CN114935722B - Lithium battery side end collaborative management method based on digital twin - Google Patents
Lithium battery side end collaborative management method based on digital twin Download PDFInfo
- Publication number
- CN114935722B CN114935722B CN202210601270.2A CN202210601270A CN114935722B CN 114935722 B CN114935722 B CN 114935722B CN 202210601270 A CN202210601270 A CN 202210601270A CN 114935722 B CN114935722 B CN 114935722B
- Authority
- CN
- China
- Prior art keywords
- soc
- battery
- state
- capacity
- initial value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 title claims abstract description 36
- 229910052744 lithium Inorganic materials 0.000 title claims abstract description 36
- 238000007726 management method Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 93
- 238000005516 engineering process Methods 0.000 claims abstract description 5
- 238000004364 calculation method Methods 0.000 claims description 36
- 238000004422 calculation algorithm Methods 0.000 claims description 33
- 230000008859 change Effects 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 17
- 230000010354 integration Effects 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 14
- 238000007599 discharging Methods 0.000 claims description 12
- 238000012937 correction Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 9
- 230000010287 polarization Effects 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000004088 simulation Methods 0.000 claims description 4
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 208000028659 discharge Diseases 0.000 description 23
- 238000002474 experimental method Methods 0.000 description 19
- 238000010586 diagram Methods 0.000 description 11
- GELKBWJHTRAYNV-UHFFFAOYSA-K lithium iron phosphate Chemical compound [Li+].[Fe+2].[O-]P([O-])([O-])=O GELKBWJHTRAYNV-UHFFFAOYSA-K 0.000 description 9
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 description 4
- 229910001416 lithium ion Inorganic materials 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- HCUOEKSZWPGJIM-YBRHCDHNSA-N (e,2e)-2-hydroxyimino-6-methoxy-4-methyl-5-nitrohex-3-enamide Chemical compound COCC([N+]([O-])=O)\C(C)=C\C(=N/O)\C(N)=O HCUOEKSZWPGJIM-YBRHCDHNSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
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]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. 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]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
- G01R31/3828—Arrangements for monitoring battery or accumulator variables, e.g. SoC using current integration
-
- 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]
- G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/18—Details relating to CAD techniques using virtual or augmented reality
-
- 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
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Secondary Cells (AREA)
Abstract
The invention discloses a lithium battery side end collaborative management method based on digital twinning, which comprises the following steps: taking a lithium battery pack to be managed as a physical entity, and constructing a lithium battery pack model by utilizing a digital twin technology aiming at the physical entity to serve as a virtual entity; s2, establishing a lithium battery pack digital model; s3, dynamically correcting an OCV-SOC table of the open-circuit voltage method in an idle state. According to the invention, the digital twin-based lithium battery side end collaborative management method is adopted, a twin model of the battery is established based on digital twin, and the management is more real-time and accurate through mutual feedback of data between the twin model and an actual battery; and the initial value of the SOC estimation is optimized by using a digital model of the battery, and meanwhile, the method for estimating the SOC is more reliable and accurate by combining a capacity learning mode.
Description
Technical Field
The invention relates to a battery management technology, in particular to a lithium battery side end collaborative management method based on digital twinning.
Background
In recent years, the battery industry has evolved rapidly. Particularly, lithium batteries are a wide choice for power batteries of new energy automobiles due to their own characteristics and advantages.
Because the battery is used as a power source of the electric automobile, the running environment is very harsh. It is therefore necessary to provide a battery management system for monitoring it in actual operation. There is a non-negligible management parameter in the battery management system: state of Charge (SOC) of the battery, which refers to the percentage of the current amount of Charge remaining in the battery to the total amount of Charge. As one of important references of the new energy automobile endurance mileage, the estimation accuracy directly influences the control experience of a user on the automobile.
The residual condition of the battery charge quantity can not be directly obtained and can only be calculated by parameters such as voltage, current and the like. Therefore, the accuracy of SOC estimation is a hot topic today, and whether the accuracy of SOC estimation is significant for the whole battery management system
However, the existing Battery Management System (BMS) mounted on the terminal has limited computing power, and when the characteristic parameters of the battery change, the battery cannot be identified, and the battery can only be calculated according to the set parameters, so that the SOC estimation accuracy is reduced, and meanwhile, the battery management system has the problems of single function, incapability of sharing data, insufficient data processing capability and the like.
Disclosure of Invention
The invention aims to provide a lithium battery side end collaborative management method based on digital twinning, which is characterized in that a twin model of a battery is established based on the digital twinning, and the management is more real-time and accurate through mutual feedback of data between the twin model and an actual battery; and the initial value of the SOC estimation is optimized by using a digital model of the battery, and meanwhile, the method for estimating the SOC is more reliable and accurate by combining a capacity learning mode.
In order to achieve the above purpose, the invention provides a lithium battery edge collaborative management method based on digital twinning, which comprises the following steps:
s1, taking a lithium battery pack to be managed as a physical entity, and constructing a lithium battery pack model by utilizing a digital twin technology aiming at the physical entity to serve as a virtual entity;
s2, establishing a lithium battery pack digital model
S20, collecting the end voltage of a single string of batteries in the lithium battery pack, the temperature information of a battery pack and the working current through a battery state monitoring unit in a charge and discharge state
S21, establishing a battery equivalent circuit model by using a Thevenin model;
s22, parameter identification based on FFRLS algorithm to obtain forgetting factors;
s23, combining a recursive least square method with forgetting factors and an equivalent circuit model to obtain parameter identification results under different SOCs;
s24, describing a system equation based on the FFRLS algorithm to obtain the system equation;
s25, obtaining an open-circuit voltage by means of terminal voltage identification based on the calculation result of the step S24, and obtaining an OCV-SOC table in the latest idle state by utilizing the edge of the mobile phone;
s3, OCV-SOC meter of dynamic correction open-circuit voltage method in idle state
S30, analyzing an SOC estimation error source to obtain an initial value estimation of the SOC and a battery capacity change;
s31, processing errors by combining a digital model of the lithium battery pack;
s32, estimating the SOC based on the battery equivalent circuit model and an ampere-hour integration method.
Preferably, in step S21, the equivalent circuit model equation is as follows:
U t ＝U oc -U 1 -IR 0 (2)
i is input current, U t Is a terminal voltage, U oc Represents an open circuit voltage, R 0 Represents ohmic internal resistance of battery, R 1 For the polarization internal resistance of the battery, C 1 U is the polarization capacitance of the battery 1 Representing the voltage drop across the RC parallel link,represents U 1 The derivative of time t.
Preferably, the step S22 specifically includes the following steps:
s220, recursive least square algorithm containing forgetting factors
For the following system:
wherein y is k As an output variable at the kth acquisition instant,for the data variable, θ, at the kth acquisition time k For the parameter variable at the kth acquisition time, e k Is system noise;
the following formula was used for calculation:
wherein K is k Algorithm gain vector for kth acquisition time, P k-1 The covariance matrix of the identification parameters at the k-1 acquisition time is shown, mu is a forgetting factor, and I is an identity matrix;
s221, selecting a forgetting factor by using a simulation experiment.
Preferably, the step S23 specifically includes the following steps:
s230, obtaining the expression of the 1 expression and the 2 expression on the complex frequency domain through Laplace transformation:
defining the output of the system as U t (s)-U oc (s) the input of the system is I(s), then the transfer equation of the system is:
and then utilizing a bilinear transformation formula:
substituting the formula 9 into the formula 8, and converting the transfer function of the system into the z domain, and obtaining after simplification:
and (3) making:
equation 10 is transformed into:
and then obtaining a discretized result through inverse transformation of z:
U t (k)＝U oc (k)-b 3 U oc (k-1)+b 3 U t (k-1)+b 1 I(k)+b 2 I(k-1) (15)
because the sampling time interval is short, U oc (k) And U oc (k-1) are approximately equal, then equation 15 is written as:
U t (k)＝(1-b 3 )U oc (k)+b 3 U t (k-1)+b 1 I(k)+b 2 I(k-1) (16)
thereby obtaining a relation between the open-circuit voltage and the terminal voltage, wherein the terminal voltage is used as a parameter identification result.
Preferably, the step S24 specifically includes the following steps:
definition y k As an output variable at the kth acquisition instant,for the data variable, θ, at the kth acquisition time k For the parameter variable at the kth acquisition time, writing the system equation into a matrix form: />Finishing 16 yields the system equation:
y k ＝U t (k) (17)
θ k ＝[(1-b 3 )U oc (k) b 3 b 1 b 2 ] (19)
substituting 11, equation 12 and equation 13 into the obtained parameter U oc ，R 0 And R is 1 。
Preferably, the step S30 specifically includes the following steps:
the open circuit voltage method and the ampere-hour integration method are adopted to estimate the SOC:
taking 1 second as an integral time interval, dividing into a front time slice and a rear time slice, carrying out acquisition operation for the front 500ms, and starting algorithm calculation in the rear 500ms, so that in the calculation time interval 1s, firstly, obtaining the SOC obtained by an open circuit voltage method ocv And one second of last round of last obtained SOC k-1 Result twoThe average value is taken to determine an initial value:
SOC k ＝(SOC k-1 +SOC ocv )/2 (21)
and integrating the current of the next 500ms by utilizing an ampere-hour integration method to obtain a change value delta SOC:
wherein t represents the time slice of 1s, Q FC Representing the rated capacity of the battery;
subtracting the change value from the initial value to obtain:
SOC k+1 ＝SOC k -ΔSOC ocv (23)
the error sources are estimated SOC initial value and battery capacity variation.
Preferably, the step S31 specifically includes the following steps:
s310, initial value error processing
The current changes along with the time change in the working state, the last calculation is read in the common process in the working state to be the initial value of this time, and the current is set to be 0 when the working state reaches the emptying condition; when the working state reaches the full condition, setting the working state to be 100%; in the idle state, correcting an initial value according to an open circuit voltage method combined with an equivalent circuit model;
s311, capacity error handling
And obtaining an OCV-SOC table obtained by utilizing digital twin model identification in an idle state through error analysis, and then carrying out table lookup to obtain an SOC as an initial value, wherein a result obtained by utilizing an ampere-hour integrating method in a working state is used as the initial value of the next calculation, and the actual capacity of the battery is recalculated by utilizing a capacity learning mode to correct the capacity error.
Preferably, the initial value correction by the open circuit voltage method in step S310 includes the following steps:
when the battery state monitoring unit of the physical entity collects working current, terminal voltage and the like and the last calculated SOC result, the working current, the terminal voltage and the like and the last calculated SOC result are uploaded to the virtual entity, the virtual entity calculates to obtain a new OCV-SOC table according to an equivalent circuit model and a recursive least square method with forgetting factors, the new OCV-SOC table is fed back to the main control unit of the physical entity, and the main control unit utilizes the new OCV-SOC table and combines an initial value calculated by an open circuit voltage method to finish correction.
Preferably, the step S311 specifically includes the following steps:
s3110, judging whether the discharging state reaches a discharging zone bit, namely whether the voltage of the single battery is smaller than a single string undervoltage threshold, if so, executing a step S3111, otherwise, executing a step S3113;
s3111, triggering a capacity learning sign, and when the capacity learning sign is valid, reaching a full charge state, completing one-time capacity learning:
wherein I is current, t is time from start of charging to full charge state, ΔQ FC Is the learned capacity;
if the state of charge is interrupted during the capacity learning process, the current battery state CurState enters a discharging or idle state, and if the capacity learning flag fails, step S3113 is executed;
s3112, update capacity:
Q FC ＝ΔQ FC (25)
s1113, finishing capacity learning.
Preferably, the step S32 specifically includes the following steps:
s320, initial value setting:
n＝0 (27)
wherein SOC is ocv The representative system is obtained by open circuit voltage table lookup during initialization, and SOC k Representing the last calculated SOC result of the last second when the system is normally operated, n generationsCounting the marks in the second of the table, and determining the number of times of calculation;
s321, state value processing:
wherein t represents the 1000ms time slice, I represents the current at the 1s, Q FC Representing the rated capacity of the battery;
s322, calculating the state of charge SOC of the battery k+1 ：
SOC k+1 ＝SOC 0 -ΔSOC (29)
n＝n+1 (30)
Wherein SOC is 0 Representing the calculated initial value at this time, wherein delta SOC represents the calculated result value in the step S321, and n is increased by 1;
s323, judging whether n is more than or equal to 2,
if n is not greater than 2, entering initial value processing:
judging whether the battery is in an idle state or not, and judging whether the current is in a current dead zone or not according to the judgment standard;
if the state is the idle state, firstly, looking up a table according to the existing OCV-SOC table, then judging the difference value calculated in the step S322, if the difference value is more than 30%, representing that the existing OCV-SOC table is invalid, then, carrying out parameter estimation of open-circuit voltage by utilizing a digital twin model, feeding back and dynamically updating the OCV-SOC table of hardware by using the model in the mobile phone terminal, and then, carrying out an open-circuit voltage method by using the new table to obtain an initial value as the next calculation;
if not, reading the initial value calculated in the step S322 as the next calculation, and entering the step S321 to continue calculation after obtaining the initial value;
if n is greater than or equal to 2, go to S324;
s324, the calculation result is output to the outside for display, and n is reset to 0:
SOC＝SOC k (31)；
s325, judging whether the voltage in the discharge state is smaller than an undervoltage threshold value, if yes, entering a step S326, and if not, entering a step S327;
s326, triggering a capacity learning mark in a emptying state, entering a capacity learning process, judging whether the current capacity learning mark is 1 and whether the current battery is in a charging state, if the current capacity learning mark is simultaneously satisfied with starting the capacity learning, the battery must be always in the charging state in the learning process, and resetting the capacity learning mark to 0 when the battery enters other states, wherein the charge amount obtained from emptying to full charge is the new capacity, and updating the battery charge state SOC of S321 k+1 ；
S327, finishing the formation of the fusion algorithm.
Therefore, the invention has the following beneficial effects:
analyzing error sources of the existing SOC estimation method, adopting an improved SOC estimation fusion algorithm, and updating an OCV-SOC table of an open-circuit voltage method by combining a digital twin model to further determine an initial value SOC of the battery 0 And simultaneously, carrying out real-time correction on the capacity in the ampere-hour integration method by utilizing a capacity learning algorithm. Based on the method, experimental verification is carried out on the SOC estimated by the ampere-hour integration method and the improved fusion algorithm, and the result proves that the estimation precision of the improved fusion algorithm is higher than that of the former, so that various errors of SOC estimation are effectively reduced.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a graph of recognition results with forgetting factor of 0.998;
FIG. 3 is a graph of recognition results with forgetting factor of 0.99;
FIG. 4 is a graph of recognition results with forgetting factor of 0.95;
FIG. 5 is a graph of recognition results with forgetting factor of 0.9;
FIG. 6 is an enlarged view of the result of forgetting factor of 0.998;
FIG. 7 is an enlarged view of the result of forgetting factor of 0.99;
FIG. 8 is a graph of actual terminal voltage versus identified terminal voltage;
FIG. 9 is a graph of end voltage error results;
FIG. 10 is a graph comparing errors of the conventional ampere-hour method and the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings, and it should be noted that, while the present embodiment provides a detailed implementation and a specific operation process on the premise of the present technical solution, the protection scope of the present invention is not limited to the present embodiment.
FIG. 1 is a flow chart of the present invention, as shown in FIG. 1, comprising the steps of:
s1, taking a lithium battery pack to be managed as a physical entity, and constructing a lithium battery pack model at the edge of a mobile phone by utilizing a digital twin technology aiming at the physical entity to serve as a virtual entity; in this embodiment, a 18650 type lithium iron phosphate battery pack composed of 16 strings of rated capacities of 1.6Ah is selected as a parameter identification object, and the parameter table of the single battery is shown in table 1.
Table 1 monomer lithium iron phosphate Battery parameter table
S2, establishing a lithium battery pack digital model
S20, collecting the end voltage of a single string of batteries in the lithium battery pack, the temperature information of a battery pack and the working current through a battery state monitoring unit in a charge and discharge state
S21, establishing a battery equivalent circuit model by using a Thevenin model to obtain the relation between terminal voltage and open-circuit voltage;
preferably, in step S21, the equivalent circuit model equation is as follows:
U t ＝U oc -U 1 -IR 0 (2)
i is input current, Ut is terminal voltage, uoc is open-circuit voltage, R0 is ohmic internal resistance of the battery, R1 is polarized internal resistance of the battery, C1 is polarized capacitance of the battery, U1 is voltage drop of RC parallel link,representing the derivative of U1 over time t.
S22, parameter identification based on FFRLS algorithm to obtain forgetting factors;
preferably, the step S22 specifically includes the following steps:
s220, recursive least square algorithm containing forgetting factors
For the following system:
wherein y is k As an output variable at the kth acquisition instant,for the data variable, θ, at the kth acquisition time k For the parameter variable at the kth acquisition time, e k Is system noise;
the following formula was used for calculation:
wherein K is k Algorithm gain vector for kth acquisition time, P k-1 Covariance matrix of identification parameters at k-1 acquisition time, μ is forgetting factor, and I is identity matrix；
S221, selecting a forgetting factor by using a simulation experiment;
in the recursive least square algorithm, a forgetting factor is a very important parameter, which represents the forgetting degree of the algorithm on the previous identification result, and determines the influence degree of a new data sampling point on the parameter identification result. The forgetting factor has a value ranging from 0 to 1. When the forgetting factor takes 1, the representative algorithm does not have any forgetting function and is degraded into a least square algorithm. When the forgetting factor is 0, the representative algorithm forgets all the identification results before forgetting, and only adopts the data at the current moment to carry out parameter identification. Normally, neither 0 nor 1 is the optimal value of the forgetting factor. When the forgetting factor takes 1 or the value is too large, the identification result of the parameter is insensitive to the input change of the system, and sometimes the model parameter at the current moment is changed, but the algorithm still takes a message from the previous identification result, so that the identified parameter cannot be changed along with the change of the model state, a data saturation phenomenon is generated, and the accuracy of parameter identification is reduced. When the forgetting factor takes 0 or the value is too small, the algorithm can refer to too few data points, and a large error can be generated in the identification result. Therefore, the proper value of the forgetting factor is selected, and the accuracy and the stability of the parameter identification result of the lithium ion battery model are greatly affected.
In order to obtain a proper forgetting factor, the influence of the forgetting factor on the parameter identification result is verified, and a series of simulation verification is performed. Firstly, terminal voltage and current data obtained by a constant current emptying experiment of a 18650 lithium iron phosphate battery with 1A are selected as data for parameter identification, then parameter identification with forgetting factors of 0.998,0.99,0.95 and 0.9 is carried out, an identification result diagram with forgetting factors of 0.998 is shown in fig. 2, an identification result diagram with forgetting factors of 0.99 is shown in fig. 3, an identification result diagram with forgetting factors of 0.95 is shown in fig. 4, an identification result diagram with forgetting factors of 0.9 is shown in fig. 5, the ordinate represents ohmic resistance obtained by identification, and the abscissa represents calculated iteration times. It can be clearly seen from the plotted curves of the ohmic resistors that the recognition stability of the ohmic resistors is poor when the forgetting factors are 0.95 and 0.9, and the recognition stability of the ohmic resistors is high when the forgetting factors are 0.998 and 0.99. The partial curves of the same iteration times of the ohmic resistance when the forgetting factor is 0.998 and 0.99 are amplified, the ohmic resistance with the forgetting factor of 0.998 is amplified to obtain a result diagram, the ohmic resistance with the forgetting factor of 0.99 is amplified to obtain a result diagram, and the result diagram is shown in fig. 6 and 7, wherein in the same iteration process from 300 to 750, the identification result with the forgetting factor of 0.99 is a plurality of negative numbers, the ohmic resistance with the forgetting factor of 0.998 is changed in a trend way along with the increase of the iteration times, and the ohmic resistance with the forgetting factor of 0.998 is changed in an irregular way, so that the identification effect of the forgetting factor of 0.998 is better than that of 0.99. The actual terminal voltage and the identification terminal voltage diagram of fig. 8, the terminal voltage error result diagram of fig. 9 are shown in fig. 8 and 9, when the forgetting factor is 0.998, the comparison error of the identified terminal voltage and the actual terminal voltage is shown in the diagram, when the forgetting factor takes 0.998, the terminal voltage error is basically between-0.0006 and 0.0006, the identification result of the parameter is relatively accurate, the rapid tracking capability is high, the dynamic characteristic in the working process of the battery can be well reflected, therefore, 0.998 is selected as the forgetting factor, and the parameter identification calculation of the recursive least square method is carried out according to the forgetting factor.
S23, combining a recursive least square method with forgetting factors and an equivalent circuit model to obtain parameter identification results under different SOCs;
preferably, the step S23 specifically includes the following steps:
s230, obtaining the expression of the 1 expression and the 2 expression on the complex frequency domain through Laplace transformation:
defining the output of the system as U t (s)-U oc (s) the input of the system is I(s), then the transfer equation of the system is:
and then utilizing a bilinear transformation formula:
substituting the formula 9 into the formula 8, and converting the transfer function of the system into the z domain, and obtaining after simplification:
and (3) making:
equation 10 is transformed into:
and then obtaining a discretized result through inverse transformation of z:
U t (k)＝U oc (k)-b 3 U oc (k-1)+b 3 U t (k-1)+b 1 I(k)+b 2 I(k-1) (15)
because the sampling time interval is short, U oc (k) And U oc (k-1) are approximately equal, then equation 15 is written as:
U t (k)＝(1-b 3 )U oc (k)+b 3 U t (k-1)+b 1 I(k)+b 2 I(k-1) (16)
thereby obtaining a relation between the open-circuit voltage and the terminal voltage, wherein the terminal voltage is used as a parameter identification result.
Experimental example:
the voltage data of the lithium battery before and after standing in the discharging process is firstly obtained through a pulse discharging experiment (HPPC), and the battery terminal voltage can be regarded as open circuit voltage at the moment because the battery can effectively eliminate the battery polarization condition after standing for a long time and no current passes through during standing, so that the relationship table of the OCV and the SOC is obtained.
The test object is a lithium iron phosphate battery with an actual capacity of 1.5Ah, and the nominal voltage is 3.2V. The flow of the experiment can be expressed as the following steps:
firstly, preparing a lithium iron phosphate single battery in a full charge state;
secondly, placing the lithium ion battery in a laboratory environment at 25 ℃ for 2 hours to achieve electrochemical balance and eliminate polarization;
thirdly, constant current discharge is carried out according to the current of 1A, the discharge time is 4 minutes and 30 seconds, namely 5 percent of SOC is discharged, and the charge quantity of 0.075Ah is discharged;
fourth, judge whether the battery is empty, judge whether the standard is that the total time of experiment reaches 20 hours 30 minutes, namely discharge 100% SOC,1.5Ah of electric charge; if yes, executing the sixth step, and if not executing the fifth step;
fifthly, stopping the discharging operation, quietly placing the lithium ion battery for 1 hour, and waiting for the polarization reaction to finish, wherein terminal voltage data is recorded every 30 minutes in the hour;
sixth, the experiment was ended.
And drawing a pulse experiment voltage curve graph of the lithium iron phosphate single battery according to terminal voltage data recorded in the experiment, as shown in figures 3-10.
Table 2 shows the relationship between OCV and SOC
As shown in Table 3-2, a continuous discharge test of a full-charged battery was then performed, similarly, a 1A constant current discharge was performed at room temperature of 25 ℃,the constant volume (10% SOC) records terminal voltage data, and the experimental mode is similar to the HPPC pulse mode, except that no standing operation is performed, and the continuous discharge is performed, so the experimental mode is not described. And the relation between the terminal voltage and the open-circuit voltage in the equivalent circuit model of the section 3.1.1 is the formula (3-2), and then the parameter identification results of the resistance and the like under different SOCs are obtained according to the forgetting factor obtained in the section, the parameter identification method and the equivalent circuit model of the section 3.1.1. Since the parameter changes faster in the later stage of discharge and the error of identification is larger, the identification result reaches 30% of SOC, and the parameter identification results under different SOCs are shown in tables 3-3. Wherein U is t For recorded terminal voltage, U ocv 、R 0 、R 1 The parameter identification result is as follows:
table 3 shows the result of parameter identification under different SOC conditions
And combining the relationship between the OCV and the SOC of the table 2 with the open circuit voltage identification result of the table 3 to obtain an open circuit voltage error comparison table under different SOCs.
Table 4 shows comparison of open circuit voltage error for different SOCs
As shown in Table 4, the parameters were identified as U ocv The error of the open-circuit voltage obtained by the pulse discharge experiment is smaller, so that the result obtained by the parameter identification can be considered to be reliable, and the method can be applied to the subsequent SOC estimation algorithm.
The purpose of obtaining a corresponding relation table between open circuit voltage and SOC is to prepare for dynamic correction of the OCV-SOC table of the open circuit voltage method in the subsequent idle state.
S24, describing a system equation based on the FFRLS algorithm to obtain the system equation;
preferably, the step S24 specifically includes the following steps:
definition y k As an output variable at the kth acquisition instant,for the data variable, θ, at the kth acquisition time k For the parameter variable at the kth acquisition time, writing the system equation into a matrix form: />Finishing 16 yields the system equation:
y k ＝U t (k) (17)
θ k ＝[(1-b 3 )U oc (k) b 3 b 1 b 2 ] (19)
substituting 11, equation 12 and equation 13 to obtain parameter R 0 And R is 1 。
S25, calculating an SOC estimation value by an open circuit voltage method based on the calculation result of the step S24 to obtain an OCV-SOC table in a charge-discharge state;
s3, OCV-SOC meter of dynamic correction open-circuit voltage method in idle state
S30, analyzing an SOC estimation error source to obtain an initial value estimation of the SOC and a battery capacity change;
preferably, the step S30 specifically includes the following steps:
the open circuit voltage method and the ampere-hour integration method are adopted to estimate the SOC:
taking 1 second as an integral time interval, dividing into a front time slice and a rear time slice, carrying out acquisition operation for the front 500ms, and starting algorithm calculation in the rear 500ms, so that in the calculation time interval 1s, firstly, obtaining the SOC obtained by an open circuit voltage method ocv And one second of last round of last obtained SOC k-1 The average of the results is taken to determine an initial value:
SOC k ＝(SOC k-1 +SOC ocv )/2 (21)
and integrating the current of the next 500ms by utilizing an ampere-hour integration method to obtain a change value delta SOC:
wherein t represents the time slice of 1s, Q FC Representing the rated capacity of the battery;
subtracting the change value from the initial value to obtain:
SOC k+1 ＝SOC k -ΔSOC ocv (23)
as can be seen from the above, two sources of error: the initial value of SOC has large estimation error, and Q in state value processing FC The rated capacity of the battery cannot be dynamically changed along with the excessive cycle times of the battery, and errors of capacity are caused by static writing death, and the errors greatly affect the final SOC estimated value, namely, the error sources are the initial value estimation of the SOC and the change of the battery capacity.
S31, processing errors by combining a digital model of the lithium battery pack;
preferably, the step S31 specifically includes the following steps:
s310, initial value error processing
The current changes along with the time change in the working state, the last calculation is read in the common process in the working state to be the initial value of this time, and the current is set to be 0 when the working state reaches the emptying condition; when the working state reaches the full condition, setting the working state to be 100%; in the idle state, correcting an initial value according to an open circuit voltage method combined with an equivalent circuit model;
preferably, the initial value correction by the open circuit voltage method in step S310 includes the following steps:
when the battery state monitoring unit of the physical entity collects working current, terminal voltage and the like and the last calculated SOC result, the working current, the terminal voltage and the like and the last calculated SOC result are uploaded to the virtual entity, the virtual entity calculates to obtain a new OCV-SOC table according to an equivalent circuit model and a recursive least square method with forgetting factors, the new OCV-SOC table is fed back to the main control unit of the physical entity, and the main control unit utilizes the new OCV-SOC table and combines an initial value calculated by an open circuit voltage method to finish correction.
Table 5 shows an OCV-SOC meter
As can be seen from Table 5, the table look-up method is as follows:
the first step: u for judging that OCV falls in table ocv Which section is listed, assuming that OCV is U ocv1 And U ocv2 Between them.
And a second step of: calculating OCV and U ocv1 ，U ocv2 The purpose of (1) is to determine that the target OCV is more toward U based on the magnitude of the deviation ocv1 Or U ocv2 Closer U ocv So that the occupied weight is large, weight l l 、l 2 The calculation is as follows:
ΔOCV1＝|OCV-U oCV1 | (32)
ΔOCV2＝|OCV-U OCV2 | (33)
and a third step of: lookup U from table ocv1 、U ocv1 Corresponding SOC ocv1 、SOC ocv2 Calculating the result SOC corresponding to the OCV by using the weight of the second step ocv ：
SOC OCV ＝l 2 ·SOC OCV1 +l 1 ·SOC OCV2 (36)
Thereby obtaining the corrected SOC ocv . And takes this as an initial value.
S311, capacity error handling
And obtaining a 0CV-SOC table obtained by utilizing digital twin model identification in an idle state through error analysis, and then carrying out table lookup to obtain an SOC as an initial value, wherein a result obtained by utilizing an ampere-hour integrating method in a working state is used as the initial value of the next calculation, and the actual capacity of the battery is recalculated by utilizing a capacity learning mode to correct the capacity error.
Preferably, the step S311 specifically includes the following steps:
s3110, judging whether the discharging state reaches a discharging zone bit, namely whether the voltage of the single battery is smaller than a single string undervoltage threshold, if so, executing a step S3111, otherwise, executing a step S3113;
s3111, triggering a capacity learning sign, and when the capacity learning sign is valid, reaching a full charge state, completing one-time capacity learning:
wherein I is current, t is time from start of charging to full charge state, ΔQ FC Is the learned capacity;
if the state of charge is interrupted during the capacity learning process, the current battery state CurState enters a discharging or idle state, and if the capacity learning flag fails, step S3113 is executed;
s3112, update capacity:
Q FC ＝ΔQ FC (25)
s1113, finishing capacity learning.
To verify the capacity learning effect, the following experiment was developed:
the experimental object: the lithium iron phosphate battery with the rated capacity of 1.6Ah is obtained by capacity learning in advance, and the actual capacity is 1.5Ah.
First experiment: carrying out discharge experiments on the battery pack in a full charge state at a constant current of 1A, predicting 1.6 hours to empty according to rated capacity, and predicting 1.5 hours to empty according to actual capacity;
second experiment: charging the battery pack in the empty state with a constant current of 0.5A, wherein the battery pack is predicted to be full for 3.2 hours according to the rated capacity and is predicted to be full for 3 hours according to the actual capacity;
each set of experiments was performed three times and the time required for each battery to actually empty and the time required for each battery to actually fill were recorded as shown in tables 7 and 8.
Table 7 is a comparison table of vent time errors
Table 8 is a comparative table of charging time errors
As is apparent from comparison of the actual capacity blow-down error and the rated capacity blow-down error of the experimental results, the full time and the blow-down time of the battery pack are obviously closer to the corresponding time of the learned capacity, which indicates that the capacity learning is effective.
S32, estimating the SOC based on the battery equivalent circuit model and an ampere-hour integration method.
Preferably, the step S32 specifically includes the following steps:
s320, initial value setting:
n＝0 (27)
wherein SOC is ocv The representative system is obtained by open circuit voltage table lookup during initialization, and SOC k Representing the last calculated SOC result of the last second when the system is in normal operation, and n represents the counting mark in the second to determine the calculation times;
s321, state value processing:
wherein t represents the 1000ms time slice, I represents the current at the 1s, Q FC Representing the rated capacity of the battery;
s322, calculating the state of charge SOC of the battery k+1 ：
SOC k+1 ＝SOC 0 -ΔSOC (29)
n＝n+1 (30)
Wherein SOC is 0 Representing the calculated initial value at this time, wherein delta SOC represents the calculated result value in the step S321, and n is increased by 1;
s323, judging whether n is more than or equal to 2,
if n is not greater than 2, entering initial value processing:
judging whether the battery is in an idle state or not, and judging whether the current is in a current dead zone or not according to the judgment standard;
if the state is the idle state, firstly, looking up a table according to the existing OCV-SOC table, then judging the difference value calculated in the step S322, if the difference value is more than 30%, representing that the existing OCV-SOC table is invalid, then, carrying out parameter estimation of open-circuit voltage by utilizing a digital twin model, feeding back and dynamically updating the OCV-SOC table of hardware by using the model in the mobile phone terminal, and then, carrying out an open-circuit voltage method by using the new table to obtain an initial value as the next calculation;
if not, reading the initial value calculated in the step S322 as the next calculation, and entering the step S321 to continue calculation after obtaining the initial value;
if n is greater than or equal to 2, go to S324;
s324, the calculation result is output to the outside for display, and n is reset to 0:
SOC＝SOC k (31)；
s325, judging whether the voltage in the discharge state is smaller than an undervoltage threshold value, if yes, entering a step S326, and if not, entering a step S327;
s326, triggering a capacity learning mark in a emptying state, entering a capacity learning flow, and judging that the current capacity learning mark isIf not, 1 is obtained, and the current battery is in a charging state, if the current battery is in a charging state, the battery must be in a charging state all the time in the learning process, and if the current battery is in other states, the capacity learning flag is reset to 0, the charge amount obtained from emptying to filling is new capacity, and the battery charging state SOC of S321 is updated k+1 ；
S327, finishing the formation of the fusion algorithm.
Experimental example:
the experiment is to compare 3 different SOC estimation methods: a discharge experiment method; wang Feng modified ampere-hour integration [57]; the SOC estimation method adopted in the paper comprises the following steps: the SOC estimation method combines a digital twin open-circuit voltage method to correct an initial value and fuses a capacity correction ampere-hour integration method. Since the discharge experiment method is a relatively accurate method for estimating the residual capacity of the battery, the discharge experiment method is used as an actual estimation value of the SOC.
The experimental study subjects were: 16 strings of lithium iron phosphate batteries with an actual capacity of 1.5Ah.
The experimental procedure was as follows:
firstly, preparing a lithium iron phosphate single battery in a full charge state;
secondly, placing the lithium ion battery in a laboratory environment at 25 ℃ for 2 hours to achieve electrochemical balance and eliminate polarization;
thirdly, constant-current discharge is carried out according to the current of 1A, the discharge time is 9 minutes (constant volume discharge), and the discharge is kept stand for 10 minutes;
fourth, judge whether the battery is empty, judge whether the standard is that discharge time reaches 1 hour 30 minutes, namely discharge the electric charge quantity of 100% SOC; if yes, executing the step 6, and if not, executing the step 5;
fifthly, recording the SOC value at the moment;
sixth, the experiment was ended.
By modifying the SOC estimation mode, each mode performs the above experimental steps, and the experimental results are recorded and plotted respectively.
Table 9 is a comparison table of SOC estimation errors
Fig. 10 is a comparison diagram of errors in the conventional ampere-hour method and the present invention, and as can be seen from fig. 10 in combination with table 9, the dashed curve is the SOC error calculated by the ampere-hour integration method only, and the solid curve is the SOC error calculated by the algorithm based on the battery digital model and the ampere-hour integration method. It can be seen that in the course of the decrease of the abscissa charge amount from 100% to 0, the SOC error obtained by the ampere-hour integration method becomes larger and larger, and the peak value reaches 19, and the analysis is that the cause of the accumulated error is the cause. The improved SOC estimation can be seen from the graph that the SOC estimation error approaches 0 in the process of the charge amount from 100% to 80%, then gradually increases to peak 6, and then decreases again in the process of the charge amount from 20% to 0, and is closer to the actual SOC estimation value. Compared with curve error values of the two methods, the improved algorithm error precision is improved by 8%. Therefore, the improved algorithm is beneficial to reducing the SOC calculation error, and a more accurate SOC estimation value can be obtained.
Therefore, the invention adopts the lithium battery side end collaborative management method based on digital twin, builds a twin model of the battery based on digital twin, and ensures that management is more real-time and accurate through mutual feedback of data between the twin model and an actual battery; and the initial value of the SOC estimation is optimized by using a digital model of the battery, and meanwhile, the method for estimating the SOC is more reliable and accurate by combining a capacity learning mode.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention and not for limiting it, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that: the technical scheme of the invention can be modified or replaced by the same, and the modified technical scheme cannot deviate from the spirit and scope of the technical scheme of the invention.
Claims (8)
1. A lithium battery side end collaborative management method based on digital twinning is characterized in that: the method comprises the following steps:
s1, taking a lithium battery pack to be managed as a physical entity, and constructing a lithium battery pack model by utilizing a digital twin technology aiming at the physical entity to serve as a virtual entity;
s2, establishing a digital twin model of the lithium battery pack;
s20, acquiring the terminal voltage of a single string of batteries in the lithium battery pack, the temperature information of a battery pack and the working current through a battery state monitoring unit in a charge-discharge state;
s21, establishing a battery equivalent circuit model by using a Thevenin model to obtain a circuit equation;
s22, parameter identification based on FFRLS algorithm to obtain forgetting factors;
s23, combining a recursive least square method with forgetting factors and an equivalent circuit model to obtain parameter identification results under different SOCs;
s24, describing a system equation based on the FFRLS algorithm to obtain the system equation;
s25, obtaining an open-circuit voltage by means of terminal voltage identification based on the calculation result of the step S24, and obtaining an OCV-SOC table in the latest idle state by using the mobile phone terminal;
s3, dynamically correcting an OCV-SOC table of an open-circuit voltage method in an idle state;
s30, analyzing an SOC estimation error source to obtain an initial value estimation of the SOC and a battery capacity change;
the step S30 specifically includes the following steps:
the open circuit voltage method and the ampere-hour integration method are adopted to estimate the SOC:
taking 1 second as an integral time interval, dividing into a front time slice and a rear time slice, carrying out acquisition operation for the front 500ms, and starting algorithm calculation in the rear 500ms, so that in the calculation time interval 1s, firstly, obtaining the SOC obtained by an open circuit voltage method ocv And one second of last round of last obtained SOC k-1 The average of the results is taken to determine an initial value:
SOC k ＝(SOC k-1 +SOC ocv )/2 (21)
and integrating the current of the next 500ms by utilizing an ampere-hour integration method to obtain a change value delta SOC:
wherein t represents the time slice of 1s, Q FC Representing the rated capacity of the battery;
subtracting the change value from the initial value to obtain:
SOC k+1 ＝SOC k -ΔSOC ocv (23)
the error sources are SOC initial value estimation and battery capacity change;
s31, processing errors by combining a digital twin model of the lithium battery pack;
the step S31 specifically includes the following steps:
s310, initial value error processing
The current changes along with the time change in the working state, the last calculation is read in the common process in the working state to be the initial value of this time, and the current is set to be 0 when the working state reaches the emptying condition; when the working state reaches the full condition, setting the working state to be 100%; in the idle state, correcting an initial value according to an open circuit voltage method combined with an equivalent circuit model;
s311, capacity error handling
Obtaining an OCV-SOC table obtained by utilizing digital twin model identification in an idle state through error analysis, then carrying out table lookup to obtain an SOC as an initial value, obtaining a result by utilizing an ampere-hour integrating method in a working state as the initial value of the next calculation, and recalculating the actual capacity of the battery by utilizing a capacity learning mode to correct the capacity error;
s32, estimating the SOC based on the battery equivalent circuit model and an ampere-hour integration method.
2. The digital twinning-based lithium battery side end collaborative management method according to claim 1, wherein the method comprises the following steps: in step S21, the equivalent circuit model equation is as follows:
U t ＝U oc -U 1 -IR 0 (2)
i is input current, U t Is a terminal voltage, U oc Represents an open circuit voltage, R 0 Represents ohmic internal resistance of battery, R 1 For the polarization internal resistance of the battery, C 1 U is the polarization capacitance of the battery 1 Representing the voltage drop across the RC parallel link,represents U 1 The derivative of time t.
3. The digital twinning-based lithium battery side end collaborative management method according to claim 2, characterized in that: the step S22 specifically includes the following steps:
s220, recursive least square algorithm containing forgetting factors
For the following system:
wherein y is k As an output variable at the kth acquisition instant,for the data variable, θ, at the kth acquisition time k For the parameter variable at the kth acquisition time, e k Is system noise;
the following formula was used for calculation:
wherein K is k Algorithm gain vector for kth acquisition time, P k-1 The covariance matrix of the identification parameters at the k-1 acquisition time is shown, mu is a forgetting factor, and I is an identity matrix;
s221, selecting a forgetting factor by using a simulation experiment.
4. The digital twinning-based lithium battery side end collaborative management method according to claim 3, wherein the method comprises the following steps: the step S23 specifically includes the following steps:
s230, obtaining the expression of the expression 1 and the expression 2 in the complex frequency domain through Laplace transformation:
defining the output of the system as U t (s)-U oc (s) the input of the system is I(s), then the transfer equation of the system is:
and then utilizing a bilinear transformation formula:
substituting equation 9 into equation 8, and converting the transfer function of the system into z domain, and simplifying to obtain:
and (3) making:
equation 10 is transformed into:
and then obtaining a discretized result through inverse transformation of z:
U t (k)＝U oc (k)-b 3 U oc (k-1)+b 3 U t (k-1)+b 1 I(k)+b 2 I(k-1) (15)
because the sampling time interval is short, U oc (k) And U oc (k-1) are approximately equal, then equation 15 is written as:
U t (k)＝(1-b 3 )U oc (k)+b 3 U t (k-1)+b 1 I(k)+b 2 I(k-1) (16)
thereby obtaining a relation between the open-circuit voltage and the terminal voltage, wherein the terminal voltage is used as a parameter identification result.
5. The digital twinning-based lithium battery side end collaborative management method according to claim 4, wherein the method comprises the following steps: the step S24 specifically includes the following steps:
definition y k As an output variable at the kth acquisition instant,for the data variable, θ, at the kth acquisition time k For the parameter variable at the kth acquisition time, writing the system equation into a matrix form: />Finishing 16 yields the system equation:
y k ＝U t (k) (17)
θ k ＝[(1-b 3 )U oc (k) b 3 b 1 b 2 ] (19)
substituting the formulas 11, 12 and 13 into the obtained parameter R 0 And R is 1 。
6. The digital twinning-based lithium battery side end collaborative management method according to claim 1, wherein the method comprises the following steps: the initial value correction by the open circuit voltage method in step S310 includes the following steps:
when the battery state monitoring unit of the physical entity collects working current, terminal voltage and the last calculated SOC result, the working current, the terminal voltage and the last calculated SOC result are uploaded to the mobile phone terminal virtual entity, the virtual entity calculates to obtain a new OCV-SOC table according to an equivalent circuit model and a recursive least square method with forgetting factors, the new OCV-SOC table is fed back to the main control unit of the physical entity, and the main control unit utilizes the new OCV-SOC table and combines an initial value calculated by an open circuit voltage method to complete correction.
7. The digital twinning-based lithium battery side end collaborative management method according to claim 6, wherein the method comprises the following steps: step S311 specifically includes the following steps:
s3110, judging whether the discharging state reaches a discharging zone bit, namely whether the voltage of the single battery is smaller than a single string undervoltage threshold, if so, executing a step S3111, otherwise, executing a step S3113;
s3111, triggering a capacity learning sign, and when the capacity learning sign is valid, reaching a full charge state, completing one-time capacity learning:
wherein I is current, t is time from start of charging to full charge state, ΔQ FC Is the learned capacity;
if the state of charge is interrupted during the capacity learning process, the current battery state CurState enters a discharging or idle state, and if the capacity learning flag fails, step S3113 is executed;
s3112, update capacity:
Q FC ＝ΔQ FC (25)
s1113, finishing capacity learning.
8. The digital twinning-based lithium battery side end collaborative management method according to claim 7, wherein the method comprises the following steps: the step S32 specifically includes the following steps:
s320, initial value setting:
n＝0 (27)
wherein SOC is ocv The representative system is obtained by open circuit voltage table lookup during initialization, and SOC k Representing the last calculated SOC result of the last second when the system is operating normally, n representing the secondA number identifier for determining the number of times of calculation;
s321, state value processing:
wherein t represents the 1000ms time slice, I represents the current at the 1s, Q FC Representing the rated capacity of the battery;
s322, calculating the state of charge SOC of the battery k+1 ：
SOC k+1 ＝SOC 0 -ΔSOC (29)
n＝n+1 (30)
Wherein SOC is 0 Representing the calculated initial value at this time, wherein delta SOC represents the calculated result value in the step S321, and n is increased by 1;
s323, judging whether n is more than or equal to 2,
if n is not greater than 2, entering initial value processing:
judging whether the battery is in an idle state or not, and judging whether the current is in a current dead zone or not according to the judgment standard;
if the state is the idle state, firstly, looking up a table according to the existing OCV-SOC table, then judging the difference value calculated in the step S322, if the difference value is more than 30%, representing that the existing OCV-SOC table is invalid, then, carrying out parameter estimation of open-circuit voltage by utilizing a digital twin model, feeding back and dynamically updating the OCV-SOC table of hardware by using the model in the mobile phone terminal, and then, carrying out an open-circuit voltage method by using the new table to obtain an initial value as the next calculation;
if not, reading the initial value calculated in the step S322 as the next calculation, and entering the step S321 to continue calculation after obtaining the initial value;
if n is greater than or equal to 2, go to S324;
s324, the calculation result is output to the outside for display, and n is reset to 0:
SOC＝SOC k (31)；
s325, judging whether the voltage in the discharge state is smaller than an undervoltage threshold value, if yes, entering a step S326, and if not, entering a step S327;
s326, triggering a capacity learning mark in a emptying state, entering a capacity learning process, judging whether the current capacity learning mark is 1 and whether the current battery is in a charging state, if the current capacity learning mark is simultaneously satisfied with starting the capacity learning, the battery must be always in the charging state in the learning process, and resetting the capacity learning mark to 0 when the battery enters other states, wherein the charge amount obtained from emptying to full charge is the new capacity, and updating the battery charge state SOC of S321 k+1 ；
S327, finishing the formation of the fusion algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210601270.2A CN114935722B (en) | 2022-05-30 | 2022-05-30 | Lithium battery side end collaborative management method based on digital twin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210601270.2A CN114935722B (en) | 2022-05-30 | 2022-05-30 | Lithium battery side end collaborative management method based on digital twin |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114935722A CN114935722A (en) | 2022-08-23 |
| CN114935722B true CN114935722B (en) | 2024-01-12 |
Family
ID=82867584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210601270.2A Active CN114935722B (en) | 2022-05-30 | 2022-05-30 | Lithium battery side end collaborative management method based on digital twin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114935722B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110082683A (en) * | 2019-05-09 | 2019-08-02 | 合肥工业大学 | Inhibit the closed loop compensation method of ampere-hour integral SOC evaluated error |
| CN110492186A (en) * | 2019-07-02 | 2019-11-22 | 北京航空航天大学 | A kind of management method of the power battery module based on cloud control technology |
| CN110534823A (en) * | 2019-07-19 | 2019-12-03 | 北京航空航天大学 | A kind of power battery balanced management system and method |
| CN112711892A (en) * | 2020-12-07 | 2021-04-27 | 深圳先进技术研究院 | Cloud battery management system and method based on digital twin and block chain technology |
| CN113495215A (en) * | 2021-06-28 | 2021-10-12 | 清华大学深圳国际研究生院 | Virtual battery cell system, operation method thereof and twin battery |
| CN113671382A (en) * | 2021-09-06 | 2021-11-19 | 中国科学院电工研究所 | Battery energy storage system state estimation method based on cloud-end digital twinning |
| CN113960476A (en) * | 2021-09-23 | 2022-01-21 | 北京理工大学 | Power battery fault monitoring method and system based on information physical fusion technology |
| CN114329760A (en) * | 2021-11-26 | 2022-04-12 | 北京交通大学 | Vehicle-mounted lithium ion battery modeling and fault diagnosis method based on digital twinning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9869725B2 (en) * | 2012-05-16 | 2018-01-16 | Robert Bosch Gmbh | Battery system and method with capacity estimator |
-
2022
- 2022-05-30 CN CN202210601270.2A patent/CN114935722B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110082683A (en) * | 2019-05-09 | 2019-08-02 | 合肥工业大学 | Inhibit the closed loop compensation method of ampere-hour integral SOC evaluated error |
| CN110492186A (en) * | 2019-07-02 | 2019-11-22 | 北京航空航天大学 | A kind of management method of the power battery module based on cloud control technology |
| CN110534823A (en) * | 2019-07-19 | 2019-12-03 | 北京航空航天大学 | A kind of power battery balanced management system and method |
| CN112711892A (en) * | 2020-12-07 | 2021-04-27 | 深圳先进技术研究院 | Cloud battery management system and method based on digital twin and block chain technology |
| CN113495215A (en) * | 2021-06-28 | 2021-10-12 | 清华大学深圳国际研究生院 | Virtual battery cell system, operation method thereof and twin battery |
| CN113671382A (en) * | 2021-09-06 | 2021-11-19 | 中国科学院电工研究所 | Battery energy storage system state estimation method based on cloud-end digital twinning |
| CN113960476A (en) * | 2021-09-23 | 2022-01-21 | 北京理工大学 | Power battery fault monitoring method and system based on information physical fusion technology |
| CN114329760A (en) * | 2021-11-26 | 2022-04-12 | 北京交通大学 | Vehicle-mounted lithium ion battery modeling and fault diagnosis method based on digital twinning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114935722A (en) | 2022-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107422269B (en) | Lithium battery online SOC measurement method | |
| CN110261779B (en) | Online collaborative estimation method for state of charge and state of health of ternary lithium battery | |
| CN107368619B (en) | Extended Kalman filtering SOC estimation method | |
| CN107402353B (en) | Method and system for carrying out filtering estimation on state of charge of lithium ion battery | |
| CN108732503B (en) | Method and device for detecting battery health state and battery capacity | |
| CN108369258B (en) | State estimation device and state estimation method | |
| CN109507611B (en) | SOH correction method and system for electric vehicle | |
| US20090048793A1 (en) | State and parameter estimator having integral and differential components for electrical energy accumulators | |
| WO2015106691A1 (en) | Soc estimation method for power battery for hybrid electric vehicle | |
| CN108445422B (en) | Battery state of charge estimation method based on polarization voltage recovery characteristics | |
| CN108693472A (en) | Battery equivalent model on-line parameter identification method | |
| CN112415411A (en) | Method and apparatus for estimating SOC of battery, vehicle, and storage medium | |
| CN112858916B (en) | Battery pack state of charge estimation method based on model and data driving fusion | |
| CN113009346B (en) | Battery system and SOC value correction method thereof | |
| CN110658466A (en) | Method for calculating residual capacity of storage battery | |
| CN110716141B (en) | Battery state of charge estimation method based on extended Kalman filtering | |
| CN111044924A (en) | Method and system for determining residual capacity of all-condition battery | |
| CN110988701A (en) | Battery available energy determination method, device, management system and storage medium | |
| CN112213644B (en) | Battery state of charge estimation method and battery management system | |
| CN114935722B (en) | Lithium battery side end collaborative management method based on digital twin | |
| CN109298340B (en) | Battery capacity online estimation method based on variable time scale | |
| CN116224073A (en) | Battery SOC estimation method, device, equipment, battery module and storage medium | |
| CN114114048A (en) | Battery SOC, SOH and SOE joint estimation method | |
| CN116134327A (en) | Method and device for estimating SOC of battery pack and battery management system | |
| CN114636936B (en) | Correction method and device for SOC prediction curve in charging stage of lead-acid battery |
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 |