer, N0>L;
A. hidden node parameter (a is generated at randomi,bi), wherein i=1 ..., L;
B. initial hidden layer output weight matrix H is calculated0:
C. initial output weights are calculatedWherein:
D. it is currently trained data block sequence number to enable k=0, k.
The study stage:
A.+1 data block of kth is obtainedWherein NjContained data for j-th of data block
Number;
B. H is calculatedk+1:
It enables
C. output weights are calculated:
D. k=k+1 is enabled to turn to step a, the optimal output weights of the network are calculated, weights are assigned to network estimation
SOC3Value.
4) battery standing time long enough can obtain the initial SOC of battery according to SOC-OCV0, in cell operation,
Battery current situation can be detected in real time using sensor, so as to calculate SOC using current integration method:
Above formula is calculation formula ideally, but in actual moving process, need to consider the discharge-rate of battery,
The influence of temperature, cell degradation needs to compensate.
Discharge-rate penalty coefficient η1:As electrochemical equation Peukert equations it is found that electricity and pass existing for discharge current
It is to be:
K=t*In
I is discharge current in formula, and t is the time, K be with the battery related constant of internal active material in itself, n is with moving
The relevant constant of model of power battery;As available from the above equation:
In formula, INFor rated current, CNFor rated capacity, K and n can be acquired by surveying out two groups of C and I, added in discharge-rate and mended
The SOC formula repaid are:
Temperature compensation coefficient η2:Power battery is influenced by temperature more serious, and temperature-compensating formula is:
η2=1-0.008 | TB-T|
In formula, TBFor normal temperature, T is practical temperature, adds in discharge-rate compensation, the SOC formula of temperature-compensating are:
Cell degradation penalty coefficient η3:The specified charge and discharge cycles number α of battery0, the number α of charge and discharge1, compensation
Coefficient is:
Ultimately join discharge-rate compensation, temperature-compensating, cell degradation compensation SOC formula be:
Battery dump energy is estimated that using revised current integration method, it, can be straight if battery is in static condition
Connect the open-circuit voltage that battery is obtained using SOC-OCV relationships.But in real time execution, electric vehicle in lasting traveling, even if
There are of short duration pause, the chemical reaction of inside battery persistently carries out, and the hysteretic characteristic of battery causes at detected open-circuit voltage
In deviation.So open circuit voltage method is utilized in the state of operation, then it will be to two aspect of hysteresis loop and arbitrary holding time
It corrects.Holding time threshold epsilon is set here, if practical holding time is more than ε, it is believed that holding time long enough, open-circuit voltage are
Accurately, if holding time is less than ε, practical holding time is divided into m stage, by linear least square fitting principle,
Obtain the empirical equation of the SOC and open-circuit voltage under m stage:
Then rule of thumb the coefficient of formula carries out least square curve fitting, obtains c3(t)、c2(t)、c1(t)、c0
(t), empirical equation is rewritten as:
SOC=c3(t).ocv3+c2(t).ocv2+c1(t).ocv+c0(t)
Obtain respectively with SOC1,SOC2,SOC3Corresponding preresearch estimates voltage value U1,U2,U3。
5) preresearch estimates voltage value U is calculated respectively1,U2,U3Deviation e between voltage Um, m=1,2,3;Work as deviation
Value emLess than voltage U 10% when, the training stage terminates.
Working stage
By 1 timing node battery status parameter U of actual measurements、Is, TsGenetic algorithm optimization BP neural network is substituted into respectively
Model, adaptive neural network-Fuzzy inference system model and OS-ELM neural network models are estimated, wherein substituting into OS-ELM nerve nets
Network model is to be directly entered the study stage, obtains three SOC result of calculations:
SOCm, m=1,2,3;Corresponding preresearch estimates voltage value U is calculated further according to SOC-OCV relationshipsmWith voltage U it
Between deviation em=U-Um, then deviation is normalized:
The final estimated value of SOC is finally obtained using normalized deviation as weighting coefficient:
SOC=∑s SOCm.P(Um|U)。
Claims (4)
1. a kind of multi-model and battery SOC evaluation method, it is characterised in that include the following steps:
1) battery status parameter, voltage U, electric current I and temperature T including battery are obtained;
2) battery status parameter is normalized respectively;
3) battery status parameter after normalized is substituted into genetic algorithm optimization BP neural network estimation model, adaptive respectively
Neuro-fuzzy inference system model and OS-ELM neural network models are answered, obtains three kinds of SOC estimation results;
4) according to SOC-OCV relationships, the three kinds of SOC estimation result preresearch estimates voltage values U obtained using step 3)m, m=1,2,
3;
5) preresearch estimates voltage value U is calculated respectivelymDeviation e between voltage Um, m=1,2,3;By deviation emNormalization
Obtain preresearch estimates voltage value UmWeighting coefficient;
6) the final estimated value of SOC is calculated according to weighting coefficient.
2. a kind of multi-model as described in claim 1 and battery SOC evaluation method, it is characterised in that:Step 4) uses
Current integration method and open circuit voltage method, which combine, obtains SOC-OCV relationships.
3. a kind of multi-model as claimed in claim 2 and battery SOC evaluation method, it is characterised in that:Current integration method
Further include the compensation of the discharge-rate to battery, temperature, cell degradation.
4. a kind of multi-model as claimed in claim 2 and battery SOC evaluation method, it is characterised in that:Step 4) is opened
Road voltage method includes the amendment to hysteresis loop and arbitrary holding time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810337096.9A CN108226809A (en) | 2018-04-13 | 2018-04-13 | A kind of multi-model and battery SOC evaluation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810337096.9A CN108226809A (en) | 2018-04-13 | 2018-04-13 | A kind of multi-model and battery SOC evaluation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108226809A true CN108226809A (en) | 2018-06-29 |
Family
ID=62657655
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810337096.9A Pending CN108226809A (en) | 2018-04-13 | 2018-04-13 | A kind of multi-model and battery SOC evaluation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108226809A (en) |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109164398A (en) * | 2018-08-03 | 2019-01-08 | 北京交通大学 | Cell capacity evaluation method in a kind of Li-ion batteries piles |
CN109216803A (en) * | 2018-09-20 | 2019-01-15 | 东北大学 | A kind of UMDs battery management system |
CN109738810A (en) * | 2019-01-21 | 2019-05-10 | 温州大学 | A kind of comprehensive estimate method of remaining battery service life |
CN110082682A (en) * | 2019-03-12 | 2019-08-02 | 浙江大学 | A kind of lithium battery charge state estimation method |
CN110110545A (en) * | 2019-03-25 | 2019-08-09 | 湖州师范学院 | The space crowdsourcing Quality Control Model detected based on location privacy protection and tricker |
CN110646737A (en) * | 2019-09-20 | 2020-01-03 | 广州市香港科大霍英东研究院 | Battery SOC dynamic estimation method and system based on multiple models and storage medium |
CN110967636A (en) * | 2019-06-24 | 2020-04-07 | 宁德时代新能源科技股份有限公司 | Battery state of charge correction method, device and system and storage medium |
CN111103549A (en) * | 2019-12-16 | 2020-05-05 | 中车大连机车车辆有限公司 | Method for judging maintenance requirement of battery system of hybrid power locomotive |
CN111103553A (en) * | 2019-12-26 | 2020-05-05 | 江苏大学 | Method for estimating health state of GRNN-adaptive electric vehicle lithium ion battery |
CN112114254A (en) * | 2020-08-25 | 2020-12-22 | 哈尔滨工业大学(威海) | Power battery open-circuit voltage model fusion method |
CN112630661A (en) * | 2020-12-28 | 2021-04-09 | 广州橙行智动汽车科技有限公司 | Battery state of charge (SOC) estimation method and device |
CN113093013A (en) * | 2021-03-31 | 2021-07-09 | 安庆师范大学 | Integrated estimation method for health state of lithium battery pack |
CN113156316A (en) * | 2021-04-26 | 2021-07-23 | 太仓中科赛诺新能源科技有限公司 | Estimation algorithm for SOC of brine battery |
CN113297129A (en) * | 2021-04-28 | 2021-08-24 | 北京天启鸿源新能源科技有限公司 | SOC (System on chip) calculation method and system of energy storage system |
EP3879619A1 (en) * | 2020-03-11 | 2021-09-15 | Micropower ED Marketing AB | Method and system for calibrating a current sensor |
CN113495213A (en) * | 2020-03-20 | 2021-10-12 | 北京京东乾石科技有限公司 | Electric quantity prediction model construction method, electric quantity prediction model carrying equipment deployment method and electric quantity prediction model carrying equipment deployment device |
WO2021217599A1 (en) * | 2020-04-30 | 2021-11-04 | 中国科学院深圳先进技术研究院 | Method for estimating state of charge of battery, and estimator |
CN114207454A (en) * | 2019-10-24 | 2022-03-18 | 株式会社Lg新能源 | Apparatus and method for diagnosing degree of deterioration of battery |
CN114218813A (en) * | 2022-02-18 | 2022-03-22 | 中国汽车技术研究中心有限公司 | Fuel cell flow resistance function construction method and flow resistance value prediction method |
CN116774086A (en) * | 2023-06-09 | 2023-09-19 | 淮阴工学院 | Lithium battery health state estimation method based on multi-sensor data fusion |
CN116840722A (en) * | 2023-06-09 | 2023-10-03 | 淮阴工学院 | Performance degradation evaluation and life prediction method for proton exchange membrane fuel cell |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105425154A (en) * | 2015-11-02 | 2016-03-23 | 北京理工大学 | Method for estimating charge state of power cell set of electric vehicle |
CN106125007A (en) * | 2016-08-31 | 2016-11-16 | 北京新能源汽车股份有限公司 | Determination method, device and the automobile of a kind of battery dump energy |
CN107367693A (en) * | 2017-07-07 | 2017-11-21 | 淮阴工学院 | SOC detection system for power battery of electric vehicle |
CN107422272A (en) * | 2017-07-07 | 2017-12-01 | 淮阴工学院 | A kind of electric automobile power battery SOC intellectualized detection devices |
CN107436409A (en) * | 2017-07-07 | 2017-12-05 | 淮阴工学院 | A kind of electric automobile power battery SOC intelligent predicting devices |
-
2018
- 2018-04-13 CN CN201810337096.9A patent/CN108226809A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105425154A (en) * | 2015-11-02 | 2016-03-23 | 北京理工大学 | Method for estimating charge state of power cell set of electric vehicle |
CN106125007A (en) * | 2016-08-31 | 2016-11-16 | 北京新能源汽车股份有限公司 | Determination method, device and the automobile of a kind of battery dump energy |
CN107367693A (en) * | 2017-07-07 | 2017-11-21 | 淮阴工学院 | SOC detection system for power battery of electric vehicle |
CN107422272A (en) * | 2017-07-07 | 2017-12-01 | 淮阴工学院 | A kind of electric automobile power battery SOC intellectualized detection devices |
CN107436409A (en) * | 2017-07-07 | 2017-12-05 | 淮阴工学院 | A kind of electric automobile power battery SOC intelligent predicting devices |
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109164398A (en) * | 2018-08-03 | 2019-01-08 | 北京交通大学 | Cell capacity evaluation method in a kind of Li-ion batteries piles |
CN109164398B (en) * | 2018-08-03 | 2019-10-11 | 北京交通大学 | Cell capacity evaluation method in a kind of Li-ion batteries piles |
CN109216803A (en) * | 2018-09-20 | 2019-01-15 | 东北大学 | A kind of UMDs battery management system |
CN109738810A (en) * | 2019-01-21 | 2019-05-10 | 温州大学 | A kind of comprehensive estimate method of remaining battery service life |
CN110082682B (en) * | 2019-03-12 | 2020-04-24 | 浙江大学 | Lithium battery state of charge estimation method |
CN110082682A (en) * | 2019-03-12 | 2019-08-02 | 浙江大学 | A kind of lithium battery charge state estimation method |
CN110110545A (en) * | 2019-03-25 | 2019-08-09 | 湖州师范学院 | The space crowdsourcing Quality Control Model detected based on location privacy protection and tricker |
CN110967636A (en) * | 2019-06-24 | 2020-04-07 | 宁德时代新能源科技股份有限公司 | Battery state of charge correction method, device and system and storage medium |
US11231467B2 (en) | 2019-06-24 | 2022-01-25 | Contemporary Amperex Technology Co., Limited | Method, device, and system, for state of charge (SOC) correction for a battery |
CN110646737A (en) * | 2019-09-20 | 2020-01-03 | 广州市香港科大霍英东研究院 | Battery SOC dynamic estimation method and system based on multiple models and storage medium |
CN110646737B (en) * | 2019-09-20 | 2022-04-22 | 广州市香港科大霍英东研究院 | Battery SOC dynamic estimation method and system based on multiple models and storage medium |
CN114207454B (en) * | 2019-10-24 | 2023-06-20 | 株式会社Lg新能源 | Battery DOD diagnosis device, battery pack and battery DOD diagnosis method |
CN114207454A (en) * | 2019-10-24 | 2022-03-18 | 株式会社Lg新能源 | Apparatus and method for diagnosing degree of deterioration of battery |
CN111103549A (en) * | 2019-12-16 | 2020-05-05 | 中车大连机车车辆有限公司 | Method for judging maintenance requirement of battery system of hybrid power locomotive |
CN111103553A (en) * | 2019-12-26 | 2020-05-05 | 江苏大学 | Method for estimating health state of GRNN-adaptive electric vehicle lithium ion battery |
CN111103553B (en) * | 2019-12-26 | 2021-11-23 | 江苏大学 | Method for estimating health state of GRNN-adaptive electric vehicle lithium ion battery |
EP3879619A1 (en) * | 2020-03-11 | 2021-09-15 | Micropower ED Marketing AB | Method and system for calibrating a current sensor |
WO2021180674A1 (en) * | 2020-03-11 | 2021-09-16 | Micropower Group Ab | Method for calibrating a current sensor of a battery system |
CN113495213A (en) * | 2020-03-20 | 2021-10-12 | 北京京东乾石科技有限公司 | Electric quantity prediction model construction method, electric quantity prediction model carrying equipment deployment method and electric quantity prediction model carrying equipment deployment device |
WO2021217599A1 (en) * | 2020-04-30 | 2021-11-04 | 中国科学院深圳先进技术研究院 | Method for estimating state of charge of battery, and estimator |
CN112114254A (en) * | 2020-08-25 | 2020-12-22 | 哈尔滨工业大学(威海) | Power battery open-circuit voltage model fusion method |
CN112114254B (en) * | 2020-08-25 | 2022-02-08 | 哈尔滨工业大学(威海) | Power battery open-circuit voltage model fusion method |
CN112630661A (en) * | 2020-12-28 | 2021-04-09 | 广州橙行智动汽车科技有限公司 | Battery state of charge (SOC) estimation method and device |
CN112630661B (en) * | 2020-12-28 | 2022-12-09 | 广州橙行智动汽车科技有限公司 | Battery state of charge (SOC) estimation method and device |
CN113093013A (en) * | 2021-03-31 | 2021-07-09 | 安庆师范大学 | Integrated estimation method for health state of lithium battery pack |
CN113156316A (en) * | 2021-04-26 | 2021-07-23 | 太仓中科赛诺新能源科技有限公司 | Estimation algorithm for SOC of brine battery |
CN113156316B (en) * | 2021-04-26 | 2023-11-07 | 太仓中科赛诺新能源科技有限公司 | Brine battery SOC estimation algorithm |
CN113297129A (en) * | 2021-04-28 | 2021-08-24 | 北京天启鸿源新能源科技有限公司 | SOC (System on chip) calculation method and system of energy storage system |
CN114218813A (en) * | 2022-02-18 | 2022-03-22 | 中国汽车技术研究中心有限公司 | Fuel cell flow resistance function construction method and flow resistance value prediction method |
CN116774086A (en) * | 2023-06-09 | 2023-09-19 | 淮阴工学院 | Lithium battery health state estimation method based on multi-sensor data fusion |
CN116840722A (en) * | 2023-06-09 | 2023-10-03 | 淮阴工学院 | Performance degradation evaluation and life prediction method for proton exchange membrane fuel cell |
CN116840722B (en) * | 2023-06-09 | 2024-02-23 | 淮阴工学院 | Performance degradation evaluation and life prediction method for proton exchange membrane fuel cell |
CN116774086B (en) * | 2023-06-09 | 2024-03-26 | 淮阴工学院 | Lithium battery health state estimation method based on multi-sensor data fusion |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108226809A (en) | A kind of multi-model and battery SOC evaluation method | |
CN110341690B (en) | PHEV energy management method based on deterministic strategy gradient learning | |
CN105116343B (en) | The electrokinetic cell state of charge method of estimation and system of least square method supporting vector machine | |
CN108872866B (en) | Dynamic evaluation and long-acting prediction fusion method for charge state of lithium ion battery | |
CN107367693B (en) | SOC detection system for power battery of electric vehicle | |
CN111007399B (en) | Lithium battery state of charge prediction method based on improved generation countermeasure network | |
CN107422272B (en) | Intelligent detection device for SOC of power battery of electric automobile | |
CN107436409B (en) | Intelligent SOC prediction device for power battery of electric automobile | |
CN110824363B (en) | Lithium battery SOC and SOE joint estimation method based on improved CKF | |
CN112858929B (en) | Battery SOC estimation method based on fuzzy logic and extended Kalman filtering | |
Wu et al. | The SOC estimation of power Li-Ion battery based on ANFIS model | |
Mamo et al. | Long short-term memory with attention mechanism for state of charge estimation of lithium-ion batteries | |
CN107367699A (en) | A kind of lithium battery SOC estimation new methods based on fractional model | |
JP2004317256A (en) | Internal state detection system for vehicle electric storage device | |
CN112630659A (en) | Lithium battery SOC estimation method based on improved BP-EKF algorithm | |
Dong et al. | Estimation of power battery SOC based on improved BP neural network | |
CN111597750A (en) | Hybrid electric vehicle energy management method based on BP neural network | |
CN107169170A (en) | A kind of Forecasting Methodology of battery remaining power | |
CN116125278A (en) | Lithium battery SOC estimation method and system based on LSTM-EKF algorithm | |
CN112269133A (en) | SOC estimation method based on pre-charging circuit model parameter identification | |
CN114217234B (en) | IDE-ASRCKF-based lithium ion battery parameter identification and SOC estimation method | |
CN115186579A (en) | Lithium battery health state estimation method based on capacity increment variable quantity curve | |
CN111337833B (en) | Lithium battery capacity integrated prediction method based on dynamic time-varying weight | |
CN107301266A (en) | A kind of ferric phosphate lithium cell LOC evaluation methods and system | |
CN112098869A (en) | Self-adaptive electric vehicle SOC estimation method based on big data |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180629 |