Multi-parameter online updating liquid metal battery state-of-charge estimation method and system

Abstract

The invention discloses a multi-parameter online updated liquid metal battery state-of-charge estimation method and system, and relates to the technical field of secondary battery application, and the method comprises the steps: constructing a liquid metal battery equivalent model, and deducing a battery state space mathematical model; acquiring a system state mathematical model and a mathematical model; and performance verification is carried out through battery multi-parameter online joint estimation. According to the method for estimating the state of charge of the liquid metal battery based on multi-parameter on-line updating provided by the invention, under a short time scale, terminal voltage is used as observed quantity to obtain model parameters and estimated values of the SOC of the battery; under the long-time scale, the SOC variation of the battery in the long-time scale is used as the observed quantity to obtain the estimated value of the battery capacity, so that the complexity of the calculation process can be reduced, a more accurate estimation result can be obtained, and the change of internal parameters of the battery can be adaptively tracked, thereby improving the real-time estimation precision and robustness of the SOC of the battery in the service cycle.

Description

Technical Field

[0001] The present invention relates to the technical field of secondary battery applications, and in particular to a method and system for estimating the state of charge of a liquid metal battery with multi-parameter online updating. Background Art

[0002] Energy storage technology helps the large-scale development and application of renewable clean energy, which is one of the necessary means to achieve my country's dual-carbon strategic goals. Among the many energy storage technologies, electrochemical energy storage technology has the advantages of short construction period, small terrain factors, strong regulation ability, etc. It can be flexibly applied to the power generation, transmission, distribution and power consumption of the power system, and has gradually become one of the mainstream energy storage technologies in the future. As a new type of electrochemical energy storage technology, liquid metal batteries have the advantages of long cycle life, low energy storage cost and easy capacity amplification due to the three-layer liquid structure design. It has broad application prospects in the field of large-scale energy storage applications for power systems. Battery systems are usually integrated in series and parallel with hundreds or even tens of thousands of single cells, which require real-time management by the battery management system to ensure the safe operation of the battery system and extend its service life, thereby improving the economic value of the battery system. The basis and core of the battery management system is the high-precision estimation of the battery state of charge (SOC). As an important state parameter reflecting the remaining power of the battery, SOC is an important reference indicator for the battery management system to issue control instructions.

[0003] Existing battery SOC estimation methods can be divided into three types: measurement signal-based methods, model-based methods, and data-driven methods. Common measurement methods include the ampere-hour integration method and the open-circuit voltage method. As an open-loop calculation method, the ampere-hour integration method is easily limited by the accuracy of the current sensor; the open-circuit voltage method requires a long period of static state to obtain a stable battery OCV value. However, under actual working conditions, it is impossible to ensure that the battery is static for a long enough time, so it is difficult to apply online. The model-based method can avoid the influence of sensor errors, is insensitive to initial values, has the characteristics of closed-loop self-correction, and can obtain more accurate SOC estimation results online. However, the estimation effect depends on the high-precision development of the model, and the model parameters are easily affected by factors such as temperature, aging, and working conditions. The data-driven method regards the battery as a black box. There is no need to conduct in-depth analysis of the mechanism of the electrochemical reaction process inside the battery. It only uses the collected battery test data to mine the nonlinear relationship between data features and SOC, but usually requires a large amount of battery data for long-term training to obtain a strong mapping relationship. The accuracy of SOC estimation is highly dependent on the quality of the training data set. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the existing battery SOC estimation method has the problems of low battery model accuracy, differences between single cells, battery aging, and how the accuracy of liquid metal battery charge state estimation is reduced.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a method for estimating the state of charge of a liquid metal battery with online updating of multiple parameters, comprising constructing an equivalent model of a liquid metal battery, deriving a mathematical model of the battery state space; obtaining a system state mathematical model and a mathematical model; and jointly estimating multiple battery parameters online and performing performance verification.

[0007] As a preferred solution of the multi-parameter online updated liquid metal battery state of charge estimation method of the present invention, wherein: the construction of the liquid metal battery equivalent model and the derivation of the battery state space mathematical model include modeling the liquid metal battery monomer using the Thevenin equivalent circuit model;

[0008] Thevenin model equation is expressed as:

[0009]

[0010] V t =V oc -V p -iR o

[0011] Wherein, Voc represents the open circuit voltage of the battery, Vp represents the polarization voltage of the battery, Vt represents the terminal voltage of the battery, Ro represents the ohmic internal resistance, Rp represents the polarization internal resistance, Cp represents the polarization capacitance, and i represents the battery current;

[0012] The Thevenin model equation is discretized to obtain the discrete equation of the model as follows:

[0013] V p (k+1)=exp(-Δt / R p C p )V p (k)+R p [1-exp(-Δt / R p C p )]I(k)

[0014] =aV p (k)+bI(k)

[0015] V t (k) = V oc (k)-V p (k)-R o I(k)

[0016] Among them, Δt represents the sampling period, and k represents the sampling step;

[0017] The nonlinear mapping relationship between the open circuit voltage Voc and SOC of the liquid metal battery is expressed as:

[0018]

[0019] Wherein, Ki represents the polynomial coefficient, which is output by fitting the battery OCV-SOC calibration curve.

[0020] As a preferred solution of the multi-parameter online updated liquid metal battery state of charge estimation method of the present invention, wherein: the acquisition of the system state mathematical model and the mathematical model include taking into account that the model parameters, battery capacity and battery SOC have different time-varying characteristics, and adopting a time scale separation method to respectively obtain the true values ​​of the variables to be estimated;

[0021] When the battery is operating under actual working conditions, the model parameters and SOC will change at any time, and the short time scale Δt is used for estimation;

[0022] The change in battery capacity is estimated using a long time scale ΔT;

[0023] The long time scale ΔT is defined as L times the short time scale Δt, ΔT = LΔt, where L is a positive integer;

[0024] According to the model discretization equation, the state mathematical model at a short time scale is expressed as:

[0025]

[0026] Among them, η represents the battery coulomb efficiency, Cn represents the battery capacity, ε k represents the battery parameters, ε k =[a(k)b(k)R o (k)] T ;

[0027] The mathematical model of observation on a short time scale is expressed as:

[0028] V t (k) = V oc [SOC(k)]-V p (k)-R o I(k)

[0029] The state mathematical model and observation mathematical model on a long time scale are:

[0030] C n (l+1)=C n (l)+ω(k)

[0031]

[0032] As a preferred solution of the multi-parameter online updated liquid metal battery state of charge estimation method of the present invention, wherein: the acquisition of the system state mathematical model and the mathematical model include using a change ΔSOC of SOC within ΔT to replace the battery terminal voltage as an observed quantity;

[0033] Therefore, the mathematical model of the system state space in short time scale is expressed as:

[0034]

[0035] Among them, xk represents the state vector at a short time scale, x k =[SOC(k)V p (k)a(k)b(k)R o (k)] T ; Control vector u k =I(k); θl represents the state vector at a long time scale, θ l =C n (l); represents the observation vector at a short time scale, w and v represent process noise and observation noise respectively; f x (x k-1 ,u k-1 ,θ l ) and g x (x k ,u k ,θ l ) are the nonlinear state transfer and system output functions that take xk and uk as independent variables respectively;

[0036] The mathematical model of the system state space on a long time scale is expressed as:

[0037]

[0038] in, represents the observation vector at a long time scale, g θ (θ l ,u k ,x k ) are the system output functions that take θl and uk as independent variables respectively.

[0039] As a preferred solution of the method for estimating the state of charge of a liquid metal battery with multi-parameter online update described in the present invention, wherein: the online joint estimation of multiple parameters of the battery and performance verification include using a multi-time scale H∞ filtering algorithm to perform online joint estimation of multiple parameters of the liquid metal battery, including;

[0040] Parameter initialization, setting state estimation And the initial value of the error covariance matrix P0 is expressed as:

[0041]

[0042] Time update—short time scale, output function f x (x k-1 ,u k-1 ,θ l )The first-order partial derivative of xk at the current moment is expressed as:

[0043]

[0044] The estimated prior state vector is expressed as:

[0045]

[0046] The updated prior covariance matrix is ​​expressed as:

[0047]

[0048] Measurement update, output function g x (x k ,u k ,θ l )The first-order partial derivative of xk at the current moment is expressed as:

[0049]

[0050] The output gain matrix is ​​expressed as:

[0051]

[0052] The updated posterior state vector estimate is expressed as:

[0053]

[0054] The updated posterior covariance matrix is ​​expressed as:

[0055]

[0056] If k%L=0, then l=l+1, otherwise jump to k=k+1 and return to time update—short time scale until the sampling time expires.

[0057] As a preferred solution of the liquid metal battery state of charge estimation method with multi-parameter online update described in the present invention, wherein: the online joint estimation of battery multi-parameters and performance verification includes time update - the long time scale includes:

[0058] The estimated prior state vector is expressed as:

[0059]

[0060] The updated prior covariance matrix is ​​expressed as:

[0061]

[0062] Measurement update, output function g θ (θ l ,u k ,x k ) at the current moment l The first-order partial derivative of is expressed as:

[0063]

[0064] The output gain matrix is ​​expressed as:

[0065]

[0066] The updated posterior state vector estimate is expressed as:

[0067]

[0068] The updated posterior covariance matrix is ​​expressed as:

[0069]

[0070] k=k+1, return time update—short time scale until the sampling time ends.

[0071] As a preferred solution of the multi-parameter online updated liquid metal battery state of charge estimation method of the present invention, wherein: the online joint estimation of multiple parameters of the battery and performance verification include inputting test data into the online joint estimation of multiple parameters of the battery, outputting battery model parameters, capacity and SOC estimation values, and verifying the accuracy and reliability of the proposed online joint estimation method of battery model parameters, capacity and SOC based on multiple time scales by comparing with the measured data;

[0072] The root mean square error (RMSE) and mean absolute error (MAE) are used to evaluate the estimation performance, which are expressed as:

[0073]

[0074] Where N represents the length of the test data, xk represents the actual measured value of the state quantity, Represents the estimated value of a state quantity.

[0075] Another object of the present invention is to provide a liquid metal battery state of charge estimation system with multi-parameter online update, which can solve the problem of low battery model accuracy in current battery SOC estimation methods through online joint estimation of multiple battery parameters.

[0076] As a preferred solution of the multi-parameter online updated liquid metal battery state of charge estimation system described in the present invention, it includes: a battery state space model derivation module, a system state model and observation model acquisition module, and a battery multi-parameter online joint estimation module; the battery state space model derivation module is used to construct a liquid metal battery equivalent model and derive the battery state space model; the system state model and observation model acquisition module is used to obtain the system state equation and the observation equation; the battery multi-parameter online joint estimation module is used to perform multi-parameter online joint estimation.

[0077] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a method for estimating the state of charge of a liquid metal battery with online updating of multiple parameters.

[0078] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for estimating the state of charge of a liquid metal battery with online updating of multiple parameters.

[0079] Beneficial effects of the present invention: The multi-parameter online updated liquid metal battery state of charge estimation method provided by the present invention uses the terminal voltage as the observed quantity to obtain the estimated value of the model parameters and the battery SOC in a short time scale; in a long time scale, the battery SOC change in a long time scale is used as the observed quantity to obtain the estimated value of the battery capacity. Compared with the existing test method, this test method can not only reduce the complexity of the calculation process, but also obtain more accurate estimation results. The multi-parameter online joint estimation method of liquid metal batteries based on the H∞ filtering algorithm is provided to adaptively track the changes in the internal parameters of the battery, thereby improving the real-time estimation accuracy and robustness of the battery SOC during the service cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0081] Figure 1A flow chart of a method for online joint estimation of liquid metal battery model parameters, capacity and SOC of a liquid metal battery state of charge estimation method with multi-parameter online updating provided in the first embodiment of the present invention.

[0082] Figure 2 A cycle performance diagram of a liquid metal battery under aging conditions according to a multi-parameter online updated liquid metal battery state of charge estimation method provided in the second embodiment of the present invention.

[0083] Figure 3 A schematic diagram of the SOC prediction results of the 22nd lap under aging conditions of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation method with multi-parameter online update provided in the second embodiment of the present invention.

[0084] Figure 4 A schematic diagram of the SOC prediction results of the 320th lap under aging conditions of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation method with multi-parameter online update provided in the second embodiment of the present invention.

[0085] Figure 5 A schematic diagram of the voltage prediction results of the 320th cycle under aging conditions of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation method with multi-parameter online update provided in the second embodiment of the present invention.

[0086] Figure 6 A liquid metal battery dynamic operating condition-test current curve diagram of a liquid metal battery state of charge estimation method with multi-parameter online update provided in the second embodiment of the present invention.

[0087] Figure 7 A schematic diagram of SOC prediction results under dynamic conditions of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation system with multi-parameter online update provided in a second embodiment of the present invention.

[0088] Figure 8 A schematic diagram of battery capacity prediction results under dynamic conditions of a multi-time scale joint estimation method for a liquid metal battery state of charge estimation system with multi-parameter online update provided in a second embodiment of the present invention.

[0089] Fig. 9 A schematic diagram of battery voltage prediction results under dynamic conditions of a multi-time scale joint estimation method for a liquid metal battery state of charge estimation system with multi-parameter online update provided in a second embodiment of the present invention.

[0090] Fig.10A second test current curve diagram of a liquid metal battery dynamic operating condition 2 of a liquid metal battery state of charge estimation system with multiple parameters updated online is provided in accordance with the second embodiment of the present invention.

[0091] Fig.11 A schematic diagram of SOC prediction results of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation system with multi-parameter online update under dynamic working condition 2 provided in a second embodiment of the present invention.

[0092] Fig.12 A schematic diagram of battery capacity prediction results under dynamic working condition 2 of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation system with multi-parameter online update provided in a second embodiment of the present invention.

[0093] Fig.13 A schematic diagram of battery voltage prediction results under dynamic working condition 2 of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation system with multi-parameter online update provided in a second embodiment of the present invention.

[0094] Fig.14 A schematic diagram comparing the SOC estimation results of a multi-time scale joint estimation method of a liquid metal battery state of charge estimation system with a multi-parameter online update under dynamic working conditions with other algorithms provided in the second embodiment of the present invention.

[0095] Fig.15 A schematic diagram showing a comparison of the battery internal resistance estimation results of other algorithms under dynamic conditions for a multi-time scale joint estimation method of a liquid metal battery state of charge estimation system with a multi-parameter online update provided in a second embodiment of the present invention.

[0096] Fig.16 An overall flow chart of a liquid metal battery state of charge estimation system with multi-parameter online update provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0097] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0098] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a method for estimating the state of charge of a liquid metal battery with multi-parameter online updating, comprising:

[0099] S1: Construct an equivalent model of liquid metal battery and derive the battery state space equation.

[0100] Furthermore, preferably, the Thevenin equivalent circuit model is used to model the liquid metal battery monomer. The Thevenin model equation is expressed as follows:

[0101]

[0102] V t =V oc -V p -iR o

[0103] Among them, Voc is the battery open circuit voltage, Vp is the battery polarization voltage, Vt is the battery terminal voltage, Ro is the ohmic internal resistance, Rp is the polarization internal resistance, Cp is the polarization capacitance, and i is the battery current.

[0104] Discretize equation (1) to obtain the discrete equation of the model:

[0105] V p (k+1)=exp(-Δt / R p C p )V p (k)+R p [1-exp(-Δt / R p C p )]I(k)

[0106] =aV p (k)+bI(k)

[0107] V t (k) = V oc (k)-V p (k)-R o I(k)

[0108] Among them, Δt represents the sampling period, and k is the sampling step. The nonlinear mapping relationship between the open circuit voltage Voc and SOC of the liquid metal battery is expressed as:

[0109]

[0110] Among them, Ki is the polynomial coefficient, which can be obtained by fitting the battery OCV-SOC calibration curve.

[0111] S2: Obtain the system state equation and observation equation.

[0112] Preferably, considering that the model parameters, battery capacity and battery SOC have different time-varying characteristics, the time scale separation method is used to obtain the true values ​​of the variables to be estimated. When the battery is operating under actual complex conditions, the model parameters and SOC will change at any time and have fast time-varying properties, so a short time scale Δt is used for estimation; the battery capacity changes relatively slowly and is a slow time-varying parameter, so a long time scale ΔT is used for estimation. The long time scale ΔT is defined as L times the short time scale Δt, that is, ΔT = LΔt, where L is a positive integer.

[0113] According to the model discretization equation in step S1, the state equation at a short time scale is:

[0114]

[0115] Among them, η represents the battery coulomb efficiency, Cn represents the battery capacity, ε k represents the battery parameters, ε k =[a(k)b(k)R o (k)] T .

[0116] The observation equation on a short time scale is:

[0117] V t (k) = V oc [SOC(k)]-V p (k)-R o I(k)

[0118] The state equation and observation equation on a long time scale are:

[0119] C n (l+1)=C n (l)+ω(k)

[0120]

[0121] Preferably, the change ΔSOC of SOC within ΔT is used instead of the battery terminal voltage as the observed quantity to improve the accuracy of battery capacity estimation.

[0122] Therefore, the state space equation of the system on a short time scale can be simplified as:

[0123]

[0124] Among them, xk represents the state vector at a short time scale, x k =[SOC(k)V p (k)a(k)b(k)R o (k)] T ; Control vector u k=I(k); θl represents the state vector at a long time scale, θ l =C n (l); represents the observation vector at a short time scale, w and v represent process noise and observation noise respectively. x (x k-1 ,u k-1 ,θ l ) and g x (x k ,u k ,θ l ) are the nonlinear state transfer and system output functions with xk and uk as independent variables respectively.

[0125] The state space equation of the system on a long time scale can be abbreviated as:

[0126]

[0127] in, represents the observation vector at a long time scale, g θ (θ l ,u k ,x k ) are the system output functions that take θl and uk as independent variables respectively.

[0128] S3: Online joint estimation of multiple battery parameters.

[0129] Preferably, the present invention adopts a multi-time scale H∞ filter algorithm (Mts-HIF) to perform online joint estimation of multiple parameters of liquid metal batteries. The specific method is implemented by the following steps:

[0130] (1) Parameter initialization

[0131] Setting state estimate And the initial value of its error covariance matrix P0:

[0132]

[0133] (2) Time update—short time scale

[0134] 1) Calculate the function f x (x k-1 ,u k-1 ,θ l )The first-order partial derivative (Jacobian matrix) of xk at the current moment:

[0135]

[0136] 2) Estimate the prior state vector

[0137]

[0138] 3) Update the prior covariance matrix

[0139]

[0140] (3) Measurement update

[0141] 1) Calculate the function g x (x k ,u k ,θ l )The first-order partial derivative of xk at the current moment:

[0142]

[0143] 2) Calculate the gain matrix

[0144]

[0145] 3) Update the posterior state vector estimate

[0146]

[0147] 4) Update the posterior covariance matrix

[0148]

[0149] (4) If k%L=0, then l=l+1, otherwise jump to process (7)

[0150] (5) Time update—long time scale

[0151] 1) Estimate the prior state vector

[0152]

[0153] 2) Update the prior covariance matrix

[0154]

[0155] (6) Measurement update

[0156] 1) Calculate the function g θ (θ l ,u k ,x k ) at the current moment l The first partial derivative of :

[0157]

[0158] 2) Calculate the gain matrix

[0159]

[0160] 3) Update the posterior state vector estimate

[0161]

[0162] 4) Update the posterior covariance matrix

[0163]

[0164] (7) k=k+1, return to process (2) until the sampling time expires.

[0165] S4: Method performance verification.

[0166] The test data is input into step S3 to obtain the battery model parameters, capacity and SOC estimation values. The accuracy and reliability of the proposed online joint estimation method of battery model parameters, capacity and SOC based on multiple time scales are verified by comparing with the measured data.

[0167] Preferably, the present invention uses root mean square error (RMSE) and mean absolute error (MAE) to evaluate the estimation performance, and the calculation formula is as follows:

[0168]

[0169] Where N represents the length of the test data, xk represents the actual measured value of the state quantity, Represents the estimated value of a state quantity.

[0170] The liquid metal cell tested is a 200Ah-class Li||Bi liquid metal battery.

[0171] Figure 1 The flow chart of the online joint estimation method of liquid metal battery model parameters, SOC and capacity based on the multi-time scale H∞ filtering algorithm provided by the present invention. The main steps include:

[0172] Perform a working condition test on the liquid metal battery and read the battery input current and battery output voltage at time k.

[0173] The battery SOC and model parameters are updated based on measurements on a short time scale.

[0174] The battery SOC and model parameters are updated on a short time scale.

[0175] When the input time k%L=0, the battery capacity is updated by measurement and time, otherwise the battery capacity is not updated.

[0176] When the test time is over, the battery parameter estimation process is stopped.

[0177] Example 2, reference Figure 2-Figure 15 , which is an embodiment of the present invention, provides a liquid metal battery state of charge estimation method with multi-parameter online update. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0178] first, Figure 2 This is a cycle performance diagram of the liquid metal battery provided by the present invention under aging conditions.

[0179] The capacity of the tested liquid metal battery gradually decayed as the battery cycled, and battery aging occurred.

[0180] Figure 3 Schematic diagram of the SOC estimation results of the aging battery under different cycle numbers using the multi-time scale joint estimation method of the present invention

[0181] When the battery is discharged for the 22nd cycle, the SOC of the battery is estimated by the estimation method provided by the present invention, and the average absolute error is 0.79%. Figure 4 As shown; when the battery is discharged for 320 cycles, the SOC of the battery is estimated again using the estimation method provided by the present invention, and the average absolute error is 0.81%. Figure 5 As shown in the figure, it can be seen from the estimation results that the battery model parameter-SOC-capacity online joint estimation method based on the H∞ filtering algorithm can accurately estimate the SOC of the battery at different aging degrees, and the estimation error is less than 1%.

[0182] Figure 7 This is a test current curve diagram of the dynamic working condition of the liquid metal battery provided by the present invention.

[0183] The batteries were tested using the Federal Urban Driving Schedule (FUDS).

[0184] Figure 8 — Fig. 9 A schematic diagram of battery parameter estimation results under dynamic conditions using the multi-time scale joint estimation method provided by the present invention.

[0185] Under FUDS working condition, the combined estimation method provided by the present invention is used to estimate the battery state. The results are as follows: Fig.10 As shown. Fig.10 It can be seen that the battery model parameter-SOC-capacity online joint estimation method based on the H∞ filtering algorithm can accurately track the changes in battery capacity and terminal voltage, and achieve high-precision estimation of battery SOC. The average absolute error of battery SOC estimation is 0.43%, and the root mean square error is 0.50%.

[0186] Fig.11 This is a test current curve diagram of the second dynamic working condition of the liquid metal battery provided by the present invention.

[0187] The Urban Dynamometer Driving Schedule (UDDS) is adopted.

[0188] Figure 12-13 This is a schematic diagram of the battery SOC, capacity, and terminal voltage prediction results under dynamic working condition 2 at multiple time scales provided by the present invention.

[0189] Under UDDS working conditions, the battery state is estimated using the combined estimation method provided by the present invention. The estimation results are as follows: Fig.10 As shown, the battery SOC estimation value quickly converges to the true value, the average absolute error of the battery SOC estimation is 0.65%, the root mean square error is 0.66%, the capacity estimation error is 0.95Ah, and the voltage prediction error is <5mV. It can be seen that the real-time update of model parameters and the online estimation of battery capacity are helpful to improve the estimation accuracy of liquid metal battery SOC. The Mts-HIF model parameter-capacity-SOC joint estimation algorithm provided by the present invention has the advantages of fast convergence speed, high estimation accuracy and strong robustness.

[0190] Fig.13 Schematic diagram of the comparison of battery state estimation results of the multi-time scale joint estimation method provided by the present invention and other algorithms under dynamic conditions

[0191] Preferably, in order to further illustrate the superiority of the joint estimation method provided by the present invention, the Mts-HIF joint estimation algorithm provided by the present invention is compared with the Adaptive Extended Kalman Filter (AEKF) algorithm with online parameter estimation and the Adaptive Unscented Kalman Filter (AUKF) algorithm. The battery SOC estimation result is as follows: Fig.13 As shown, the battery internal resistance estimation result is shown in 14, and the statistical analysis results of the three algorithms are shown in Table 1. Comparing the estimation results of the battery parameters by the three algorithms, it can be seen that the Mts-HIF joint estimation algorithm has a higher estimation accuracy for the battery SOC, and has the best tracking effect on the battery internal resistance with the smallest error. It can be seen that the method provided by the present invention has strong applicability and can accurately and reliably estimate the state of the liquid metal battery in actual complex working conditions.

[0192] Table 1 Statistical analysis of battery SOC estimation results of different algorithms under FUDS conditions

[0193] Mts-HIF AEKF AUKF MAE(%) 0.4305 0.8745 0.8117 RMSE(%) 0.4961 1.0813 1.1053

[0194] Example 3, reference Fig.16 , as an embodiment of the present invention, provides a liquid metal battery state of charge estimation system with multi-parameter online update, including a battery state space model derivation module, a system state model and observation model acquisition module, and a battery multi-parameter online joint estimation module.

[0195] The battery state space model derivation module is used to construct a liquid metal battery equivalent model and derive the battery state space model. The system state model and observation model acquisition module is used to obtain the system state equation and observation equation. The battery multi-parameter online joint estimation module is used to perform multi-parameter online joint estimation.

[0196] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0197] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0198] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0199] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0200] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for estimating the state of charge of a liquid metal battery with multi-parameter online updating, characterized in that: include: Construct a liquid metal battery equivalent model and derive a battery state space mathematical model; Obtaining mathematical models of system states and mathematical models; Online joint estimation of multiple battery parameters and performance verification.

2. The method for estimating the state of charge of a liquid metal battery with multi-parameter online updating according to claim 1, characterized in that: The construction of the liquid metal battery equivalent model and the derivation of the battery state space mathematical model include modeling the liquid metal battery monomer using the Thevenin equivalent circuit model; Thevenin model equation is expressed as: V t =V oc -V p -iR o Wherein, Voc represents the open circuit voltage of the battery, Vp represents the polarization voltage of the battery, Vt represents the terminal voltage of the battery, Ro represents the ohmic internal resistance, Rp represents the polarization internal resistance, Cp represents the polarization capacitance, and i represents the battery current; The Thevenin model equation is discretized to obtain the discrete equation of the model as follows: V t (k)=V oc (k)-V p (k)-R o I(k) Among them, Δt represents the sampling period, and k represents the sampling step; The nonlinear mapping relationship between the open circuit voltage Voc and SOC of the liquid metal battery is expressed as: Wherein, Ki represents the polynomial coefficient, which is output by fitting the battery OCV-SOC calibration curve.

3. The method for estimating the state of charge of a liquid metal battery with multi-parameter online updating as claimed in claim 2, characterized in that: The system state mathematical model and the mathematical model are obtained by taking into account that the model parameters, battery capacity and battery SOC have different time-varying characteristics, and adopting a time scale separation method to respectively obtain the true values ​​of the variables to be estimated; When the battery is operating under actual working conditions, the model parameters and SOC will change at any time, and the short time scale Δt is used for estimation; The change in battery capacity is estimated using a long time scale ΔT; The long time scale ΔT is defined as L times the short time scale Δt, ΔT = LΔt, where L is a positive integer; According to the model discretization equation, the state mathematical model at a short time scale is expressed as: Among them, η represents the battery coulomb efficiency, Cn represents the battery capacity, ε k represents the battery parameters, ε k =[a(k) b(k) R o (k)] T ; The mathematical model of observation on a short time scale is expressed as: V t (k)=V oc [SOC(k)]-V p (k)-R o I(k) The state mathematical model and observation mathematical model on a long time scale are: C n (l+1)=C n (l)+ω(k) 4. The method for estimating the state of charge of a liquid metal battery with multi-parameter online updating as claimed in claim 3, characterized in that: The system state mathematical model and the mathematical model include using a change ΔSOC of SOC within ΔT to replace the battery terminal voltage as an observed quantity; Therefore, the mathematical model of the system state space in short time scale is expressed as: Among them, xk represents the state vector at a short time scale, x k =[SOC(k) V p (k) a(k) b(k) R o (k)] T ; Control vector u k =I(k); θl represents the state vector at a long time scale, θ l =C n (l); represents the observation vector at a short time scale, w and v represent process noise and observation noise respectively; f x (x k-1 ,u k-1 ,θ l ) and g x (x k ,u k ,θ l ) are the nonlinear state transfer and system output functions that take xk and uk as independent variables respectively; The mathematical model of the system state space on a long time scale is expressed as: in, represents the observation vector at a long time scale, g θ (θ l ,u k ,x k ) are the system output functions that take θl and uk as independent variables respectively.

5. The method for estimating the state of charge of a liquid metal battery with multi-parameter online updating as claimed in claim 4, characterized in that: The method of online joint estimation of multiple parameters of the battery and performance verification includes adopting a multi-time scale H∞ filtering algorithm to perform online joint estimation of multiple parameters of the liquid metal battery, including: Parameter initialization, setting state estimation And the initial value of the error covariance matrix P0 is expressed as: Time update—short time scale, output function f x (x k-1 ,u k-1 ,θ l )The first-order partial derivative of xk at the current moment is expressed as: The estimated prior state vector is expressed as: The updated prior covariance matrix is ​​expressed as: Measurement update, output function g x (x k ,u k ,θ l )The first-order partial derivative of xk at the current moment is expressed as: The output gain matrix is ​​expressed as: The updated posterior state vector estimate is expressed as: The updated posterior covariance matrix is ​​expressed as: If k%L=0, then l=l+1, otherwise jump to k=k+1 and return to time update—short time scale until the sampling time expires.

6. The method for estimating the state of charge of a liquid metal battery with multi-parameter online updating as claimed in claim 5, characterized in that: The online joint estimation of multiple battery parameters and performance verification includes time updates - long time scales include: The estimated prior state vector is expressed as: The updated prior covariance matrix is ​​expressed as: Measurement update, output function g θ (θ l ,u k ,x k ) at the current moment l The first-order partial derivative of is expressed as: The output gain matrix is ​​expressed as: The updated posterior state vector estimate is expressed as: The updated posterior covariance matrix is ​​expressed as: k=k+1, return time update—short time scale until the sampling time ends.

7. The method for estimating the state of charge of a liquid metal battery with multi-parameter online updating according to claim 6, characterized in that: The method of performing performance verification by online joint estimation of multiple battery parameters includes inputting test data into the online joint estimation of multiple battery parameters, outputting battery model parameters, capacity and SOC estimation values, and verifying the accuracy and reliability of the proposed online joint estimation method of battery model parameters, capacity and SOC based on multiple time scales by comparing with the measured data; The root mean square error (RMSE) and mean absolute error (MAE) are used to evaluate the estimation performance, which are expressed as: Where N represents the length of the test data, xk represents the actual measured value of the state quantity, Represents the estimated value of a state quantity.

8. A system using the method for estimating the state of charge of a liquid metal battery with online multi-parameter updates as claimed in any one of claims 1 to 7, characterized in that: It includes battery state space model derivation module, system state model and observation model acquisition module, and battery multi-parameter online joint estimation module; The battery state space model derivation module is used to construct a liquid metal battery equivalent model and derive a battery state space model; The system state model and observation model acquisition module is used to acquire the system state equation and observation equation; The battery multi-parameter online joint estimation module is used for performing multi-parameter online joint estimation.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for estimating the state of charge of a liquid metal battery with multi-parameter online update as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for estimating the state of charge of a liquid metal battery with multi-parameter online update according to any one of claims 1 to 7 are implemented.

Images