CN116224099B - Method for dynamically and adaptively estimating battery SOC - Google Patents
Method for dynamically and adaptively estimating battery SOC Download PDFInfo
- Publication number
- CN116224099B CN116224099B CN202310498571.1A CN202310498571A CN116224099B CN 116224099 B CN116224099 B CN 116224099B CN 202310498571 A CN202310498571 A CN 202310498571A CN 116224099 B CN116224099 B CN 116224099B
- Authority
- CN
- China
- Prior art keywords
- battery
- gain matrix
- error
- value
- iteration
- 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
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]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
- G01R31/3842—Arrangements for monitoring battery or accumulator variables, e.g. SoC combining voltage and current measurements
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/32—Circuit design at the digital level
- G06F30/33—Design verification, e.g. functional simulation or model checking
- G06F30/3308—Design verification, e.g. functional simulation or model checking using simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/32—Circuit design at the digital level
- G06F30/33—Design verification, e.g. functional simulation or model checking
- G06F30/3323—Design verification, e.g. functional simulation or model checking using formal methods, e.g. equivalence checking or property checking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Abstract
The invention discloses a method for dynamically and adaptively estimating battery SOC, which relates to the technical field of automobile BMS (battery management system), and the method establishes a battery state observer, carries out self-adaptive correction of the state observer through error feedback, carries out iterative optimization on a gain matrix of the state observer in real time, introduces an error-based dynamic and self-adaptive estimation method, and enables the selection process and the optimizing process of the gain matrix to be continuously and automatically adjusted along with the change of the battery in the continuous iterative updating process, thereby improving the estimation speed and the estimation precision of the battery state, and simultaneously having stronger convergence and higher robustness and stability. According to the invention, the gain matrix is adaptively adjusted along with the change of the battery, so that the optimal observation effect is achieved, and the problem that the model parameters are not matched after the battery is aged and the high precision of SOC estimation cannot be realized is solved.
Description
Technical Field
The invention relates to the technical field of automobile BMS (battery management system), in particular to a method for dynamically and adaptively estimating battery SOC.
Background
SOC refers to the state of charge of a battery, which is used to reflect the remaining capacity of the battery, and is defined numerically as the ratio of the remaining capacity to the battery capacity, and is usually expressed as a percentage; the value range is [0,1], which indicates that the battery is fully discharged when soc=0 and that the battery is fully charged when soc=1. The battery SOC cannot be directly measured, and the size of the battery can be estimated only through parameters such as the voltage of the battery terminal, the charge and discharge current, the internal resistance and the like.
The lithium battery is widely applied as clean and efficient energy sources in production and living, and the SOC estimation of the lithium battery is affected by the nonlinearity of open-circuit voltage, instantaneous current, charge-discharge multiplying power, ambient temperature, battery temperature, parking time, self-discharge rate, coulomb efficiency, resistance characteristics, initial value of SOC, DOD and other external characteristics, the external characteristics affect each other, and the external characteristics are also affected by factors such as different materials, different processes and the like of the lithium battery, so that the problem of uncertainty exists in a battery model, and the SOC estimation precision is not high, is difficult to stabilize, and has poor reliability and robustness.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a method for dynamically and adaptively estimating the SOC of a battery, which is used for establishing a state observer of the battery, adaptively adjusting a gain matrix of the state observer along with the change of the battery aiming at self time-varying parameters of the battery so as to achieve the optimal observation effect and solve the problems that model parameters are not matched after the battery ages and the SOC estimation cannot realize high precision.
In order to achieve the above purpose, the present invention adopts the following technical scheme, including:
a method of dynamically adaptively estimating battery SOC, comprising the steps of:
s1, establishing an equivalent circuit model of a battery;
s2, obtaining model parameters of an equivalent circuit model;
s2, converting the equivalent circuit model into a state equation, and obtaining an actual state value x of the battery according to model parameters;
s4, a state observer of the battery is established, the state observer is used for outputting an estimated state value px of the battery, and an error e between the actual state value x and the estimated state value px is fed back to the state observer and used for correcting the state observer; the specific mode of correction is as follows: configuring a gain matrix L of the state observer to minimize an error e, wherein the gain matrix L obtained by configuration is an initial optimal gain matrix L (0), and the error between an estimated state value px and an actual state value x output by the state observer is e (0) based on the optimal gain matrix L (0);
s5, defining a scale factor lambda, wherein the error e is in direct proportion to the scale factor lambda, namely, the larger the error e is, the larger the scale factor lambda is, the more 0 is less than or equal to lambda is less than or equal to 1;
s6, as the use of the battery changes, the gain matrix L of the state observer is continuously subjected to iterative optimization; the t-th iteration mode is as follows:
taking the optimal gain matrix L (0) as an iteration center, and restraining by a scale factor lambda to obtain a gain matrix L (t) of the t-th iteration:
L(t)=L(0)+λ[L(t-1)-L(0)]；
wherein L (t) is the gain matrix of the t-th iteration; l (t-1) is the gain matrix of the last, i.e., the t-1 th iteration;
substituting the gain matrix L (t) of the t-th iteration into the state observer, and calculating an error e (t) between an estimated state value px and an actual state value x output by the state observer after the t-th iteration;
if the error e (t) is smaller than the error e (0) calculated based on the iteration center, namely the optimal gain matrix L (0), namely e (t) < e (0), updating the optimal gain matrix L (0) into a gain matrix L (t) of the t-th iteration to obtain an updated optimal gain matrix L (0), and continuing the next iteration by taking the updated optimal gain matrix L (0) as the iteration center according to the mode of the step S6;
otherwise, the optimal gain matrix L (0) is not updated, the optimal gain matrix L (0) which is not updated is continuously used as an iteration center, and the next iteration is carried out in a mode of the step S6;
s7, substituting the optimal gain matrix L (0) obtained in real time in the iteration process of the step S6 into a state observer to obtain the estimated state value px of the battery in real time.
Preferably, in step S1, an equivalent circuit model of the battery is as follows:
。
wherein U (t) represents the terminal voltage value of the battery at the time t; uocv (SOC) is an open circuit voltage value corresponding to the SOC value of the battery; r0 represents the internal resistance of the equivalent circuit; i (t) represents the current value of the battery at the time t; t represents a period, i.e., a time period of formula operation; rk represents the resistance value of the kth order RC network; τk represents the time coefficient of the kth order RC network;
k is an order number of the RC network, representing a kth order RC network, k=1, 2.
Preferably, in step S3, the equivalent circuit model is converted into a state equation, and the state equation is as follows:
y=Cx；
d(x)=Ax+Bu；
in the method, in the process of the invention,
；/>；/>；；
wherein u represents an input, in particular a current value of the battery; x represents the output, specifically the state value of the equivalent circuit model, namely the actual state value; d (x) is the derivative of the actual state value x; uk is the voltage value of the kth order RC network, k=1, 2,..k; SOC is the SOC value of the battery; y is the terminal voltage value of the battery, namely the actual terminal voltage; q represents battery capacity; A. b, C are coefficients.
Preferably, in step S4, the model formula of the state observer is as follows:
d(px)=Apx+Bu+L(y-py)=Apx+Bu+L(y-Cpx)=(A-LC)px+Bu+Ly；
wherein px is an estimated state value output by the state observer, and d (px) is a derivative of the estimated state value px; py is the estimated terminal voltage of the battery, py=cpx.
Preferably, in step S4, the error e=x-px between the actual state value x and the estimated state value px is obtained:
d(e)=d(x)-d(px)=Ax+Bu-[(A-LC)px+Bu+Ly]=Ax-(A-LC)px-LCx=(A-LC)e；
where e is the error of the state observer and d (e) is the derivative of the error e.
Preferably, in step S2, model parameters of the equivalent circuit model are obtained through experiments; the model parameters include: internal resistance R0 of the equivalent circuit, resistance Rk of the k-th order RC network in the equivalent circuit, and time coefficient τk of the k-th order RC network in the equivalent circuit.
Preferably, in step S5, an adaptive factor sigma is defined, wherein sigma is not less than 0 and not more than 1, and the adaptive factor sigma is calculated as follows:
if abs (e) is less than or equal to thr, σ=1/[ abs (e) +1];
if abs (e) > thr, σ=0;
wherein thr is the set upper error limit; abs () is an absolute function; e is the error between the actual state value x and the estimated state value px;
scale factor λ=1- σ.
The invention has the advantages that:
(1) The invention establishes the state observer of the battery, and adaptively adjusts the gain matrix of the state observer along with the change of the battery aiming at the self time-varying parameter of the battery so as to achieve the optimal observation effect, thereby solving the problems that the model parameters are not matched after the battery is aged and the high precision of SOC estimation cannot be realized.
(2) The invention introduces a dynamic self-adaptive estimation method based on errors, and in the continuous iterative updating process, the selection process and the optimizing process of the gain matrix are continuously and automatically adjusted along with the change of the battery, so that the estimation speed and the estimation precision of the battery state are improved.
(3) In the invention, when the error e is larger, the obtained scale factor lambda becomes larger, which is equivalent to the larger neighborhood searching pace, so as to improve the searching speed; when the error e is smaller, the obtained scale factor lambda becomes smaller, which is equivalent to the smaller neighborhood searching steps, so as to improve the searching precision.
(4) The dynamic self-adaptive estimation method is suitable for the application of random charging and discharging and severe fluctuation of current and voltage of electric vehicles, combines a high-precision SOC-OCV curve, considers various influencing factors such as temperature, current, internal resistance of a battery, total capacity, sensor drift and the like in a multi-scale manner, carries out self-adaptive correction of a state observer through error feedback of actual terminal voltage and estimated terminal voltage, carries out iterative optimization on a gain matrix of the state observer in real time, realizes high-precision estimation of a battery state, namely SOC, has stronger stability and convergence, can quickly converge in a short time even if the SOC error is larger, achieves higher precision, and has higher robustness and stability.
(5) The invention establishes a high-precision SOC dynamic self-adaptive estimation method, realizes that the estimation precision of the SOC can be corrected to be within 2 percent according to the initial error within 5 minutes.
Drawings
Fig. 1 is a flowchart of a method for dynamically adaptively estimating a battery SOC according to the present invention.
Fig. 2 is an equivalent circuit diagram of a lithium battery.
Fig. 3 is a schematic diagram of a state equation and a state observer for a lithium battery.
Fig. 4 is a schematic diagram of error feedback for a state observer.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only 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.
Battery models can be classified into electrochemical models, neural network models, and equivalent circuit models according to the principle of lithium batteries. The electrochemical model is mainly used for describing the internal mechanism of the battery, and relates to more parameters in the aspects of battery internal chemistry and materials, huge operation amount and extremely complex model. The neural network model regards the battery as a black box, does not pay attention to the internal mechanism of the battery, only pays attention to the mapping relation between the external characteristics and the input of the battery, and a large amount of data is often required to train the neural network model in battery modeling. The battery model widely used at present is an equivalent circuit model, and the equivalent circuit model describes the dynamic characteristics of the battery by using an equivalent circuit diagram formed by basic circuit elements such as resistance, capacitance, voltage source and the like. Fig. 2 is an equivalent circuit diagram of the lithium battery of the present embodiment, including a 3 rd order RC network.
As shown in fig. 1, a method for dynamically adaptively estimating SOC of a battery according to the present invention includes the steps of:
s1, obtaining an equivalent circuit model of the battery according to an equivalent circuit of the lithium battery shown in FIG. 2, wherein the equivalent circuit model is specifically shown as the following formula:
。
wherein U (t) represents the terminal voltage value of the battery at the time t; uocv (SOC) is an open-circuit voltage value corresponding to the SOC value of the battery at the t moment, and is known by using an SOC-OCV curve; i (t) represents the current value of the battery at the time t; k is the order number of the RC network, which represents the kth order RC network, in this embodiment, k=1, 2,3; rk represents the resistance value of the kth order RC network; τk represents the time coefficient of the kth order RC network, τk=rk×ck, ck represents the capacitance value of the kth order RC network; r0 represents the internal resistance of the equivalent circuit; t represents a period, i.e. a time period of formula operation.
S2, obtaining model parameters of an equivalent circuit model through a load pulse power characteristic test working condition offline experiment (hybrid pulse power characterization test, HPPC).
The model parameters include: r0, R1, τ1, R2, τ2, R3, τ3.
S3, converting an equivalent circuit model of the battery into a state equation, wherein the state equation is specifically shown as the following formula:
y=Cx；
d(x)=Ax+Bu；
in the method, in the process of the invention,
；/>；/>；；
wherein u represents an input, in particular a current value of the battery; x represents the output, specifically the state value of the equivalent circuit model, namely the actual state value, and d (x) is the derivative of the actual state value x; u1, U2 and U3 are the voltage values of the RC networks of the 1 st order, the 2 nd order and the 3 rd order respectively; SOC is the SOC value of the battery; y is the terminal voltage value of the battery, namely the actual terminal voltage; q represents battery capacity; A. b, C are coefficients.
S4, establishing a state observer of the battery, wherein the state observer is used for outputting an estimated state value px. The error e between the actual state value x and the estimated state value px is fed back to the state observer, and the state observer is corrected by the error feedback. The specific mode of correction is as follows: the gain matrix L of the state observer is configured to minimize the error e, so that the estimated state value px output by the state observer approximates to the actual state value x, and the gain matrix L obtained by configuration is the initial optimal gain matrix L (0), and the error between the estimated state value px output by the state observer and the actual state value x is e (0) based on the optimal gain matrix L (0).
In the correction process of this embodiment, the error e may also have a suitable attenuation rate, where the attenuation rate is a rate at which the error becomes smaller, and the greater the attenuation rate, the faster the error becomes smaller.
A schematic of the state equations and state observer is shown in fig. 3. A schematic of the correction of the state observer, i.e. the error feedback system, is shown in fig. 4.
L is a gain matrix of the state observer, and a model formula of the state observer is specifically shown as follows:
d(px)=Apx+Bu+L(y-py)=Apx+Bu+L(y-Cpx)=(A-LC)px+Bu+Ly；
wherein px is an estimated state value output by the state observer, and d (px) is a derivative of the estimated state value px; py is the estimated terminal voltage of the battery, py=cpx.
Calculating an error e, e=x-px from the estimated state value px and the actual state value x, and obtaining:
d(e)=d(x)-d(px)=Ax+Bu-[(A-LC)px+Bu+Ly]=Ax-(A-LC)px-LCx=(A-LC)e；
where e is the error of the state observer and d (e) is the derivative of the error e.
Because the time-varying parameters of the battery are changed continuously along with the use, the gain matrix L of the state observer is also adaptively adjusted along with the change of the battery, so that the optimal observation effect is achieved.
S5, defining an adaptive factor sigma, wherein sigma is more than or equal to 0 and less than or equal to 1, and the calculation mode of the adaptive factor sigma is as follows:
if abs (e) is less than or equal to thr, σ=1/[ abs (e) +1]; if abs (e) > thr, σ=0.
Wherein thr is the set upper error limit; abs () is an absolute function; e is the error between the actual state value x and the estimated state value px.
That is, if the error e is greater than the set upper error limit thr, the gain matrix L is deemed unsuitable and should be discarded and reconfigured; if the error e is smaller than or equal to the set upper error limit thr, it is considered that, within the upper error limit thr, a larger error e indicates a smaller probability that the gain matrix L is optimally corrected, and a smaller adaptive factor σ is obtained, and a larger scale factor λ is obtained later.
Defining a scale factor lambda, wherein lambda is more than or equal to 0 and less than or equal to 1, and the error e is in direct proportion to the scale factor lambda, i.e. the larger the error e is, the larger the scale factor lambda is; the scale factor lambda is calculated by the self-adaptive factor sigma in the following calculation mode: λ=1- σ.
S6, as the use of the battery changes, the gain matrix L of the state observer is continuously subjected to iterative optimization; the t-th iteration mode is as follows:
taking the optimal gain matrix L (0) as an iteration center, and restraining by a scale factor lambda to obtain a gain matrix L (t) of the t-th iteration:
L(t)=L(0)+λ[L(t-1)-L(0)]；
wherein L (t) is the gain matrix of the t-th iteration; l (t-1) is the gain matrix of the last, i.e., the t-1 th iteration.
Substituting the gain matrix L (t) of the t-th iteration into the state observer, and calculating an error e (t) between an estimated state value px and an actual state value x output by the state observer after the t-th iteration;
if the error e (t) is smaller than the error e (0) calculated based on the iteration center, namely the optimal gain matrix L (0), namely e (t) < e (0), updating the optimal gain matrix L (0) into a gain matrix L (t) of the t-th iteration to obtain an updated optimal gain matrix L (0), and continuing the next iteration by taking the updated optimal gain matrix L (0) as the iteration center according to the mode of the step S6;
otherwise, the optimal gain matrix L (0) is not updated, and the next iteration is performed by taking the non-updated optimal gain matrix L (0) as an iteration center according to the mode of the step S6.
The optimal gain matrix L (0) is iteratively found continuously during operation.
S7, substituting the optimal gain matrix L (0) obtained in real time in the iteration process of the step S6 into a state observer to obtain the estimated state value px of the battery in real time.
In the invention, when the error e is larger, the obtained self-adaptive factor sigma is smaller, the scale factor lambda is larger, which is equivalent to the large neighborhood searching pace, so as to improve the searching speed; when the error e is smaller, the obtained self-adaptive factor sigma is larger, the scale factor lambda is smaller, which is equivalent to the smaller neighborhood searching steps, so that the searching precision is improved.
The invention introduces a dynamic self-adaptive estimation method based on errors, and in the continuous iterative updating process, the selection process and the optimizing process of the gain matrix are continuously and automatically adjusted along with the change of the battery, so that the estimation speed and the estimation precision of the battery state are improved.
According to the invention, a multi-order RC circuit model of a battery monomer or a battery module is established according to the temperature, current and voltage characteristics of the battery, a state observer of the battery is established according to the dynamic experimental data of the battery, the battery equivalent parameters after the battery cell is aged or abnormal are identified by using a recursive least square method on line in the operation process, the battery equivalent parameters are continuously and iteratively updated, the self-adaptive correction is used, the gain matrix of the state observer is dynamically and adaptively adjusted according to the difference between the actual terminal voltage and the estimated terminal voltage of the battery, the SOC estimation of the battery is iterated and quickly corrected, and the battery is quickly converged in a short time.
The dynamic self-adaptive estimation method is suitable for the application of random charging and discharging and severe fluctuation of current and voltage of electric vehicles, combines a high-precision SOC-OCV curve, considers various influencing factors such as temperature, current, internal resistance of a battery, total capacity, sensor drift and the like in a multi-scale manner, carries out self-adaptive correction of a state observer through error feedback of actual terminal voltage and estimated terminal voltage, carries out iterative optimization on a gain matrix of the state observer in real time, realizes high-precision estimation of the SOC, has stronger stability and convergence, and can quickly converge in a short time even if the SOC error is larger, namely the larger the SOC error is, the larger the feedback correction value is, the larger the correction speed is, and not only can reach higher precision, but also has higher robustness and stability.
Based on the method for dynamically and adaptively estimating the battery SOC, the SOC estimation precision can be further improved, and the SOC estimation precision of the whole working condition and the whole service life is improved to be within +/-2%. The problems of groveling in running, even over-charge and over-discharge, service life reduction and the like caused by inaccurate SOC estimation are remarkably reduced, customer satisfaction and user experience are improved, and product competitiveness is improved. The invention can correct the SOC estimation accuracy within 2%, and the current domestic and foreign existing technologies in the same field can only correct the SOC estimation accuracy within 5% -3%, compared with the domestic and foreign existing technologies, the SOC estimation accuracy of the invention is improved by at least 33%, and the calculation mode is (3% -2%)/3% = 33%.
The meaning of the related English abbreviations in the invention is shown as follows:
BMS refers to a battery management system.
SOC refers to the state of charge of a battery to reflect the remaining capacity of the battery, and is defined numerically as the ratio of the remaining capacity of the battery to the total capacity of the battery.
DOD refers to the depth of discharge of a battery, which is defined numerically as the ratio of the amount of discharge of the battery to the total capacity of the battery.
Uocv refers to the open circuit voltage value of the cell.
The SOC-OCV curve refers to a relationship between the SOC value of the battery and the open circuit voltage value of the battery.
Uocv (SOC) refers to an open circuit voltage value corresponding to the SOC value of the battery.
RC network refers to a circuit network composed of resistors and capacitors.
The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention, and any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (3)
1. A method for dynamically adaptively estimating battery SOC, comprising the steps of:
s1, establishing an equivalent circuit model of a battery;
s2, obtaining model parameters of an equivalent circuit model;
s3, converting the equivalent circuit model into a state equation, and obtaining an actual state value x of the battery according to model parameters;
s4, a state observer of the battery is established, the state observer is used for outputting an estimated state value px of the battery, and an error e between the actual state value x and the estimated state value px is fed back to the state observer and used for correcting the state observer; the specific mode of correction is as follows: configuring a gain matrix L of the state observer to minimize an error e, wherein the gain matrix L obtained by configuration is an initial optimal gain matrix L (0), and the error between an estimated state value px and an actual state value x output by the state observer is e (0) based on the optimal gain matrix L (0);
s5, defining a scale factor lambda, wherein the error e is in direct proportion to the scale factor lambda, namely, the larger the error e is, the larger the scale factor lambda is, the more 0 is less than or equal to lambda is less than or equal to 1;
s6, as the use of the battery changes, the gain matrix L of the state observer is continuously subjected to iterative optimization; the t-th iteration mode is as follows:
taking the optimal gain matrix L (0) as an iteration center, and restraining by a scale factor lambda to obtain a gain matrix L (t) of the t-th iteration:
L(t)=L(0)+λ[L(t-1)-L(0)]；
wherein L (t) is the gain matrix of the t-th iteration; l (t-1) is the gain matrix of the last, i.e., the t-1 th iteration;
substituting the gain matrix L (t) of the t-th iteration into the state observer, and calculating an error e (t) between an estimated state value px and an actual state value x output by the state observer after the t-th iteration;
if the error e (t) is smaller than the error e (0) calculated based on the iteration center, namely the optimal gain matrix L (0), namely e (t) < e (0), updating the optimal gain matrix L (0) into a gain matrix L (t) of the t-th iteration to obtain an updated optimal gain matrix L (0), and continuing the next iteration by taking the updated optimal gain matrix L (0) as the iteration center according to the mode of the step S6;
otherwise, the optimal gain matrix L (0) is not updated, the optimal gain matrix L (0) which is not updated is continuously used as an iteration center, and the next iteration is carried out in a mode of the step S6;
s7, substituting the optimal gain matrix L (0) obtained in real time in the iteration process of the step S6 into a state observer to obtain an estimated state value px of the battery in real time;
in step S1, the equivalent circuit model of the battery is represented by the following formula:
wherein U (t) represents the terminal voltage value of the battery at the time t; uocv (SOC) is an open circuit voltage value corresponding to the SOC value of the battery; r0 represents the internal resistance of the equivalent circuit; i (t) represents the current value of the battery at the time t; t represents a period, i.e., a time period of formula operation; r is Rk represents the resistance value of the kth order RC network; τk represents the time coefficient of the kth order RC network;
k is an order number of the RC network, representing a kth order RC network, k=1, 2.
In step S3, the equivalent circuit model is converted into a state equation, which is shown as follows:
y=Cx；
d(x)=Ax+Bu；
in the method, in the process of the invention,
wherein u represents an input, in particular a current value of the battery; x represents the output, specifically the state value of the equivalent circuit model, namely the actual state value; d (x) is the derivative of the actual state value x; uk is the voltage value of the kth order RC network, k=1, 2,..k; SOC is the SOC value of the battery; y is the terminal voltage value of the battery, namely the actual terminal voltage; q represents battery capacity; A. b, C are all coefficients;
in step S4, the model formula of the state observer is as follows:
d(px)=Apx+Bu+L(y-py)=Apx+Bu+L(y-Cpx)=(A-LC)px+Bu+Ly；
wherein px is an estimated state value output by the state observer, and d (px) is a derivative of the estimated state value px; py is the estimated terminal voltage of the battery, py=cpx;
in step S5, an adaptive factor sigma is defined, wherein sigma is more than or equal to 0 and less than or equal to 1, and the calculation mode of the adaptive factor sigma is as follows:
if abs (e) is less than or equal to thr, σ=1/[ abs (e) +1];
if abs (e) > thr, σ=0;
wherein thr is the set upper error limit; abs () is an absolute function; e is the error between the actual state value x and the estimated state value px;
scale factor λ=1- σ.
2. The method according to claim 1, wherein in step S4, an error e=x-px between the actual state value x and the estimated state value px is obtained:
d(e)=d(x)-d(px)=Ax+Bu-[(A-LC)px+Bu+Ly]=Ax-(A-LC)px-LCx=(A-LC)e；
where e is the error of the state observer and d (e) is the derivative of the error e.
3. The method for dynamically adaptively estimating a battery SOC according to claim 1, wherein in step S2, model parameters of an equivalent circuit model are obtained through experiments; the model parameters include: internal resistance R0 of the equivalent circuit, resistance Rk of the k-th order RC network in the equivalent circuit, and time coefficient τk of the k-th order RC network in the equivalent circuit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310498571.1A CN116224099B (en) | 2023-05-06 | 2023-05-06 | Method for dynamically and adaptively estimating battery SOC |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310498571.1A CN116224099B (en) | 2023-05-06 | 2023-05-06 | Method for dynamically and adaptively estimating battery SOC |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116224099A CN116224099A (en) | 2023-06-06 |
| CN116224099B true CN116224099B (en) | 2023-07-21 |
Family
ID=86571658
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310498571.1A Active CN116224099B (en) | 2023-05-06 | 2023-05-06 | Method for dynamically and adaptively estimating battery SOC |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116224099B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018163395A (en) * | 2017-03-24 | 2018-10-18 | 富士通株式会社 | Control method, control device and control program |
Family Cites Families (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0756452B2 (en) * | 1991-03-29 | 1995-06-14 | 日本電信電話株式会社 | State observation method |
| JPH08129420A (en) * | 1994-11-02 | 1996-05-21 | Nippon Telegr & Teleph Corp <Ntt> | Direction and position controller for tunnel robot and constitution method for n-dimensional linear discrete time system probability model |
| DE102004035858A1 (en) * | 2004-07-23 | 2006-02-16 | Robert Bosch Gmbh | State and parameter estimator with integral and differential component for electrical energy storage |
| CN205898985U (en) * | 2016-07-15 | 2017-01-18 | 南京航空航天大学 | Power lithium cell SOC estimates system for pure electric vehicles |
| JP7036605B2 (en) * | 2018-01-30 | 2022-03-15 | プライムアースＥｖエナジー株式会社 | Battery state estimation device and battery state estimation method |
| CN108872865B (en) * | 2018-05-29 | 2020-12-29 | 太原理工大学 | Lithium battery SOC estimation method for preventing filtering divergence |
| CN108594135A (en) * | 2018-06-28 | 2018-09-28 | 南京理工大学 | A kind of SOC estimation method for the control of lithium battery balance charge/discharge |
| CN110261779B (en) * | 2019-06-25 | 2021-07-27 | 西安石油大学 | Online collaborative estimation method for state of charge and state of health of ternary lithium battery |
| KR20210133744A (en) * | 2020-04-29 | 2021-11-08 | 한국전력공사 | Apparatus and method for Disturbance estimation |
| CN111812530A (en) * | 2020-06-22 | 2020-10-23 | 南京邮电大学 | Lithium battery state of charge estimation method |
| CN111830418B (en) * | 2020-07-09 | 2021-05-11 | 南京航空航天大学 | SOC estimation method considering external environment influence of soft package battery |
| CN112310991B (en) * | 2020-11-30 | 2022-07-29 | 上海海事大学 | Energy storage converter improved finite set model prediction control method based on disturbance observation |
| CN113702838B (en) * | 2021-08-27 | 2024-03-12 | 四川长虹电源有限责任公司 | Lithium ion battery state of charge estimation method based on disturbance observer |
| CN113777510A (en) * | 2021-09-07 | 2021-12-10 | 国网江苏省电力有限公司电力科学研究院 | Lithium battery state of charge estimation method and device |
| CN113805075A (en) * | 2021-09-15 | 2021-12-17 | 上海电机学院 | BCRLS-UKF-based lithium battery state of charge estimation method |
| CN114910796A (en) * | 2022-05-31 | 2022-08-16 | 上海电机学院 | Lithium ion battery state of charge estimation method based on MIAUKF algorithm |
| CN115902647B (en) * | 2023-02-23 | 2023-05-16 | 新乡医学院 | Intelligent battery state monitoring method |
-
2023
- 2023-05-06 CN CN202310498571.1A patent/CN116224099B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018163395A (en) * | 2017-03-24 | 2018-10-18 | 富士通株式会社 | Control method, control device and control program |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116224099A (en) | 2023-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110221222B (en) | Battery safety cut-off voltage prediction method and device and battery management system | |
| CN107368619B (en) | Extended Kalman filtering SOC estimation method | |
| CN111007399B (en) | Lithium battery state of charge prediction method based on improved generation countermeasure network | |
| CN109633479B (en) | Lithium battery SOC online estimation method based on embedded type volume Kalman filtering | |
| CN114355211A (en) | Lithium ion power battery residual capacity estimation method | |
| CN111913109B (en) | Method and device for predicting peak power of battery | |
| CN112858920B (en) | SOC estimation method of all-vanadium redox flow battery fusion model based on adaptive unscented Kalman filtering | |
| CN112557906A (en) | SOC and capacity online joint estimation method in full life cycle of power battery | |
| CN111914485B (en) | Adaptive power battery SOC estimation method and system based on fractional order technology | |
| CN116113837A (en) | Method for estimating state of charge of battery | |
| CN115932605A (en) | Method and device for correcting residual electric quantity of energy storage battery and battery management system | |
| CN114740357A (en) | Joint estimation method for branch current, charge state and power state of parallel battery pack | |
| CN112098893B (en) | Internal short-circuit fault diagnosis method based on initial capacity difference and suitable for electric automobile | |
| Goh et al. | Successive-approximation algorithm for estimating capacity of Li-ion batteries | |
| CN112462282A (en) | Method for determining real-time state of charge of battery pack based on mechanism model | |
| CN114371408B (en) | Method for estimating state of charge of battery, method and device for extracting charging curve | |
| CN114217234B (en) | IDE-ASRCKF-based lithium ion battery parameter identification and SOC estimation method | |
| CN106772097B (en) | Method for correcting SOC (State of Charge) by using charger | |
| Baba et al. | State of charge estimation of lithium-ion battery using Kalman filters | |
| CN116224099B (en) | Method for dynamically and adaptively estimating battery SOC | |
| CN116520167A (en) | SOC estimation method and device for self-adjusting zero point, electronic equipment and storage medium | |
| Bobobee et al. | Improved fixed range forgetting factor-adaptive extended Kalman filtering (FRFF-AEKF) algorithm for the state of charge estimation of high-power lithium-ion batteries | |
| CN115113053A (en) | Lithium battery soc estimation method based on high-adaptivity filtering algorithm | |
| CN115856681A (en) | Battery SOC estimation method based on EKF adaptive temperature regulation | |
| CN115575835A (en) | Lithium ion battery peak power estimation method |
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 |