CN115327389A - Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering - Google Patents

Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering Download PDF

Info

Publication number
CN115327389A
CN115327389A CN202211054450.XA CN202211054450A CN115327389A CN 115327389 A CN115327389 A CN 115327389A CN 202211054450 A CN202211054450 A CN 202211054450A CN 115327389 A CN115327389 A CN 115327389A
Authority
CN
China
Prior art keywords
lithium battery
soc
battery
genetic algorithm
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.)
Pending
Application number
CN202211054450.XA
Other languages
Chinese (zh)
Inventor
江兵
杨怡
仲美秋
杨阳
王子博
巢一帆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN202211054450.XA priority Critical patent/CN115327389A/en
Publication of CN115327389A publication Critical patent/CN115327389A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/385—Arrangements for measuring battery or accumulator variables
    • G01R31/387—Determining ampere-hour charge capacity or SoC
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/385—Arrangements for measuring battery or accumulator variables
    • G01R31/387—Determining ampere-hour charge capacity or SoC
    • G01R31/388—Determining ampere-hour charge capacity or SoC involving voltage measurements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
    • H03H17/00—Networks using digital techniques
    • H03H17/02—Frequency selective networks
    • H03H17/0202—Two or more dimensional filters; Filters for complex signals
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
    • H03H17/00—Networks using digital techniques
    • H03H17/02—Frequency selective networks
    • H03H17/0202—Two or more dimensional filters; Filters for complex signals
    • H03H2017/0205—Kalman filters

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Evolutionary Biology (AREA)
  • Computational Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Optimization (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Mathematical Analysis (AREA)
  • Genetics & Genomics (AREA)
  • General Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Operations Research (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • Algebra (AREA)
  • Physiology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Geometry (AREA)
  • Tests Of Electric Status Of Batteries (AREA)

Abstract

The invention discloses a lithium battery SOC estimation method based on a genetic algorithm improved double-Kalman filtering, which comprises the following steps: establishing an equivalent circuit model and a dynamic system equation of the lithium battery; performing OCV test on the lithium battery, and fitting test data to obtain a relation curve of open-circuit voltage and SOC; optimizing the parameter identification process in the lithium battery equivalent circuit model by using a genetic algorithm to obtain optimal identification parameters; and establishing a discrete nonlinear system equation of the dual Kalman filter according to the lithium battery equivalent circuit model, and performing dual Kalman filter iteration by using the optimal identification parameters, the terminal voltage and the current of the lithium battery. The lithium battery SOC estimation method provided by the invention obtains the optimized model parameters by real number encoding genetic algorithm identification, the voltage output characteristics of the optimized model are basically consistent with the actual voltage output characteristics of the battery, the identification precision is higher, and the identified parameters are substituted into the double-Kalman filtering to carry out SOC prediction, so that the accuracy of the estimation model is greatly improved.

Description

Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering
Technical Field
The invention relates to a power lithium battery SOC technology, in particular to a lithium battery SOC estimation method based on double Kalman filtering improved by a genetic algorithm.
Background
While environmental pollution is becoming more serious, the problem of shortage of non-renewable energy sources such as petroleum and coal is becoming more prominent, and these are all great challenges faced by the global automobile industry. In the face of the increasingly severe environmental and energy crisis, more and more people are beginning to pay attention to the research and development of electric vehicles. The electric automobile has the characteristics of small pollution, low noise and high efficiency, can effectively relieve the practical situation of energy shortage in China, reduces the urban atmospheric pollution degree, and has a good development prospect. The lithium ion battery is one of the key parts of the electric automobile as a main energy source of the electric automobile, and the characteristics of the lithium ion battery directly influence the performance of the electric automobile.
The battery SOC estimation is one of core technologies of a battery management system, and is a basis for preventing overcharge and overdischarge of a battery and realizing balanced management, and the higher the SOC estimation precision is, the higher the service efficiency of the battery is, and the longer the service life of the battery is. Meanwhile, the SOC is also important information transmitted to the vehicle control system during the driving of the vehicle, and the vehicle control system estimates the remaining driving distance and the energy fed back to the battery by the regenerative braking system by obtaining the remaining battery capacity and the change of the battery capacity from the SOC. Therefore, theoretical research and engineering realization of SOC estimation have long-term significance for development of electric automobiles.
Common SOC estimation methods include an ampere-hour integration method, an open-circuit voltage method, a neural network method, a kalman filter method, and the like. Ampere-hour integration, also known as coulomb counting, is a classical estimation method that integrates the current and estimates SOC by accumulating the charge that is going in and out. In practical use, however, certain errors are generated in battery sampling and battery capacity change, the problem of how to select sampling precision and intervals is also faced, meanwhile, the sampling precision and the intervals are also influenced by factors such as temperature and charging and discharging multiplying power, and the errors become larger and larger along with the lapse of time; the open-circuit voltage method has the characteristics of simplicity, practicability and strong operability, and has the defect that the SOC value can be accurately estimated only by stable open-circuit voltage, so that the method needs to be stood for a long time, the electric automobile is frequently started, and the open-circuit voltage is difficult to stabilize; the neural network is a novel algorithm for processing a nonlinear system by simulating a human brain and neurons, for the neural network method, the larger the input data is, the more diversified the parameters are, and the more accurate the obtained SOC estimation is, but the neural network method needs a large amount of calculation, and the cost is higher compared with other methods; the Kalman filtering is suitable for linear, discrete and finite-dimension systems, can estimate the SOC and can also give the error range, so the Kalman filtering algorithm is widely applied, but the Kalman filtering algorithm has higher dependence on a battery model and higher requirement on the parameter precision of the battery model, and the parameter precision of the existing battery model cannot meet the high requirement.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the problems, the invention aims to provide a lithium battery SOC estimation method based on a genetic algorithm improved double Kalman filtering, the identification process of the equivalent circuit model parameters of the lithium battery is optimized through a legacy algorithm, the identification precision is improved, and then the iterative recursive calculation is carried out through the double Kalman filtering, so that the high-precision SOC online estimation is realized.
The technical scheme is as follows: the invention discloses a lithium battery SOC estimation method based on a genetic algorithm improved double Kalman filtering, which comprises the following steps:
step 1, establishing a lithium battery equivalent circuit model and a dynamic system equation;
step 2, performing OCV test on the lithium battery, measuring the open-circuit voltage value of the lithium battery under different SOC conditions, and fitting the test data to obtain a relation curve of the open-circuit voltage and the SOC;
step 3, optimizing the parameter identification process in the lithium battery equivalent circuit model by using a genetic algorithm to obtain optimal identification parameters;
and 4, establishing a dual-Kalman filter discrete nonlinear system equation according to the lithium battery equivalent circuit model, and performing dual-Kalman filter iteration by using the optimal identification parameters, the terminal voltage and the current of the lithium battery to realize the online estimation of the SOC of the lithium battery under the circulating working condition.
Further, in the step 1, the lithium battery equivalent circuit model adopts a second-order RC lithium battery equivalent circuit model, and parameters in the second-order RC lithium battery equivalent circuit model comprise electrochemical polarization internal resistance R 1 Electrochemical polarization capacitance C 1 Concentration polarization resistance R 2 Sum concentration polarization capacitance C 2 。
Further, the dynamic system equation expression is:
Figure BDA0003824419900000021
in the formula of U OC For terminal voltage, U, across the lithium battery L Is the terminal voltage of the battery, R 0 Is an ohmic resistance value, R 1 、R 2 Respectively electrochemical polarization internal resistance and concentration polarization resistance, C 1 、C 2 Respectively electrochemical polarization capacitance and concentration polarization capacitance, U 0 、U 1 、U 2 Are each R 0 、R 1 、R 2 The terminal voltage at two ends of the battery, SOC (0) is the initial value of the state of charge of the lithium battery, SOC (t) is the residual electric quantity of the battery at the moment t, I L Is the charging and discharging current of the battery, and t is the charging and discharging time。
Further, the step 3 specifically includes: acquiring charge and discharge data of the lithium battery, including current and voltage, calculating the SOC value of the battery at each time interval, and performing pulse discharge to obtain an ohmic resistance value R under an offline condition 0 The relationship between the open-circuit voltage OCV and the SOC, and the ohmic resistance value R 0 Inputting the charging and discharging data into a genetic algorithm for optimization to obtain a parameter R 1 、R 2 、C 1 And C 2 And identifying the result value.
Further, the step of optimizing the genetic algorithm specifically comprises:
step 301, mapping a search space of a parameter to be identified to a genetic space, presetting a population scale for each group of parameter values which are a chromosome or an individual in a solution space, and generating an initial population by using a random sequence;
step 302, encoding the initial population and converting the initial population into a corresponding binary code; the binary codes represent genes of individuals, and any parameter to be identified is called as a gene;
step 303, hybridizing the current population, exchanging partial gene segments and generating new individuals; the new individual is a child;
step 304, simulating variation behavior in the biological evolution process, and randomly generating variation in the offspring gene;
step 305, decoding the filial generation after crossing and mutation to obtain the numerical value of the filial generation;
step 306, substituting the decoded offspring into a fitness function, judging whether the individual meets a preset condition, and if so, taking the individual which best meets the condition as an identification parameter; if not, performing elite selection on the generated filial generation to generate a new population, and skipping to the step 302 to the step 306 to perform loop operation until an individual meets the adaptive condition or the reproduction algebra exceeds a preset algebra, and stopping the loop operation.
Further, the step 4 specifically includes:
step 401, establishing a state equation and an output equation of the lithium battery model, which are respectively:
Figure BDA0003824419900000031
U L (k)＝U OC (SOC(k))-U 1 (k)-U 2 (k)-R 0 I L (k-1)+v(k)
in the formula, τ 1 And τ 2 Representing two different time constants, U 1 And U 2 Respectively a polarization capacitance C 1 And C 2 SOC (k) is an estimated value of SOC at time k; i is L (k) The current at the moment k is the input variable of the state equation; q N Is the rated capacity of the battery, T is the sampling period, U L (k) For the total voltage estimate of polarization at time k, U 1 (k) And U 2 (k) Respectively at time k 1 And R 2 V (k) and ω (k) are Gaussian noise;
step 402, establishing a dual kalman filter discrete nonlinear system equation, wherein the expression is as follows:
Figure BDA0003824419900000041
in the formula, x k Is a state vector, f (x) k-1 ，θ k-1 ，u k-1 ) Is a process equation, θ k-1 Is a model parameter vector, u k-1 As an input vector, ω k-1 For process excitation noise, z k To observe the vector, h (x) k-1 ，θ k-1 ，u k-1 ) To observe the equation, v k-1 To observe noise;
and 403, using the obtained voltage and current data, performing optimization identification parameters by using a genetic algorithm under iterative recursion calculation of the double-Kalman filter, and repeating iteration to finally complete online estimation of the SOC under the cyclic working condition.
Has the advantages that: compared with the prior art, the invention has the following remarkable advantages: the lithium battery SOC estimation method provided by the invention optimizes the equivalent circuit model parameters by using a real number coding genetic algorithm, identifies to obtain the optimized model parameters, has high identification precision because the voltage output characteristic of the optimized model is basically consistent with the actual voltage output characteristic of the battery, and substitutes the identified parameters into the double-Kalman filtering to predict the SOC, thereby greatly improving the accuracy of the estimation model.
Drawings
FIG. 1 is a flow chart of a lithium battery SOC estimation method of the invention;
FIG. 2 is a schematic diagram of an equivalent circuit model of a lithium battery according to the present invention;
FIG. 3 is a calibration curve diagram of OCV-SOC of the lithium battery in the embodiment;
FIG. 4 is a graph of the results of terminal voltage after parameter identification by genetic algorithm;
FIG. 5 is a graph of the error results of terminal voltage after parameter identification by genetic algorithm;
FIG. 6 is a diagram of an SOC estimation of the modified DEKF.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments.
Referring to fig. 1, the present embodiment provides a lithium battery SOC estimation method based on a genetic algorithm improved dual kalman filter, where the estimation method includes the following steps:
step 1, establishing a lithium battery equivalent circuit model and a dynamic system equation;
as shown in fig. 2, the lithium battery equivalent circuit model adopts a second-order RC lithium battery equivalent circuit model, and parameters in the second-order RC lithium battery equivalent circuit model include electrochemical polarization internal resistance R 1 Electrochemical polarization capacitance C 1 Concentration polarization resistance R 2 Sum concentration polarization capacitance C 2 。U OC The terminal voltage of two ends of the power lithium battery is shown, and has a certain functional relation with the SOC; r 0 The ohmic resistance of the battery is composed of electrode materials, electrolyte and other resistances; in the RC loop, the polarization process of the battery is simulated in a mode of overlapping two resistance-capacitance links, and the method is used for simulating the process that the battery tends to be stable after the discharge is finished and the voltage is suddenly changed.
The above dynamic system equation expression is:
Figure BDA0003824419900000051
in the formula of U OC For terminal voltage, U, across the lithium battery L Is the terminal voltage of the battery, R 0 Is an ohmic resistance value, R 1 、R 2 Respectively electrochemical polarization internal resistance and concentration polarization resistance, C 1 、C 2 Electrochemical polarization capacitance and concentration polarization capacitance, U 0 、U 1 、U 2 Are each R 0 、R 1 、R 2 The terminal voltage at two ends of the battery, SOC (0) is the initial value of the state of charge of the lithium battery, SOC (t) is the residual electric quantity of the battery at t moment, I L Is the battery charge-discharge current, and t is the charge-discharge time.
And 2, performing OCV test on the lithium battery at normal temperature, measuring the open-circuit voltage value of the lithium battery under different SOC conditions, and fitting the test data to obtain a relation curve between the open-circuit voltage and the SOC, wherein the SOC and the OCV have a linear fitting relation as shown in FIG. 3.
The OCV test implementation process of the lithium battery comprises the following steps:
1) Placing the lithium battery into a temperature control box at normal temperature for fully standing, then performing constant current discharge at a current of 0.3A, and placing the lithium battery to a cut-off voltage (2.8V), wherein the SOC is considered to be 0 at the moment;
2) After sufficiently standing for 2 hours, the battery was charged at a constant current and a constant voltage of 0.3A current and a cutoff current of 0.03A, and after reaching a cutoff voltage (4.2V) and until the charging current was 0.03A or less, the battery was allowed to stand for 2 hours with SOC = 100%;
3) OCV test was started: the OCV test was completed by standing for 10 seconds, followed by constant current discharge for 2 hours, and then standing for 2 hours, at which time the battery SOC was considered to have dropped to 90%, and cycling was performed ten times until the battery SOC was 0.
In this embodiment, the open-circuit voltage value OCV and SOC obtained by the test are subjected to exponential fitting by using an MATLAB tool box to obtain a relationship curve between the open-circuit voltage and the SOC.
Step 3, optimizing the parameter identification process in the lithium battery equivalent circuit model by using a genetic algorithm to obtain optimal identification parameters;
the step of optimizing the parameter identification process in the lithium battery equivalent circuit model by using the genetic algorithm specifically comprises the following steps: under the UDDS working condition, the charging and discharging data of the lithium battery, including current and voltage, can be obtained, the SOC value of the lithium battery in each time period can be calculated by utilizing an ampere-hour integration method, and pulse discharging is carried out to obtain the ohmic resistance value R under the offline condition 0 The relationship between the open-circuit voltage OCV and the SOC, and the ohmic resistance value R 0 Inputting the charging and discharging data into a genetic algorithm for optimization to obtain a parameter R 1 、R 2 、C 1 And C 2 And identifying the result value. As shown in fig. 4 and 5, the voltage output characteristics of the identified optimized model substantially match the actual voltage output characteristics of the battery, and the error of the terminal voltage is within 2%, so that the identification precision is high.
The principle of calculating the SOC value by using the ampere-hour integration method is as follows: when the initial SOC value of the battery is known, the initial SOC value is used to subtract the varying charge amount and the rated capacity Q of the battery N The ratio Δ SOC, the remaining is the current SOC value, and the calculation formula is as follows:
Figure BDA0003824419900000061
in the formula, SOC 0 Is the SOC value at the initial time, eta is the battery charge-discharge efficiency, Q N Is the rated capacity of the battery,
Figure BDA0003824419900000062
the amount of charge released and charged by the battery over time t is indicated, and the current is defined as positive during discharge and negative during charge.
The ohmic resistance value is obtained by pulse discharge, and the calculation formula is as follows:
Figure BDA0003824419900000063
in the formula of U D Indicating the voltage value, U, at which the zero input response has just started C Indicating the final voltage value of the zero state response;
the optimization of the genetic algorithm specifically comprises the following steps:
step 301, mapping a search space of a parameter to be identified to a genetic space, presetting a population scale for each group of parameter values which are a chromosome or an individual in a solution space, and generating an initial population by using a random sequence;
step 302, encoding the initial population and converting the initial population into a corresponding binary code; the binary codes represent genes of individuals, and any parameter to be identified is called as a gene;
step 303, hybridizing the current population, exchanging partial gene segments and generating new individuals; the new individual is a child;
step 304, simulating variation behavior in the biological evolution process, and randomly generating variation in the offspring gene;
step 305, decoding the filial generation after crossing and mutation to obtain the numerical value of the filial generation;
step 306, substituting the decoded offspring into a fitness function, judging whether the individual meets a preset condition, and if so, taking the individual which best meets the condition as an identification parameter; if not, performing elite selection on the generated filial generation to generate a new population, and skipping to the step 302 to the step 306 to perform loop operation until an individual meets the adaptive condition or the reproduction algebra exceeds a preset algebra, and stopping the loop operation.
And 4, establishing a dual-Kalman filter discrete nonlinear system equation according to the lithium battery equivalent circuit model, and performing dual-Kalman filter iteration by using the optimal identification parameters, the terminal voltage and the current of the lithium battery to realize the online estimation of the SOC of the lithium battery under the circulating working condition. The prediction result of the SOC is shown in FIG. 6, the result of estimating the SOC by the improved double-Kalman filtering with the genetic algorithm is better than that of the traditional double-Kalman filtering, and as shown in Table 1, the root mean square error, the mean square error and the average absolute error of the estimated SOC after the improvement are far smaller than those of the traditional estimation algorithm. The method specifically comprises the following steps:
step 401, establishing a state equation and an output equation of the lithium battery model, which are respectively:
Figure BDA0003824419900000071
U L (k)＝U OC (SOC(k))-U 1 (k)-U 2 (k)-R 0 I L (k-1)+v(k)
in the formula, τ 1 And τ 2 Representing two different time constants, U 1 And U 2 Are respectively a polarization capacitance C 1 And C 2 SOC (k) is an estimated value of SOC at time k; i is L (k) The current at the moment k is the input variable of the state equation; q N Is the rated capacity of the battery, T is the sampling period, U L (k) Is an estimate of the total voltage of polarization at time k, U 1 (k) And U 2 (k) Are respectively at the k time R 1 And R 2 V (k) and ω (k) are Gaussian noise;
step 402, establishing a dual kalman filter discrete nonlinear system equation, wherein the expression is as follows:
Figure BDA0003824419900000072
in the formula, x k Is a state vector, f (x) k-1 ，θ k-1 ，u k-1 ) Is a process equation, θ k-1 Is a model parameter vector, u k-1 As an input vector, ω k-1 For process excitation noise, z k To observe the vector, h (x) k-1 ，θ k-1 ，u k-1 ) To observe the equation, v k-1 To observe noise;
and step 403, using the obtained voltage and current data, performing optimized identification parameters by using a genetic algorithm under iterative recursion calculation of the double-Kalman filter, and repeatedly iterating to finally complete online estimation of the SOC under the circulation condition.
Specifically, the SOC is estimated by using DEKF in the steps, the principle is that an ampere-hour integral algorithm is used as a basis, an SOC estimation value is obtained through an ampere-hour integral method, then online Kalman correction is carried out on an ampere-hour integral model by using the SOC value obtained through extended Kalman filtering, and finally the value of the SOC estimated by the double Kalman filtering algorithm is obtained, and the specific steps are as follows:
(a) Setting initial value x of state of system k|k-1 Sum covariance P k-1|k-1 。
(b) Calculating the prior estimated values of the system state and the noise covariance at the current moment by the set state initial value and covariance:
x k|k-1 ＝A k-1 x k-1|k-1 +B k-1 I k-1
Figure BDA0003824419900000081
wherein x is k|k-1 A priori estimate representing SOC, I k-1 Representing inputs from a previous moment of the system, P k|k-1 A priori estimate, Q, representing the covariance of the system k-1 Representing a noise covariance matrix of a model in an ampere-hour integration method algorithm;
(c) Obtaining a Kalman gain matrix K of the current moment k The expression is:
Figure BDA0003824419900000082
wherein, C k Representing the system observation matrix, R k Representing an error covariance matrix of the SOC estimated by the extended Kalman filtering algorithm;
(d) Obtaining the output of the system through a state equation, correcting the system output and the output obtained through measurement through a Kalman filtering algorithm prior estimation value to obtain an updated state value and a system covariance value at the current moment, wherein the expression is as follows:
y k ＝x k-1|k-1
z k ＝y EKF -y k
x k|k ＝x k|k-1 +K k z k
P k|k ＝(E-K k C k )P k|k-1
wherein, y k SOC estimation, y, representing an ampere-hour integration algorithm EKF A value representing the SOC estimation of the battery model by the extended Kalman filtering algorithm; x is the number of k|k Representing the state update value of SOC, E representing the identity matrix, P k|k An update value representing a covariance matrix;
(e) Updating the system state and covariance, wherein the expression is as follows:
x k-1|k-1 ＝x k|k
P k|k ＝P k1|k-1
(f) Returning to the step (a) and carrying out new iterative operation, and iteratively estimating the most output of the current system through the steps (a) - (e).
TABLE 1
Figure BDA0003824419900000091

Claims (6)

1. The lithium battery SOC estimation method based on the improved dual Kalman filtering of the genetic algorithm is characterized by comprising the following steps of:
step 1, establishing a lithium battery equivalent circuit model and a dynamic system equation;
step 2, performing OCV test on the lithium battery, measuring the open-circuit voltage value of the lithium battery under different SOC conditions, and fitting the test data to obtain a relation curve of the open-circuit voltage and the SOC;
step 3, optimizing the parameter identification process in the lithium battery equivalent circuit model by using a genetic algorithm to obtain optimal identification parameters;
and 4, establishing a dual-Kalman filter discrete nonlinear system equation according to the lithium battery equivalent circuit model, and performing dual-Kalman filter iteration by using the optimal identification parameters, the terminal voltage and the current of the lithium battery to realize the online estimation of the SOC of the lithium battery under the circulating working condition.
2. The method of claim 1, wherein in step 1, the lithium battery equivalent circuit model adopts a second-order RC lithium battery equivalent circuit model, and the parameters in the second-order RC lithium battery equivalent circuit model comprise electrochemical polarization internal resistance R 1 Electrochemical polarization capacitance C 1 Concentration polarization resistance R 2 Sum concentration polarization capacitance C 2 。
3. The lithium battery SOC estimation method of claim 2, wherein the dynamic system equation expression is:
Figure FDA0003824419890000011
in the formula of U OC For terminal voltage, U, across the lithium battery L Is the terminal voltage of the battery, R 0 Is an ohmic resistance value, R 1 、R 2 Electrochemical polarization internal resistance and concentration polarization resistance, C 1 、C 2 Respectively electrochemical polarization capacitance and concentration polarization capacitance, U 0 、U 1 、U 2 Are each R 0 、R 1 、R 2 The terminal voltage at two ends of the battery, SOC (0) is the initial value of the state of charge of the lithium battery, SOC (t) is the residual electric quantity of the battery at the moment t, I L Is the battery charge-discharge current, and t is the charge-discharge time.
4. The lithium battery SOC estimation method according to claim 2, wherein the step 3 specifically includes: acquiring charge and discharge data of the lithium battery, including current and voltage, calculating the SOC value of the battery at each time interval, and performing pulse discharge to obtain an ohmic resistance value R under an offline condition 0 The relationship between the open-circuit voltage OCV and the SOC, and the ohmic resistance value R 0 Inputting the charging and discharging data into a genetic algorithm for optimization to obtain a parameter R 1 、R 2 And C 1 、C 2 And identifying the result value.
5. The method for estimating the SOC of the lithium battery as claimed in claim 4, wherein the step of optimizing the genetic algorithm specifically includes:
step 301, mapping a search space of parameters to be identified to a genetic space, presetting a population scale for each set of parameter values which are a chromosome or an individual in a solution space, and generating an initial population by using a random sequence;
step 302, encoding the initial population and converting the initial population into a corresponding binary code; the binary codes represent genes of individuals, and any parameter to be identified is called as a gene;
step 303, hybridizing the current population, exchanging partial gene segments and generating new individuals; the new individual is a child;
step 304, simulating variation behaviors in the biological evolution process, and randomly generating variation in offspring genes;
step 305, decoding the filial generation after crossing and mutation to obtain the numerical value of the filial generation;
step 306, substituting the decoded offspring into a fitness function, judging whether the individual meets a preset condition, and if so, taking the individual which best meets the condition as an identification parameter; if not, performing elite selection on the generated filial generation to generate a new population, and skipping to the step 302 to the step 306 to perform loop operation until an individual meets the adaptive condition or the reproduction algebra exceeds a preset algebra, and stopping the loop operation.
6. The lithium battery SOC estimation method according to any one of claims 1 to 5, wherein the step 4 specifically includes:
step 401, establishing a state equation and an output equation of the lithium battery model, which are respectively:
Figure FDA0003824419890000021
U L (k)＝U OC (SOC(k))-U 1 (k)-U 2 (k)-R 0 I L (k-1)+υ(k)
in the formula, τ 1 And τ 2 Representing two different time constants, U 1 And U 2 Respectively a polarization capacitance C 1 And C 2 SOC (k) is an estimated value of SOC at time k; i is L (k) Current at time k, I L (k) Is an input variable of the state equation; q N Is the rated capacity of the battery, T is the sampling period, U L (k) Is an estimate of the total voltage of polarization at time k, U 1 (k) And U 2 (k) Respectively at time k 1 And R 2 V (k) and ω (k) are Gaussian noise;
step 402, establishing a dual kalman filter discrete nonlinear system equation, wherein the expression is as follows:
Figure FDA0003824419890000031
in the formula, x k Is a state vector, f (x) k-1 ，θ k-1 ，u k-1 ) Is a process equation, θ k-1 Is a model parameter vector, u k-1 As input vector, ω k-1 For process excitation noise, z k To observe the vector, h (x) k-1 ，θ k-1 ，u k-1 ) To observe the equation, v k-1 To observe noise;
and 403, using the obtained voltage and current data, performing optimization identification parameters by using a genetic algorithm under iterative recursion calculation of the double-Kalman filter, and repeating iteration to finally complete online estimation of the SOC under the cyclic working condition.
CN202211054450.XA 2022-08-31 2022-08-31 Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering Pending CN115327389A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211054450.XA CN115327389A (en) 2022-08-31 2022-08-31 Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211054450.XA CN115327389A (en) 2022-08-31 2022-08-31 Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering

Publications (1)

Publication Number Publication Date
CN115327389A true CN115327389A (en) 2022-11-11

Family

ID=83928117

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211054450.XA Pending CN115327389A (en) 2022-08-31 2022-08-31 Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering

Country Status (1)

Country Link
CN (1) CN115327389A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116699415A (en) * 2023-05-26 2023-09-05 云储新能源科技有限公司 Method and system for estimating electric quantity of dynamic reconfigurable battery system and electronic equipment

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116699415A (en) * 2023-05-26 2023-09-05 云储新能源科技有限公司 Method and system for estimating electric quantity of dynamic reconfigurable battery system and electronic equipment
CN116699415B (en) * 2023-05-26 2024-06-11 云储新能源科技有限公司 Method and system for estimating electric quantity of dynamic reconfigurable battery system and electronic equipment

Similar Documents

Publication Publication Date Title
CN111090047B (en) Lithium battery health state estimation method based on multi-model fusion
CN112034356B (en) GP-UKF-based online SOH estimation method for electric vehicle power battery
CN110488194B (en) Lithium battery SOC estimation method and system based on electrochemical impedance model
CN112051511A (en) Power battery state of health estimation method and system based on multichannel technology
CN113702843B (en) Lithium battery parameter identification and SOC estimation method based on suburb optimization algorithm
CN112989690B (en) Multi-time-scale state-of-charge estimation method for lithium battery of hybrid electric vehicle
CN111812515A (en) XGboost model-based lithium ion battery state of charge estimation
CN113917334B (en) Battery health state estimation method based on evolution LSTM self-encoder
CN113109717B (en) Lithium battery state of charge estimation method based on characteristic curve optimization
CN112269133B (en) SOC estimation method based on pre-charging circuit model parameter identification
CN112858916A (en) Battery pack state of charge estimation method based on model and data drive fusion
CN114660464A (en) Lithium ion battery state of charge estimation method
CN112462282A (en) Method for determining real-time state of charge of battery pack based on mechanism model
CN113484771A (en) Method for estimating wide-temperature full-life SOC and capacity of lithium ion battery
CN114861545A (en) Lithium battery SOP online estimation method based on RNN neural network and multi-parameter constraint
CN112946481A (en) Based on federation H∞Filtering sliding-mode observer lithium ion battery SOC estimation method and battery management system
CN116125278A (en) Lithium battery SOC estimation method and system based on LSTM-EKF algorithm
CN116298904A (en) Method for jointly estimating SOC and SOH of lithium battery based on fractional order multi-innovation double unscented Kalman filtering
CN113608126B (en) Lithium battery SOC online prediction method under different temperatures
CN115327389A (en) Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering
CN113495214B (en) Super-capacitor state-of-charge estimation method based on temperature change model
CN112946480B (en) Lithium battery circuit model simplification method for improving SOC estimation real-time performance
CN116819364A (en) Battery health determination method, application method, electronic device and storage medium
CN116679213A (en) SOH estimation method for electric vehicle power battery based on integrated deep learning
CN116449219A (en) Lithium battery model parameter identification method for improving VFFRLS

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