CN106154168A - The method for estimating charge state of power cell of data-driven - Google Patents
The method for estimating charge state of power cell of data-driven Download PDFInfo
- Publication number
- CN106154168A CN106154168A CN201610205604.9A CN201610205604A CN106154168A CN 106154168 A CN106154168 A CN 106154168A CN 201610205604 A CN201610205604 A CN 201610205604A CN 106154168 A CN106154168 A CN 106154168A
- Authority
- CN
- China
- Prior art keywords
- state
- value
- data
- soc
- estimation
- 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
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]
- G01R31/3644—Constructional arrangements
- G01R31/3648—Constructional arrangements comprising digital calculation means, e.g. for performing an algorithm
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The present invention relates to the method for estimating charge state of power cell of a kind of data-driven, including step: (1), off-line training, it is thus achieved that the Gaussian process model of battery state-of-charge SOC value under setting state；(2), On-line Estimation, under battery actual motion state, gather the data such as the terminal voltage in each moment, operating current and temperature, estimate state-of-charge SOC value according to the state-of-charge SOC Gaussian process model that off-line training obtains, and calculate average and the variance yields of the state-of-charge SOC value of estimation；Then the state-of-charge SOC value estimated according to variance yields correction.Compared with prior art, the invention has the beneficial effects as follows the electrokinetic cell SOC method of estimation of a kind of combination lot of experimental data and dynamic model, can effectively utilize the mass data obtained at laboratory, system model, the uncertainty of collection data can be considered again in actual moving process, the average of dynamic estimation SOC value and error, thus obtain a high accuracy, sane electrokinetic cell SOC method of estimation.
Description
Technical field
The present invention relates to the method for estimating charge state of power cell of data-driven.
Background technology
Its state-of-charge method of estimation of the most widely used lithium ion battery of existing electrokinetic cell is broadly divided into three classes:
The first kind is electric quantity accumulation method, is also called ampere-hour method, estimates electricity by the battery electricity when charging and discharging
The state-of-charge (state of charge, SOC) in pond, and according to battery temperature and discharge rate etc., SOC is modified, this side
Method is simple, and algorithm is easier to realize, but the subject matter existed has: the parameter that (1) relates to is more, if parameter measurement is forbidden,
Easily cause error；(2) cell degradation and cycle-index are not compensated；(3) by current measurement precision and correction factor etc. because of
The impact of element is bigger；(4) needing the initial SOC value of given accuracy, this is difficult to be given in actual applications.
Equations of The Second Kind is voltage measurement method, according to the relation between open-circuit voltage and the depth of discharge of battery of battery, passes through
The open-circuit voltage measuring battery estimates SOC value, and the method is simple, but wants in actual applications to obtain accurate SOC value,
Must be stood for a long time by battery, the most just can determine that SOC value, during real work, electric current is big ups and downs,
The most less for practical application such as electric automobiles, but as the criterion of battery charging and discharging cut-off.
3rd class is at battery-end electricity based on non-linear modeling methods such as neutral net, fuzzy neural network and Gaussian processes
Nonlinear model is set up, according to great many of experiments curve sum between the input parameter such as pressure, temperature, electric current and SOC value of battery output
It is trained according to system.The subject matter of this type of method is that not account for battery operation be a dynamic process, and is
System runs and there is bigger uncertainty, it is impossible to estimate uncertainty degree the correction model of dynamical system.
4th class is method based on battery Type Equivalent Circuit Model, uses current source, resistance and electric capacity to LiFePO4
Charge-discharge circuit carries out mathematical modeling, off-line or on-line identification model parameter, such as internal resistance and capacitance, is regarded as by battery SOC
The one-component of internal system state vector, uses EKF (extended Kalman filter, EKF) or nothing
The methods such as mark Kalman filtering (unscented Kalman filter, UKF) come dynamic estimation SOC value and new breath, and it is main
It has a problem in that: state equation is typically expressed as linear model by (1) this type of method, and observational equation is expressed as nonlinear equation, so
And SOC value is affected by non-linear factors such as battery pack temperature, charging and discharging currents and use times, its linear modelling can not reflect
Its actual physical process；(2) side that the non-linear relation fitting of a polynomial of the SOC value in observational equation and open-circuit voltage approaches
Method, the terminal voltage surveyed during work determines further according to equivalent circuit with the relation of open-circuit voltage, therefore also relies on system
Modeling accuracy and parameter identification precision.
Summary of the invention
An object of the present invention is to overcome deficiency of the prior art, it is provided that a kind of high accuracy, sane power
Battery SOC method of estimation.
For realizing object above, it is achieved through the following technical solutions:
The method for estimating charge state of power cell of data-driven, it is characterised in that include step:
(1) off-line training, it is thus achieved that the Gaussian process model of battery state-of-charge SOC value under setting state；
(2) On-line Estimation, under battery actual motion state, gathers the terminal voltage in each moment, operating current and temperature etc.
Data, estimate state-of-charge SOC value according to the state-of-charge SOC Gaussian process model that off-line training obtains, and calculate estimation
The average of state-of-charge SOC value and variance yields；Then the state-of-charge SOC value estimated according to variance yields correction.
Preferably, the described state that sets as battery multiplying power or ambient temperature, described off-line estimating step include (1.a),
Gather terminal voltage v of the battery each sampling instant t at a temperature of different multiplying, varying environmentt, operating current itAnd temperature value
ct, and carry out SOC value estimation, obtain SOCt, and be standardized all data processing so that meet all of Gauss distribution
State-of-charge SOC value average is 0.
Preferably, what described off-line was estimated specifically comprises the following steps that
If the input of system is: ut=[it；ct], wherein itFor the current value of t sampling, ctTemperature for t sampling
Angle value, utFor the input vector of t, current value and temperature value by the sampling of t are constituted；
The state variable of system is:
xt=SOCt
Wherein SOCtThe state-of-charge SOC estimation of the t demarcated when estimating for off-line, xtThe state of expression system becomes
Amount, just by the state-of-charge SOC of the t demarcatedtConstitute；
The observational variable of system is:
yt=vt
Wherein vtFor the battery terminal voltage value of t sampling, ytIt is the observational variable of t, t the electricity sampled
Pond terminal voltage value vtConstitute；
(1.b) according to the data study dynamic Gaussian process of SOC of Real-time Collection:
The data composition matrix in k moment before wherein collecting:
I-th row of matrix constitutes a data vector
If the SOC value in k moment obeys Gaussian process, i.e.
WhereinRepresent average be 0 vector, covariance matrix be KgGauss distribution, covariance matrix KgEvery
Individual element is set to:
WhereinRepresenting matrixI-th row m row element, its parameter θg=(wg1, Wg2, Wg3, τg0, αg0, αg1,
σg0) it is model parameter to be learned, δijFor delta operator；Choose N k time data of section continuous print, utilize the maximum likelihood science of law
Practise model parameter θg, its object function is:
Its subscript n represents the n-th segment data, uses gradient method to optimize this object function and can be obtained by model parameter θgEstimate
Evaluation；
In like manner, the dynamic Gaussian process of study observation model:
Wherein according to data and data one matrix of composition of front k-1 of t:
If the terminal voltage observation in k moment obeys Gaussian process, i.e.
Wherein covariance matrix KhEach element be set to:
WhereinRepresenting matrixI-th row m row element, matrix parameter θh=(wh1, wh2, wh3, τh0, αh0,
αh1, σh0) it is model parameter to be learned；According to corresponding N k time data of section continuous print, utilize method of maximum likelihood learning model
Parameter θh, its object function is:
Same employing gradient method optimizes this object function and can be obtained by model parameter θgEstimated value.
Preferably, described On-line Estimation comprises the following steps:
(2.a) area update during state estimation:
First according to the estimated value of previous moment SOC valueProduce three Sigma points:
HereinThe dynamic model Gaussian process parameter corresponding predictive value of generation according to training:
Herein
According toThe core constitutedThe first row；
Therefore the SOC value prior estimate in t can be obtained:
Wherein
(2.b) area update during variance
Use the prior estimate of SOC value, obtain the prior estimate of variance:
Wherein
(2.c) according to the output estimation value that the observation model Gaussian process parameter generation Sigma point trained is corresponding:
Herein
According toThe core constitutedThe first row.Its output estimation value is:
(2.d) gain is estimated
Calculate
Obtain yield value:
(2.e) estimated value of state variable and estimate of variance after being filtered:
It is the estimated value of t SOC.
Compared with prior art, the invention has the beneficial effects as follows a kind of combination lot of experimental data and the power of dynamic model
Battery SOC method of estimation, can effectively utilize the mass data obtained at laboratory, can consider again system in actual moving process
System model, gather the uncertainty of data, the average of dynamic estimation SOC value and error, thus obtain a high accuracy, sane
Electrokinetic cell SOC method of estimation.
Accompanying drawing explanation
Fig. 1 is the flow chart of patent working of the present invention.
Detailed description of the invention
Below in conjunction with the accompanying drawings and the present invention is described in detail by embodiment:
As it is shown in figure 1, the method for estimating charge state of power cell of data-driven, it is characterised in that include step:
(1) off-line training, its process is as follows:
(1.a) utilize the test instrunments such as battery general performance test and calorstat, record battery in different multiplying, difference
At a temperature of terminal voltage v of each sampling instant tt, operating current itWith temperature value ct, and carry out SOC value estimation, obtain SOCt；Institute
Data are had all to be standardized processing so that obeying average is the Gauss distribution of 0.
If the input of system is: ut=[it；ct], wherein itFor the current value of t sampling, ctTemperature for t sampling
Angle value, utFor the input vector of t, current value and temperature value by the sampling of t are constituted.
The state variable of system is: xt=SOCt, wherein SOCtThe state-of-charge SOC of the t demarcated when estimating for off-line
Estimated value, xtThe state variable of expression system, just by the state-of-charge SOC of the t demarcatedtConstitute.
The observational variable of system is: yt=vt
Wherein vtFor the battery terminal voltage value of t sampling, ytIt is the observational variable of t, t the electricity sampled
Pond terminal voltage value vtConstitute.
(1.b) according to the data study dynamic Gaussian process of SOC of Real-time Collection:
The data composition matrix in k moment before wherein collecting:
I-th row of matrix constitutes a data vector
If the SOC value in k moment obeys Gaussian process, i.e.
Wherein covariance matrix KgEach element be set to:
WhereinRepresenting matrixI-th row m row element, its parameter θg=(wg1, wg2, wg3, τg0, αg0, αg1,
σg0) it is model parameter to be learned, δijFor delta operator.Choose N k time data of section continuous print, utilize the maximum likelihood science of law
Practise model parameter θg, its object function is:
Its subscript n represents the n-th segment data, uses gradient method to optimize this object function and can be obtained by model parameter θgEstimate
Evaluation.
In like manner, the dynamic Gaussian process of study observation model:
Wherein according to data and data one matrix of composition of front k-1 of t:
If the terminal voltage observation in k moment obeys Gaussian process, i.e.
Wherein covariance matrix KhEach element be set to:
WhereinRepresenting matrixI-th row m row element, its parameter θh=(wh1, Wh2, wh3, τh0, αh0, αh1,
σh0) it is model parameter to be learned.According to corresponding N k time data of section continuous print, utilize method of maximum likelihood learning model parameter
θh, its object function is:
Same employing gradient method optimizes this object function and can be obtained by model parameter θgEstimated value.
(2) On-line Estimation
First from training data, choose representational M section continuous data be stored in data cell, according to work electricity
Flow valuve and temperature value choose the immediate one group of data initial value as On-line Estimation link Gaussian process core.SetVariance yields is determined according to training dataInitial value.It is worth emphasizing that training data has been marked
Quasi-ization processes, and for obtaining actual result, needs the average and the variance that use according to the result estimated and standardization to repair
Just restore.Next utilizing UKF to carry out On-line Estimation, its process is as follows:
(2.a) area update during state estimation
First according to the estimated value of previous moment SOC valueProduce three Sigma points:
HereinThe dynamic model Gaussian process parameter corresponding predictive value of generation according to training:
Herein
According toThe core constitutedThe first row.
Therefore the SOC value prior estimate in t can be obtained:
Wherein
(2.b) area update during variance
Use the prior estimate of SOC value, obtain the prior estimate of variance:
Wherein
(2.c) according to the output estimation value that the observation model Gaussian process parameter generation Sigma point trained is corresponding:
Herein
According toThe core constitutedThe first row.Its output estimation value is:
(2.d) gain is estimated
Calculate
Obtain:
(2.e) estimated value of state variable and estimate of variance after being filtered:
ThereforeIt is the estimated value of t SOC.
Owing to this method is a kind of method of data-driven, the method being not based on circuit model, therefore battery SOC is dynamic
The dynamic process of state process and observation data is all modeled by Gaussian process, and the training of Gaussian process model is then according to going through
History data obtain.SOC value method of estimation based on data-driven methods such as neutral nets is to set up between input and output
One fixing nonlinear mapping relation, can not carry out dynamic corrections according to the observation in a upper moment, and side of the present invention
Method, owing to establishing a dynamic model, can carry out dynamic corrections according to system operation data, therefore has the most dynamically
Adaptation ability.
For guaranteeing the effectiveness of the inventive method, need to collect lot of experimental data and carry out model training, and test
Checking, and then adjust model parameter, it is thus achieved that optimum Gaussian process model.
The present invention has two advantages: (1) reduces the dependence that SOC estimates initial set value.Traditional ampere-hour method needs mark
Determine the initial value of SOC, typically realized by deep discharge.Owing to the method is data-driven, can be according to historical data
Carry out SOC value according to a preliminary estimate, simultaneously because estimation procedure is a dynamic process, it is possible to dynamic corrections adjusts SOC estimation, thus
Even if in the case of SOC initial value is given not accurately, also can obtain more accurate SOC and estimate；(2) estimation difference is little.Pass through
Test to ferric phosphate lithium cell laboratory ruuning situation and the simulation test according to electric automobile work condition operation, this method
SOC estimates that absolute value error is less than 3% in most cases, and the SOC of ampere-hour method estimates that absolute value error is generally 5%.
Embodiment in the present invention is only used for that the present invention will be described, is not intended that the restriction to right,
Those skilled in that art it is contemplated that other replacements being substantially equal to, the most within the scope of the present invention.
Claims (4)
1. the method for estimating charge state of power cell of data-driven, it is characterised in that include step:
(1) off-line training, it is thus achieved that the Gaussian process model of battery state-of-charge SOC value under setting state；
(2) On-line Estimation, under battery actual motion state, gathers the data such as the terminal voltage in each moment, operating current and temperature,
Estimate state-of-charge SOC value according to the state-of-charge SOC Gaussian process model that off-line training obtains, and calculate the charged shape of estimation
The average of state SOC value and variance yields；Then the state-of-charge SOC value estimated according to variance yields correction.
The method for estimating charge state of power cell of data-driven the most according to claim 1, it is characterised in that described in set
Determining state is battery multiplying power or ambient temperature, and described off-line estimating step includes (1.a), gathers battery in different multiplying, difference
Terminal voltage v of each sampling instant t under ambient temperaturet, operating current itWith temperature value ct, and carry out SOC value estimation, obtain
SOCt, and be standardized all data processing so that all state-of-charge SOC value averages meeting Gauss distribution are 0.
The method for estimating charge state of power cell of data-driven the most according to claim 2, it is characterised in that described from
What line was estimated specifically comprises the following steps that
If the input of system is:
ut=[it；ct]
Wherein itFor the current value of t sampling, ctFor the temperature value of t sampling, utFor the input vector of t, during by t
The current value of the sampling carved and temperature value are constituted；
The state variable of system is:
xt=SOCt
Wherein SOCtThe state-of-charge SOC estimation of the t demarcated when estimating for off-line, xtThe state variable of expression system, just
By the state-of-charge SOC of the t demarcatedtConstitute；
The observational variable of system is:
yt=vt
Wherein vtFor the battery terminal voltage value of t sampling, ytIt is the observational variable of t, t the battery-end sampled
Magnitude of voltage vtConstitute；
(1.b) according to the data study dynamic Gaussian process of SOC of Real-time Collection:
The data composition matrix in k moment before wherein collecting:
I-th row of matrix constitutes a data vector
If the SOC value in k moment obeys Gaussian process, i.e.
WhereinRepresent average be 0 vector, covariance matrix be KgGauss distribution, covariance matrix KgEach unit
Element is set to:
WhereinRepresenting matrixI-th row m row element, its parameter θg=(wg1, wg2, wg3, τg0, αg0, αg1, σg0) it is
Model parameter to be learned, δijFor delta operator；Choose N k time data of section continuous print, utilize method of maximum likelihood learning model
Parameter θg, its object function is:
Its subscript n represents the n-th segment data, uses gradient method to optimize this object function and can be obtained by model parameter θgEstimated value；
In like manner, the dynamic Gaussian process of study observation model:
Wherein according to data and data one matrix of composition of front k-1 of t:
If the terminal voltage observation in k moment obeys Gaussian process, i.e.
Wherein covariance matrix KhEach element be set to:
WhereinRepresenting matrixI-th row m row element, matrix parameter θh=(wh1, wh2, wh3, τh0, αh0, αh1, σh0)
For model parameter to be learned；According to corresponding N k time data of section continuous print, utilize method of maximum likelihood learning model parameter θh,
Its object function is:
Same employing gradient method optimizes this object function and can be obtained by model parameter θgEstimated value.
The method for estimating charge state of power cell of data-driven the most according to claim 1, it is characterised in that described
Line is estimated to comprise the following steps:
(2.a) area update during state estimation:
First according to the estimated value of previous moment SOC valueProduce three Sigma points:
HereinThe dynamic model Gaussian process parameter corresponding predictive value of generation according to training:
Herein
According toThe core constitutedThe first row；
Therefore the SOC value prior estimate in t can be obtained:
Wherein
(2.b) area update during variance
Use the prior estimate of SOC value, obtain the prior estimate of variance:
Wherein
(2.c) according to the output estimation value that the observation model Gaussian process parameter generation Sigma point trained is corresponding:
Herein
According toThe core constitutedThe first row.Its output estimation value is:
(2.d) gain is estimated
Calculate
Obtain yield value:
(2.e) estimated value of state variable and estimate of variance after being filtered:
It is the estimated value of t SOC.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610205604.9A CN106154168B (en) | 2016-04-01 | 2016-04-01 | The method for estimating charge state of power cell of data-driven |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610205604.9A CN106154168B (en) | 2016-04-01 | 2016-04-01 | The method for estimating charge state of power cell of data-driven |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106154168A true CN106154168A (en) | 2016-11-23 |
| CN106154168B CN106154168B (en) | 2019-03-05 |
Family
ID=57353684
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610205604.9A Active CN106154168B (en) | 2016-04-01 | 2016-04-01 | The method for estimating charge state of power cell of data-driven |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106154168B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108318823A (en) * | 2017-12-28 | 2018-07-24 | 上海交通大学 | A kind of lithium battery charge state evaluation method based on noise tracking |
| CN108732508A (en) * | 2018-05-23 | 2018-11-02 | 北京航空航天大学 | A kind of real-time estimation method of capacity of lithium ion battery |
| CN112368588A (en) * | 2018-06-15 | 2021-02-12 | 大和制罐株式会社 | Charge-discharge curve estimation device and charge-discharge curve estimation method for storage battery |
| JP2021103141A (en) * | 2019-12-25 | 2021-07-15 | 本田技研工業株式会社 | Machine learning device, machine learning method, charging rate estimation device, and charging rate estimation system |
| WO2021197038A1 (en) * | 2020-03-31 | 2021-10-07 | 比亚迪股份有限公司 | Method and device for determining state of charge of battery, and battery management system |
| CN113933725A (en) * | 2021-09-08 | 2022-01-14 | 深圳大学 | Method for determining power battery charge state based on data driving |
| CN115079012A (en) * | 2022-07-22 | 2022-09-20 | 羿动新能源科技有限公司 | Characteristic data extraction system and method for big data prediction battery SOC |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169168A (en) * | 2011-05-17 | 2011-08-31 | 杭州电子科技大学 | Battery dump energy estimation method based on particle filtering |
| US20120041698A1 (en) * | 2011-10-27 | 2012-02-16 | Sakti3, Inc. | Method and system for operating a battery in a selected application |
| CN102798823A (en) * | 2012-06-15 | 2012-11-28 | 哈尔滨工业大学 | Gaussian process regression-based method for predicting state of health (SOH) of lithium batteries |
| CN103399276A (en) * | 2013-07-25 | 2013-11-20 | 哈尔滨工业大学 | Lithium-ion battery capacity estimation and residual cycling life prediction method |
| CN105277891A (en) * | 2014-07-18 | 2016-01-27 | 三星电子株式会社 | Method and apparatus for estimating state of battery |
| CN105425153A (en) * | 2015-11-02 | 2016-03-23 | 北京理工大学 | Method for estimating charge state of power cell of electric vehicle |
-
2016
- 2016-04-01 CN CN201610205604.9A patent/CN106154168B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169168A (en) * | 2011-05-17 | 2011-08-31 | 杭州电子科技大学 | Battery dump energy estimation method based on particle filtering |
| US20120041698A1 (en) * | 2011-10-27 | 2012-02-16 | Sakti3, Inc. | Method and system for operating a battery in a selected application |
| CN102798823A (en) * | 2012-06-15 | 2012-11-28 | 哈尔滨工业大学 | Gaussian process regression-based method for predicting state of health (SOH) of lithium batteries |
| CN103399276A (en) * | 2013-07-25 | 2013-11-20 | 哈尔滨工业大学 | Lithium-ion battery capacity estimation and residual cycling life prediction method |
| CN105277891A (en) * | 2014-07-18 | 2016-01-27 | 三星电子株式会社 | Method and apparatus for estimating state of battery |
| CN105425153A (en) * | 2015-11-02 | 2016-03-23 | 北京理工大学 | Method for estimating charge state of power cell of electric vehicle |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108318823A (en) * | 2017-12-28 | 2018-07-24 | 上海交通大学 | A kind of lithium battery charge state evaluation method based on noise tracking |
| CN108318823B (en) * | 2017-12-28 | 2020-06-02 | 上海交通大学 | Lithium battery state of charge estimation method based on noise tracking |
| CN108732508A (en) * | 2018-05-23 | 2018-11-02 | 北京航空航天大学 | A kind of real-time estimation method of capacity of lithium ion battery |
| CN108732508B (en) * | 2018-05-23 | 2020-10-09 | 北京航空航天大学 | Real-time estimation method for lithium ion battery capacity |
| CN112368588A (en) * | 2018-06-15 | 2021-02-12 | 大和制罐株式会社 | Charge-discharge curve estimation device and charge-discharge curve estimation method for storage battery |
| JP2021103141A (en) * | 2019-12-25 | 2021-07-15 | 本田技研工業株式会社 | Machine learning device, machine learning method, charging rate estimation device, and charging rate estimation system |
| JP7082603B2 (en) | 2019-12-25 | 2022-06-08 | 本田技研工業株式会社 | Machine learning device, machine learning method, charge rate estimation device, and charge rate estimation system |
| WO2021197038A1 (en) * | 2020-03-31 | 2021-10-07 | 比亚迪股份有限公司 | Method and device for determining state of charge of battery, and battery management system |
| CN113933725A (en) * | 2021-09-08 | 2022-01-14 | 深圳大学 | Method for determining power battery charge state based on data driving |
| CN113933725B (en) * | 2021-09-08 | 2023-09-12 | 深圳大学 | Method for determining state of charge of power battery based on data driving |
| CN115079012A (en) * | 2022-07-22 | 2022-09-20 | 羿动新能源科技有限公司 | Characteristic data extraction system and method for big data prediction battery SOC |
| CN115079012B (en) * | 2022-07-22 | 2022-11-29 | 羿动新能源科技有限公司 | Characteristic data extraction system and method for big data prediction battery SOC |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106154168B (en) | 2019-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105116343B (en) | The electrokinetic cell state of charge method of estimation and system of least square method supporting vector machine | |
| Lin et al. | State of charge estimation with the adaptive unscented Kalman filter based on an accurate equivalent circuit model | |
| CN110502778B (en) | Adaptive optimization method for estimating battery SOC based on Kalman filtering framework | |
| CN110441694B (en) | Lithium battery state-of-charge estimation method based on multiple fading factors Kalman filtering | |
| CN106154168A (en) | The method for estimating charge state of power cell of data-driven | |
| Li et al. | Enhanced online model identification and state of charge estimation for lithium-ion battery under noise corrupted measurements by bias compensation recursive least squares | |
| Tian et al. | State of charge estimation of lithium-ion batteries using an optimal adaptive gain nonlinear observer | |
| Sun et al. | Adaptive unscented Kalman filtering for state of charge estimation of a lithium-ion battery for electric vehicles | |
| CN104617623B (en) | A kind of electric automobile power battery group balance control method | |
| Kim | The novel state of charge estimation method for lithium battery using sliding mode observer | |
| Liu et al. | A novel model-based state of charge estimation for lithium-ion battery using adaptive robust iterative cubature Kalman filter | |
| CN110596593A (en) | Lithium ion battery SOC estimation method based on intelligent adaptive extended Kalman filtering | |
| CN104267261B (en) | On-line secondary battery simplified impedance spectroscopy model parameter estimating method based on fractional order united Kalman filtering | |
| Li et al. | A combination state of charge estimation method for ternary polymer lithium battery considering temperature influence | |
| CN103472398A (en) | Power battery SOC (state of charge) estimation method based on expansion Kalman particle filter algorithm | |
| CN105093122A (en) | Strong-tracking self-adaptive-SQKF-based SOC estimation method of emergency lamp battery | |
| CN111337832A (en) | Power battery multidimensional fusion SOC and SOH online joint estimation method | |
| CN105717460A (en) | Power battery SOC estimation method and system based on nonlinear observer | |
| CN105203963A (en) | Charge state estimation method based on open-circuit voltage hysteretic characteristics | |
| CN114184962B (en) | Multi-algorithm fusion lithium ion battery SOC and SOH joint estimation method | |
| CN107290678A (en) | A kind of electrokinetic cell health status on-line monitoring method | |
| CN105223512A (en) | Based on the method for the dynamic rectification dump energy of battery behavior | |
| CN112269133B (en) | SOC estimation method based on pre-charging circuit model parameter identification | |
| CN112098849A (en) | Lithium battery residual capacity estimation method based on integral Kalman filtering | |
| CN110196395A (en) | Battery SOC estimation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200318 Address after: Room 1712, 1713 and 1714, building 7, Zhifu center, Xiuzhou New District, Jiaxing City, Zhejiang Province Patentee after: Zhejiang reservoir Yingxin energy science and technology limited company Address before: 200131, Shanghai, Pudong New Area, China (Shanghai) free trade test zone, 29, Tai 1 road, East, room 304-A22 Patentee before: TRUEWIN RENEWABLES TECHNOLOGY (SHANGHAI) Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210108 Address after: 1 / F, 14 / F, 295, Section 2, Guangfu Road, lvshuili, East District, Hsinchu, Taiwan, China Patentee after: Chuying Technology Co.,Ltd. Address before: Rooms 1712, 1713 and 1714, building 7, Zhifu center, Xiuzhou New District, Jiaxing City, Zhejiang Province, 314001 Patentee before: Zhejiang reservoir Yingxin energy science and technology Ltd. |