CN114325447B - Method, system and device for establishing battery health evaluation model and evaluation - Google Patents
Method, system and device for establishing battery health evaluation model and evaluation Download PDFInfo
- Publication number
- CN114325447B CN114325447B CN202111590382.4A CN202111590382A CN114325447B CN 114325447 B CN114325447 B CN 114325447B CN 202111590382 A CN202111590382 A CN 202111590382A CN 114325447 B CN114325447 B CN 114325447B
- Authority
- CN
- China
- Prior art keywords
- battery
- model
- charge
- establishing
- voltage
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 120
- 230000036541 health Effects 0.000 title claims abstract description 100
- 238000013210 evaluation model Methods 0.000 title claims abstract description 43
- 238000011156 evaluation Methods 0.000 title claims abstract description 40
- 230000008569 process Effects 0.000 claims abstract description 57
- 238000007600 charging Methods 0.000 claims abstract description 23
- 230000006870 function Effects 0.000 claims description 54
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000010277 constant-current charging Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 2
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 21
- 229910052744 lithium Inorganic materials 0.000 description 21
- 238000010586 diagram Methods 0.000 description 20
- 238000009826 distribution Methods 0.000 description 17
- 238000005259 measurement Methods 0.000 description 15
- 238000012549 training Methods 0.000 description 11
- 238000012360 testing method Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 239000012634 fragment Substances 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- FFBHFFJDDLITSX-UHFFFAOYSA-N benzyl N-[2-hydroxy-4-(3-oxomorpholin-4-yl)phenyl]carbamate Chemical compound OC1=C(NC(=O)OCC2=CC=CC=C2)C=CC(=C1)N1CCOCC1=O FFBHFFJDDLITSX-UHFFFAOYSA-N 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000002939 conjugate gradient method Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Abstract
The invention discloses a method, a system and a device for establishing a battery health evaluation model and battery health evaluation, comprising the steps of establishing the battery health evaluation model; acquiring battery charging process data on line; and inputting the charging process data into a battery health evaluation model to obtain battery health evaluation. According to the invention, the evaluation cost is reduced by acquiring the charging data on line and establishing the single-charge capacity prediction model of the battery.
Description
Technical Field
The present invention relates to the field of battery health assessment, and in particular, to a method, system and apparatus for establishing a battery health assessment model and assessment.
Background
The State Of Health (SOH) Of the lithium battery is an important parameter for fault diagnosis and safety early warning Of the lithium battery Of the electric automobile in the full life cycle, and the accurate evaluation Of the SOH has important significance for improving the overall performance Of the lithium battery. At present, research students at home and abroad pay more attention to the research of lithium battery health state assessment, particularly an online assessment method, however, the online assessment result has larger deviation under the condition that the daily practical training data of the lithium battery is insufficient. The data under different time scales have different advantages and disadvantages, and the data is focused on predicting the cycle times of the available discharge capacity under the macroscopic time scale representing the capacity attenuation characteristic of the full life cycle of the lithium battery, but the use condition of the lithium battery in the daily discharge process is uncertain, so that the available discharge capacity can not be directly obtained; the single available capacity can be estimated on line under the microcosmic time scale of the characteristic of single charge and discharge of the lithium battery, but because the lithium battery is not completely charged and discharged in daily use and the individual difference of the lithium battery exists, the establishment of an accurate full life cycle lithium battery SOH evaluation model has difficulty.
At present, the estimation method of the battery state of health is mainly used for predicting single batteries, and the batteries used by the vehicle are formed by connecting hundreds of batteries in series, so that the single batteries in the battery stack can be mutually influenced in the use process, and the uncertainty of the battery state of health estimation is increased. Meanwhile, in the current estimation method, discharge process data are mostly adopted for analysis, special equipment and sites are needed, and a new energy electric vehicle is subjected to a charge and discharge process, so that the data are obtained for analysis, and the cost for one time of analysis is high.
Disclosure of Invention
The invention aims to provide a method, a system and a device for establishing a battery health evaluation model and evaluation, and aims to solve the problem of establishing the battery health evaluation model and evaluation.
The invention provides a method for establishing a battery health evaluation model, which comprises the following steps:
s11, establishing a single charge capacity prediction model of the battery on line;
s12, establishing a full life cycle file of the battery, and correcting a single charge capacity prediction model of the battery according to data in the full life cycle file of the battery;
and S13, obtaining the single-charge predicted capacity of the battery according to the single-charge capacity prediction model of the battery, and dividing the single-charge predicted capacity of the battery by the original capacity of the battery to obtain a battery health evaluation model.
The invention also provides a battery health evaluation method, which comprises the following steps of:
s1, acquiring battery charging process data on line;
s2, inputting the charging process data into a battery health degree evaluation model to obtain battery health degree evaluation.
The invention also provides a system for establishing a battery health evaluation model, which comprises,
and a prediction module: the method comprises the steps of establishing a single charge capacity prediction model of a battery on line;
and (3) a correction module: the method comprises the steps of establishing a full life cycle file of a battery, and correcting a single charge capacity prediction model of the battery according to data in the full life cycle file of the battery;
and the calculation module is used for obtaining the single-charge predicted capacity of the battery according to the single-charge capacity prediction model of the battery, and dividing the single-charge predicted capacity of the battery by the original capacity of the battery to be used as a battery health evaluation model.
The invention also provides a battery health evaluation system, which comprises:
the acquisition module is used for: the method comprises the steps of acquiring battery charging process data on line;
and an evaluation module: and the battery health evaluation module is used for inputting the charging process data into a battery health evaluation model to obtain battery health evaluation.
The embodiment of the invention also provides a device for establishing the battery health evaluation model, which comprises the following steps: a memory, a processor, and a computer program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of a method of building a battery health assessment model.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores an information transmission implementation program, and the program is executed by a processor to implement the steps of the method for establishing the battery health evaluation model.
The embodiment of the invention also provides a battery health degree evaluation device, which comprises: a memory, a processor, and a computer program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of a method of battery health assessment.
By adopting the embodiment of the invention, the low-cost battery health evaluation can be realized by acquiring the charging data on line and establishing the evaluation model on line.
The foregoing description is only an overview of the present invention, and is intended to be implemented in accordance with the teachings of the present invention in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present invention more readily apparent.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method of establishing a battery health assessment model in accordance with an embodiment of the present invention;
FIG. 2 is a flowchart of a battery health assessment method according to an embodiment of the present invention;
FIG. 3 is a prior schematic diagram of Gaussian process regression for the method of establishing a battery health assessment model according to an embodiment of the present invention;
FIG. 4 is a Gaussian process regression posterior schematic diagram of a method for establishing a battery health assessment model according to an embodiment of the invention;
FIG. 5 is a schematic diagram of a Gaussian process model for establishing a battery health assessment model method according to an embodiment of the invention;
FIG. 6 is a single square index covariance schematic;
FIG. 7 is a schematic diagram of two square-index covariances;
FIG. 8 is a neural network covariance schematic;
FIG. 9 is a schematic diagram of a neural network and Matern covariance combination modeling for battery health according to an embodiment of the invention;
fig. 10 is a schematic diagram of a feature length l=1 of the battery health evaluation model according to the embodiment of the present invention;
fig. 11 is a schematic diagram of a feature length l=2 of the battery health evaluation model according to the embodiment of the present invention;
fig. 12 is a schematic diagram of a feature length l=3 of the battery health evaluation model according to the embodiment of the present invention;
FIG. 13 is a schematic representation of Gaussian process regression prediction for establishing a battery health assessment model according to an embodiment of the present invention;
FIG. 14 is a graph of error distribution between predicted data points and measured values for establishing a battery health assessment model in accordance with an embodiment of the present invention;
FIG. 15 is a schematic view of a fitted curve of a Gaussian process for establishing a battery health assessment model according to an embodiment of the invention;
FIG. 16 is a schematic view of fitting errors of a Gaussian process for modeling battery health assessment according to an embodiment of the present invention;
FIG. 17 is a schematic diagram of a measurement power uncertainty evaluation result of a battery health evaluation model according to an embodiment of the present invention;
FIG. 18 is a schematic diagram of a system for modeling battery health assessment in accordance with an embodiment of the present invention;
FIG. 19 is a schematic diagram of a battery health assessment system according to an embodiment of the present invention;
fig. 20 is a schematic diagram of an apparatus for establishing a battery health evaluation model according to an embodiment of the present invention.
Fig. 21 is a schematic view of a battery health assessment apparatus according to an embodiment of the present invention.
Reference numerals illustrate:
1810: a prediction module; 1820: a correction module; 1830: a computing module; 1910: an acquisition module; 1920: and an evaluation module.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise. Furthermore, the terms "mounted," "connected," "coupled," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
Method embodiment
According to an embodiment of the present invention, there is provided a method for establishing a battery health degree evaluation model, and fig. 1 is a flowchart of a method for establishing a battery health degree evaluation model according to an embodiment of the present invention, as shown in fig. 1, specifically including:
s11, establishing a single charge capacity prediction model of the battery on line;
s11 specifically comprises:
s111, establishing a measured electric quantity model in a constant current charging mode:
f (u) -the corresponding time at the cut-off voltage; q (Q) 1 For the charge quantity, u is the termination voltage;
d-the number of current values included in the interval;
i-includes an intra-interval current value;
i (k) -comprising an inter-zone current value;
s112, fitting and predicting the charge and discharge data of the second half section according to the Gaussian process to obtain the corresponding time and the cutoff voltage of the single cutoff voltage;
s112 specifically includes:
fitting and predicting charge and discharge data according to the sum of the neural network covariance function and the Matern covariance function to obtain the corresponding time and the cutoff voltage of the single cutoff voltage.
And S113, carrying the corresponding time of the single cut-off voltage and the cut-off voltage into a measured electric quantity model to calculate the single charge prediction capacity of the battery.
S12, establishing a full life cycle file of the battery, and correcting a single charge capacity prediction model of the battery according to data in the full life cycle file of the battery;
and S13, obtaining the single-charge predicted capacity of the battery according to the single-charge capacity prediction model of the battery, and dividing the single-charge predicted capacity of the battery by the original capacity of the battery to obtain a battery health evaluation model.
According to an embodiment of the present invention, there is further provided a method for evaluating the health of a battery, and fig. 2 is a flowchart of the method for evaluating the health of a battery according to the embodiment of the present invention, as shown in the drawings, specifically including:
s1, acquiring battery charging process data on line;
s2, inputting the charging process data into a battery health degree evaluation model to obtain battery health degree evaluation.
The specific implementation is as follows:
data model fusion is generally a model built to overcome the shortcomings of single data driven algorithms, fusing various conditions. According to the method, a capacity estimation model of daily single fragment data under a micro time scale of a lithium battery is fused with a full life cycle attenuation prediction model under a macro time scale to carry out on-line evaluation on SOH, a full life cycle file of the battery is established through vehicle vin coding or license plate number, battery coding and other parameters through capacity estimation of the single fragment data, and then the single fragment estimation model is continuously corrected through vehicle running condition characteristics, so that model fusion calibration of the full life cycle is finally realized, and therefore, the SOH of the battery under approximate conditions can be accurately predicted. Meanwhile, the uncertainty of SOH estimation is quantitatively estimated, and the reliability of an estimation result is guaranteed;
the method mainly solves the problem that the SOH estimation model is difficult to cover completely under the condition of insufficient daily training data of the lithium battery and the problem of individual variability of the capacity estimation model, and provides a model for establishing and estimating the available charge capacity of the lithium battery based on the segment charge data for solving the problems.
Battery state of health (SOH) =current estimated capacity/original capacity.
The problem of individual variability of a capacity online estimation model is solved by adopting a machine learning method of directly mapping the capacity by parameters;
and (3) establishing a current available charge capacity model of the estimated lithium battery on line based on daily segment charge data, and converting machine learning from macroscopic time into prediction and estimation of a single charge curve of the lithium battery under microscopic time scale. According to the method, on the basis of determining a lithium battery charging working voltage platform through a capacity increment method, a relation between voltage and time is provided as a measurement equation, and an iterative Gaussian Kalman filtering algorithm is provided by combining nonlinear fitting capacity of a Gaussian process regression (Gaussian Process Regression, GPR) algorithm and historical iterative learning capacity of an extended Kalman filter (Extended Kalman Filtering, EKF), and lithium battery capacity increment (Incremental Capacity, IC) curve change of the lithium battery in microscopic time is estimated by using estimated battery per-se daily segment charging data. IC method is a common method for studying SOH. The principle of the method is that a constant current charge and discharge test is carried out on a battery, an Open-Circuit Voltage (OCV) and charge and discharge capacity of the battery are continuously recorded, and a battery capacity change curve corresponding to the OCV is drawn. The OCV versus battery capacity curve may change as the battery ages, reflecting the battery capacity degradation characteristics.
And (3) obtaining a single full charge curve to further estimate the current charge available capacity of the lithium battery.
The Gaussian Process (GP) is a novel algorithm applied to the field of machine learning, and is mainly modeled by using a function space theory, and an objective function is defined as a priori distribution of the gaussian process, and bayesian inference is performed in the space range of the objective function.
In performing Gaussian process regression predictions we mark the mean function as m (x), and the covariance function as k (x, x'), defined as follows:
m(x)＝E(f(x))
k(x,x')＝E[f(x)-m(x)][f(x')-m(x')]
consider a simple regression model:
y＝f(x)+εf(x)＝X T w
training set { (x) i ,y i ) I=1,.. i (i=1,., n) is the input value, f is the function value, y i (i=1.,), n) is the output value of the output signal, epsilon represents an independent co-distributed Gaussian white noise with a mean of 0 and a variance of sigma 2 The following forms can be obtained:
ε:N(0,σ 2 )
the probability density function of the observed value y is obtained as follows:
the a priori distribution of observations y is:
y|X,w:N(X T w,k(x,x')|σ 2 I)
when applied to the estimation of the predicted value, the observed value y and the predicted value f * The distribution of (2) is as follows:
wherein k (X, X) * ) Represents the covariance between training data and predicted values, k (x * ,x * ) Is the covariance between the predicted values.
When the observed value contains noise, the covariance of the observed value isThe joint distribution of observations and test values is:
according to Bayesian inference, the posterior distribution of the weights is obtained, and the posterior distribution inference of the Bayesian linear model is as follows:
f * |X,f,x * :N(k(x * ,X)k(X,X) -1 f,k(x * ,x * )-k(x * ,X)k(X,X) -1 k(X,x * ))
FIG. 3 is a prior schematic representation of a Gaussian process regression for establishing a battery health assessment model in accordance with an embodiment of the present invention; as shown in fig. 3: from the gaussian process a priori, 3 sample functions are derived, where the discrete points are the output values produced by the gaussian process and the other two consecutive samples are produced by a large number of points. FIG. 4 is a Gaussian process regression posterior schematic of a method of modeling battery health according to an embodiment of the invention; as shown in fig. 4: three posterior derived functions were obtained from 5 noise-free observations, with the gray areas in the graph representing the mean of the posterior probabilities and 95% confidence intervals.
FIG. 5 is a schematic illustration of a Gaussian process model for modeling battery health assessment in accordance with an embodiment of the present invention; as shown in fig. 5:
the nature of the Gaussian Process (GP) model is determined by its covariance function (i.e., kernel function), and the key to gaussian process modeling is to determine the type of covariance function. Gaussian process modeling requires that we must first choose the mean function and covariance function, which we typically set to zero.
GP is often parameterized, with respect to covariance function, there are several popular covariance functions:
(1) Square index covariance function:
the covariance function is an infinitely differentiable function, where l is the characteristic length.
(2) Constant covariance function:
(3) Periodic covariance function
These covariance functions are fixed and are often used for prediction to give good results.
In the fitting prediction of the battery charging and discharging process, a neural network and a Matern function are combined to be used as covariance functions, the neural network is suitable for fitting data, x in different positive and negative directions are allowed to be saturated, and the distribution predicted by using three covariance functions is compared; the mean and 95% confidence interval for the noise signal is shown as gray region in 64 data points (standard deviation of gaussian noise is σ=0.1) generated by a step function. FIG. 6 is a single square index covariance schematic; as shown in fig. 6: is a single square index covariance (SE); FIG. 7 is a schematic diagram of two square-index covariances; as shown in fig. 7: the sum of the two square-index covariances, this covariance function is more flexible than the first covariance function is predicted because it has two parameters of size and length. The predicted distributions look better, but none of them is an ideal fit, fig. 8 is a neural network covariance schematic; as shown in fig. 8: the use of neural network covariance functions (NNs) is an ideal fit that contains x with different values in the positive and negative directions. FIG. 9 is a schematic diagram of a neural network and Matern covariance combination modeling for battery health according to an embodiment of the invention; as shown in fig. 9: is a combined form of a neural network and a Matern function (nn+matern).
The covariance function used herein is as follows:
(1) Neural network covariance function:
(2) Maternard covariance function:
the Maternard covariance function f is r d F is a function of 1 (t)＝1,f 3 (t)＝1+t,r d Is the distance between x and x'>
(3) Neural network and Maternard k(x,x') The sum is still the covariance function:
some of the free parameters of the covariance function, commonly referred to as hyper-parameters, are variable, and the difference in predicted data will result in a large fitting error, and optimization and maximization of log-likelihood are necessary to obtain a proper hyper-parameter, assuming X as the training input, y as the training output, where θ is the hyper-parameter vector, deduced by Bayesian:
the super-parameters are determined by training data, noting that the length l is the core variance, signal varianceAnd noise signal sigma 2 Is a variation ofA kind of electronic device. By experimentation we will find that when different initial values are set, different fitted curves will be obtained, and the error between them will be smaller. According to the proposed problem we should select an appropriate initial value. In the subsequent verification process, different parameters are selected for X according to different inputs of training data.
And obtaining the maximum value of the log likelihood function of the training sample by adopting a conjugate gradient method, obtaining the optimal super-parameter, and taking the optimal super-parameter as the average predicted value of the super-parameter. Taking square-index covariance as an example, the super-parameters include (l, sigma) f ,σ n ) Fig. 10 is a schematic diagram of a characteristic length l=1 of a method of establishing a battery health evaluation model according to an embodiment of the present invention; as shown in fig. 10; fig. 11 is a schematic diagram of a characteristic length l=2 of a method of establishing a battery health evaluation model according to an embodiment of the present invention; as shown in fig. 11; fig. 12 is a schematic diagram of a feature length l=3 of the battery health evaluation model according to the embodiment of the present invention; as shown in fig. 12; taking the Gaussian process estimation results at different values, it can be seen that the variation of different super parameters can influence the prediction result, and we have to perform super parameter optimization in order to obtain a better prediction result.
FIG. 13 is a schematic representation of Gaussian process regression prediction for establishing a battery health assessment model according to an embodiment of the present invention; as shown in fig. 13;
a combination of neural network covariance functions (NN) and Matern covariance functions is selected as a kernel function for Gaussian process prediction. And carrying out charge and discharge data prediction by using Gaussian Process Regression (GPR), selecting training data and test data by using lithium battery test data, limiting prior distribution by using the input training data, and calculating to obtain an output predicted value of the GP posterior distribution function by using a Bayesian framework.
The distribution of the cut-off time is predicted by a Gaussian process in the electric quantity measurement model. The measured data is analyzed, and the current rises briefly in the initial stage of the constant current charging process, and then the current is stabilized in a section.
F (u) in the formula, namely the corresponding time(s) when the voltage is cut off;
d-the number of current values included in the interval;
i-comprises an intra-interval current value (A);
i (k) -contains the inter-zone current value (A).
The model for measuring the electric quantity established in the constant-current charging mode is obtained by performing fitting prediction in the constant-current charging mode by using a Gaussian process-based method. And selecting constant-current charging data of the battery to carry out fitting prediction to obtain cut-off voltage of 4.2V, and substituting the cut-off voltage into an electric quantity model to calculate single capacity of the battery.
The complete constant-current charging curve is predicted by using a Gaussian process, and the error distribution between predicted data points and measured values can be obtained: FIG. 14 is a graph of error distribution between predicted data points and measured values for establishing a battery health assessment model in accordance with an embodiment of the present invention; as shown in fig. 14;
fitting prediction in a constant current charging mode is performed by using a Gaussian process, and the key point is to predict the time under the voltage cut-off condition, so that the second half data of the charging process is predicted to obtain a fitting curve based on the Gaussian process and fitting error distribution of the fitting curve; FIG. 15 is a schematic view of a fitted curve of a Gaussian process for establishing a battery health assessment model according to an embodiment of the invention; as shown in fig. 15;
FIG. 16 is a schematic view of fitting errors of a Gaussian process for modeling battery health assessment according to an embodiment of the present invention; as shown in fig. 16;
the functional relation between the time and the voltage of the voltage cut-off point can be obtained from the prediction, the Gaussian distribution is satisfied, and the probability density function is as follows:
N(3230.5，2.3909)
the probability density function of each input quantity in the electric quantity measurement model in the CCC mode is determined, firstly, the voltage indication value of the universal meter is analyzed, the proper inclusion probability (95% or 99%) is selected, and the inclusion interval corresponding to the probability is determined. The inclusion interval should include the indication value of the stable voltage corresponding to the stable charging current, and the voltage value in the inclusion interval is set to be U, so as to represent the true value of the current. The voltage value U (k) outside the interval is included as an unstable voltage to characterize the current true value.
The electric quantity measurement model in the constant-current charging mode can know the total electric quantity Q in the constant-current charging process 1 The measurement uncertainty of (2) depends on the following input quantities:and determining probability density functions of the input quantities according to related technical parameters of each measuring device, and listing distribution types met by the density functions, wherein the distribution types are shown in the following table.
Input quantity probability density function table of electric quantity measurement model in CCC mode
The sampling period in the table is selected to be 1s, the standard resistor is 1 omega, lambda 1 ,λ 2 Digital multimeter voltage and current reading measuring precision, delta, respectively 34411A 1 ,δ 2 Is a fixed error determined by the measurement range. Taking the measurement Range of 34411A DC voltage measurement 100V as an example, the error is (+/-) (0.0035%. Times.U+0.0006%. Times.Range), delta 1 The content was recorded as 0.0006% by range.
The uncertainty of electric quantity measurement is assessed by using a Matlab self-adaptive MCM method, and the experimental parameters are set as follows:
(1) The inclusion probability P is set to 99%;
(2) The set value of the electric quantity standard uncertainty effective number ndig is 1;
(3) In the experiment, the accuracy of the result is ensured by increasing the test times of the Monte Carlo method in consideration of the fact that an electric quantity measurement model is complex. The number of trials M was set to max (10000/1-P, 1000000).
And (3) evaluating the uncertainty of the electric quantity measurement in the CCC mode by utilizing an MCM method, wherein the evaluation result of the uncertainty of the electric quantity measurement is as follows: FIG. 17 is a schematic diagram of a measurement power uncertainty evaluation result of a battery health evaluation model according to an embodiment of the present invention; as shown in fig. 17;
electric quantity uncertainty evaluation result table in CCC mode
The battery health state evaluation can provide reliable evaluation basis for prolonging the service life of the power battery and support the gradient utilization of the power battery;
by the method, under the existing condition, the evaluation of the SOH of the battery can be realized through collecting the charging process data of the charging pile. The method can greatly reduce the evaluation cost of the existing new energy electric vehicle battery, improve the accuracy of battery health evaluation and reduce the influence on the battery.
Meanwhile, by the application of the method, the operation difficulty of battery evaluation can be greatly simplified, and the threshold of new energy electric vehicle battery evaluation is reduced.
System embodiment
According to an embodiment of the present invention, a system for establishing a battery health evaluation model is provided, and fig. 18 is a schematic diagram of the system for establishing a battery health evaluation model according to the embodiment of the present invention, as shown in fig. 18, specifically including:
prediction module 1810: the method comprises the steps of establishing a single charge capacity prediction model of a battery on line;
correction module 1820: the method comprises the steps of establishing a full life cycle file of a battery, and correcting a single charge capacity prediction model of the battery according to data in the full life cycle file of the battery;
the calculating module 1830 is configured to obtain a predicted single-charge capacity of the battery according to the predicted single-charge capacity of the battery, and divide the predicted single-charge capacity of the battery by the original capacity of the battery to obtain a health evaluation model of the battery.
The prediction module 1810 includes:
and the electric quantity measuring module is used for: the method is used for establishing a measured electric quantity model in a constant current charging mode:
f (u) -the corresponding time at the cut-off voltage; q (Q) 1 Is the charge quantity; u is the termination voltage;
d-the number of current values included in the interval;
i-includes an intra-interval current value;
i (k) -comprising an inter-zone current value;
fitting module: the method comprises the steps of fitting and predicting the charge and discharge data of the second half section according to a Gaussian process to obtain the corresponding time of a single cut-off voltage and the cut-off voltage;
single charge capacity module: the method comprises the steps of carrying the corresponding time of the single cut-off voltage and the cut-off voltage into a measured electric quantity model to calculate the single charge prediction capacity of the battery;
the fitting module is specifically used for:
fitting and predicting charge and discharge data according to the sum of the neural network covariance function and the Matern covariance function to obtain the corresponding time and the cutoff voltage of the single cutoff voltage.
According to an embodiment of the present invention, a battery health evaluation system is provided, and fig. 19 is a schematic diagram of a system for establishing a battery health evaluation model according to an embodiment of the present invention, as shown in fig. 19, specifically including:
acquisition module 1910: the method comprises the steps of acquiring battery charging process data on line;
the evaluation module 1920: and the battery health evaluation module is used for inputting the charging process data into a battery health evaluation model to obtain battery health evaluation.
The embodiment of the present invention is a system embodiment corresponding to the above method embodiment, and specific operations of each module may be understood by referring to the description of the method embodiment, which is not repeated herein.
Device embodiment 1
An embodiment of the present invention provides a device for establishing a battery health evaluation model, as shown in fig. 20, including: memory 200, processor 202, and a computer program stored on memory 200 and executable on processor 202, which when executed by the processor, performs the steps of the method embodiments described above.
Device example two
An embodiment of the present invention provides a computer readable storage medium, where a program for implementing information transmission is stored, where the program when executed by the processor 202 implements the steps in the above-described method embodiments.
Device example III
An embodiment of the present invention provides a battery health evaluation device, as shown in fig. 21, including: memory 210, processor 212, and a computer program stored on memory 210 and executable on processor 212, which when executed by the processor, performs the steps of the method embodiments described above.
The computer readable storage medium of the present embodiment includes, but is not limited to: ROM, RAM, magnetic or optical disks, etc.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may alternatively be implemented in program code executable by computing devices, so that they may be stored in a memory device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; these modifications or alternatives to the embodiments of the present invention do not depart from the spirit and scope of the present invention.
Claims (8)
1. A method of modeling a battery health assessment, comprising:
s11, establishing a single charge capacity prediction model of the battery on line;
the step S11 specifically comprises the following steps:
s111, establishing a measured electric quantity model in a constant current charging mode:
f (u) -the corresponding time at the cut-off voltage; q (Q) 1 Is the charge quantity; u is the termination voltage;
d-the number of current values included in the interval;
i-includes an intra-interval current value;
i (k) -comprising an inter-zone current value;
s112, fitting and predicting the charge and discharge data of the second half section according to the Gaussian process to obtain the corresponding time and the cutoff voltage of the single cutoff voltage;
s113, carrying the corresponding time of the single cut-off voltage and the cut-off voltage into a measured electric quantity model to calculate the single charge prediction capacity of the battery;
s12, establishing a full life cycle file of the battery, and correcting a single charge capacity prediction model of the battery according to data in the full life cycle file of the battery;
and S13, obtaining the single-charge predicted capacity of the battery according to the single-charge capacity prediction model of the battery, and dividing the single-charge predicted capacity of the battery by the original capacity of the battery to obtain a battery health evaluation model.
2. The method according to claim 1, wherein S112 specifically comprises:
fitting and predicting charge and discharge data according to the sum of the neural network covariance function and the Matern covariance function to obtain the corresponding time and the cutoff voltage of the single cutoff voltage.
3. A battery health evaluation method, characterized by comprising, based on the battery health evaluation model according to any one of claims 1 to 2:
s1, acquiring battery charging process data on line;
s2, inputting the charging process data into a battery health degree evaluation model to obtain battery health degree evaluation.
4. A system for establishing a battery health evaluation model is characterized by comprising,
and a prediction module: the method comprises the steps of establishing a single charge capacity prediction model of a battery on line;
and (3) a correction module: the method comprises the steps of establishing a full life cycle file of a battery, and correcting a single charge capacity prediction model of the battery according to data in the full life cycle file of the battery;
the calculation module is used for obtaining the single-charge predicted capacity of the battery according to the single-charge capacity prediction model of the battery, and dividing the single-charge predicted capacity of the battery by the original capacity of the battery to be used as a battery health evaluation model;
the prediction module includes:
and the electric quantity measuring module is used for: the method is used for establishing a measured electric quantity model in a constant current charging mode:
f (u) -the corresponding time at the cut-off voltage; q (Q) 1 Is the charge quantity; u is the termination voltage;
d-the number of current values included in the interval;
i-includes an intra-interval current value;
i (k) -comprising an inter-zone current value;
fitting module: the method comprises the steps of fitting and predicting the charge and discharge data of the second half section according to a Gaussian process to obtain the corresponding time of a single cut-off voltage and the cut-off voltage;
single charge capacity module: the method comprises the steps of carrying the corresponding time of the single cut-off voltage and the cut-off voltage into a measured electric quantity model to calculate the single charge prediction capacity of the battery;
the fitting module is specifically configured to:
fitting and predicting charge and discharge data according to the sum of the neural network covariance function and the Matern covariance function to obtain the corresponding time and the cutoff voltage of the single cutoff voltage.
5. A battery health assessment system, characterized by comprising, based on the battery health assessment model of any one of claims 1 to 2:
the acquisition module is used for: the method comprises the steps of acquiring battery charging process data on line;
and an evaluation module: and the battery health evaluation module is used for inputting the charging process data into a battery health evaluation model to obtain battery health evaluation.
6. An apparatus for evaluating the health of a battery, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor performs the steps of the battery health assessment method as claimed in claim 3.
7. An apparatus for modeling a battery health assessment, comprising: memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor carries out the steps of the method of establishing a battery health assessment model according to any one of claims 1 to 2.
8. A computer-readable storage medium, wherein a program for realizing information transfer is stored on the computer-readable storage medium, which program, when executed by a processor, realizes the steps of the method for establishing a battery health evaluation model according to any one of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111590382.4A CN114325447B (en) | 2021-12-23 | 2021-12-23 | Method, system and device for establishing battery health evaluation model and evaluation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111590382.4A CN114325447B (en) | 2021-12-23 | 2021-12-23 | Method, system and device for establishing battery health evaluation model and evaluation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114325447A CN114325447A (en) | 2022-04-12 |
| CN114325447B true CN114325447B (en) | 2023-12-29 |
Family
ID=81053656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111590382.4A Active CN114325447B (en) | 2021-12-23 | 2021-12-23 | Method, system and device for establishing battery health evaluation model and evaluation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114325447B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116298930B (en) * | 2023-05-11 | 2023-09-05 | 苏州清研精准汽车科技有限公司 | Method, device and server for measuring health degree of battery pack |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045291A (en) * | 2019-04-26 | 2019-07-23 | 上海什弋维新能源科技有限公司 | A kind of lithium battery capacity estimation method |
| CN110988695A (en) * | 2019-12-24 | 2020-04-10 | 珠海银隆电器有限公司 | Power battery health state evaluation method and device, storage medium and electronic equipment |
| US20210055353A1 (en) * | 2017-12-07 | 2021-02-25 | Yazami Ip Pte. Ltd. | Method and system for online assessing state of health of a battery |
| CN113030744A (en) * | 2021-02-24 | 2021-06-25 | 上海交通大学 | Battery health condition prediction method, system and medium based on health factor extraction |
| WO2021136493A1 (en) * | 2020-01-03 | 2021-07-08 | 恒大新能源汽车投资控股集团有限公司 | Charging control method and apparatus, electronic device, and system |
| WO2021179699A1 (en) * | 2020-03-11 | 2021-09-16 | 深圳易马达科技有限公司 | Method and apparatus for detecting state of health of battery, and electronic device and storage medium |
| CN113721159A (en) * | 2021-09-27 | 2021-11-30 | 重庆大学 | Data-driven battery health estimation method oriented to random local charging |
-
2021
- 2021-12-23 CN CN202111590382.4A patent/CN114325447B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210055353A1 (en) * | 2017-12-07 | 2021-02-25 | Yazami Ip Pte. Ltd. | Method and system for online assessing state of health of a battery |
| CN110045291A (en) * | 2019-04-26 | 2019-07-23 | 上海什弋维新能源科技有限公司 | A kind of lithium battery capacity estimation method |
| CN110988695A (en) * | 2019-12-24 | 2020-04-10 | 珠海银隆电器有限公司 | Power battery health state evaluation method and device, storage medium and electronic equipment |
| WO2021136493A1 (en) * | 2020-01-03 | 2021-07-08 | 恒大新能源汽车投资控股集团有限公司 | Charging control method and apparatus, electronic device, and system |
| WO2021179699A1 (en) * | 2020-03-11 | 2021-09-16 | 深圳易马达科技有限公司 | Method and apparatus for detecting state of health of battery, and electronic device and storage medium |
| CN113030744A (en) * | 2021-02-24 | 2021-06-25 | 上海交通大学 | Battery health condition prediction method, system and medium based on health factor extraction |
| CN113721159A (en) * | 2021-09-27 | 2021-11-30 | 重庆大学 | Data-driven battery health estimation method oriented to random local charging |
Non-Patent Citations (3)
| Title |
|---|
| 基于粒子群算法估计实际工况下锂电池SOH;南金瑞等;《北京理工大学学报》;第41卷(第1期);全文 * |
| 数据驱动的锂离子电池健康状态综合评分及异常电池筛选;贾俊等;《机械工程学报》;第57卷(第14期);全文 * |
| 锂电池容量衰减和循环寿命影响因素浅析;井冰等;《中国安全防范技术与应用》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114325447A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109946623B (en) | SOC (State of Charge) online estimation method for lithium battery | |
| Kong et al. | Voltage-temperature health feature extraction to improve prognostics and health management of lithium-ion batteries | |
| WO2021185308A1 (en) | Online determination method and system for state of health of power battery pack of electric vehicle | |
| KR102652848B1 (en) | Method and device for determining the state of charge and health of lithium sulfur batteries | |
| CN110488194B (en) | Lithium battery SOC estimation method and system based on electrochemical impedance model | |
| CN105334462B (en) | Battery capacity loses estimation on line method | |
| Li et al. | On-line estimation method of lithium-ion battery health status based on PSO-SVM | |
| CN106918789A (en) | A kind of SOC SOH combine online real-time estimation and on-line amending method | |
| CN106055775B (en) | A kind of service life of secondary cell prediction technique that particle filter is combined with mechanism model | |
| CN111624495B (en) | Lithium battery SOC interval estimation method and system based on EKF (extended Kalman Filter) optimization by deep belief network | |
| Che et al. | Lifetime and aging degradation prognostics for lithium-ion battery packs based on a cell to pack method | |
| CN112051511A (en) | Power battery state of health estimation method and system based on multichannel technology | |
| JP2008522152A (en) | Battery state and parameter estimation system and method | |
| CN113917334B (en) | Battery health state estimation method based on evolution LSTM self-encoder | |
| CN108646199A (en) | A kind of power battery SOC estimation method and system | |
| CN111426957A (en) | SOC estimation optimization method for power battery under simulated vehicle working condition | |
| CN115587527A (en) | Battery life prediction method, system, terminal device and computer readable medium | |
| CN115158076A (en) | Metering error evaluation method, device and computer readable storage medium | |
| CN114325447B (en) | Method, system and device for establishing battery health evaluation model and evaluation | |
| CN112528472A (en) | Multi-innovation hybrid Kalman filtering and H-infinity filtering algorithm | |
| CN114545262A (en) | Lithium ion battery parameter identification and SOC estimation method aiming at loss data | |
| KR100878123B1 (en) | Method and system for battery state and parameter estimation | |
| CN113785209B (en) | Method for detecting abnormal battery cell | |
| Ren et al. | Novel strategy based on improved Kalman filter algorithm for state of health evaluation of hybrid electric vehicles Li-ion batteries during short-and longer term operating conditions | |
| US20220334191A1 (en) | Method and Apparatus for Operating a System for Providing an Electrochemical Battery Module for a Device Battery for a Device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |