Battery SOC determination method and battery SOC detection system

The battery parameters are updated through the least squares method and the online gradient method, and the current sensor data is corrected using the bias function, which solves the problem of low battery SOC estimation accuracy, achieving higher SOC estimation accuracy and battery health monitoring.

CN118914859APending Publication Date: 2024-11-08ZHANJIANG POWER SUPPLY BUREAU OF GUANGDONG POWER GRID CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411000268.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In the prior art, the accuracy of battery SOC estimation is low, resulting in large errors in SOC estimation.

Method used

The initial parameters are updated using the least squares method and the online gradient method, and the current value of the current sensor is corrected by the bias function to calculate the battery SOC value.

Benefits of technology

It improves the accuracy of SOC estimation, reduces estimation errors, and enhances the reliability of parameter estimation and the accuracy of battery health monitoring.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118914859A_ABST
    Figure CN118914859A_ABST
Patent Text Reader

Abstract

The invention provides a battery SOC determination method and a battery SOC detection system. The method comprises the steps that initial parameters of a battery are obtained, the initial parameters comprise a first initial parameter and a second initial parameter, the first initial parameter comprises at least one of an initial capacitance value, an initial voltage value, an initial resistance value and an initial battery capacity, and the second initial parameter comprises an initial current value input by a current sensor; an updating algorithm is adopted to update the first initial parameter to obtain a first parameter, and the updating algorithm comprises a least square method and / or an online gradient method; updating the second initial parameter by adopting a bias function to obtain a second parameter; and calculating the SOC value of the battery according to the first parameter and the second parameter. According to the scheme, the problem of low accuracy of SOC estimation in the prior art is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of battery safety management, and in particular, to a method, device, computer program product, and battery SOC detection system for determining the SOC of a battery. Background Art

[0002] Currently, SoC estimation techniques are roughly divided into direct methods and indirect methods. Coulomb counting is a computationally simple method for directly estimating SoC, but it is prone to errors due to its sensitivity to initial conditions and current sensor bias. Indirect methods for SoC estimation include Kalman filter (KF), extended Kalman filter (EKF), adaptive extended Kalman filter, nonlinear prediction filter, and Luenberger observer.

[0003] However, the current SOC estimation method has a large error compared with the actual situation, resulting in a low accuracy of SOC estimation. Summary of the invention

[0004] The main purpose of the present application is to provide a method, device, computer program product and battery SOC detection system for determining the SOC of a battery, so as to at least solve the problem of low accuracy of SOC estimation in the prior art.

[0005] To achieve the above-mentioned purpose, according to one aspect of the present application, a method for determining the SOC of a battery is provided, comprising: obtaining initial parameters of the battery, wherein the initial parameters include a first initial parameter and a second initial parameter, the first initial parameter includes at least one of an initial capacitance value, an initial voltage value, an initial resistance value and an initial battery capacity, and the second initial parameter includes an initial current value input by a current sensor; updating the first initial parameter using an update algorithm to obtain a first parameter, wherein the update algorithm includes a least squares method and / or an online gradient method; updating the second initial parameter using a bias function to obtain a second parameter; and calculating the SOC value of the battery based on the first parameter and the second parameter.

[0006] Optionally, obtaining the initial parameters of the battery includes: calculating the initial parameters according to a first calculation formula, wherein the first calculation formula is:

[0007]

[0008] V(t)=f(z(t))-V c (t)-I(t)R0,

[0009] Indicates V c (t), R1 represents the first initial resistance value, C1 represents the initial capacitance value, Vc I(t) represents the initial voltage value of the capacitor, I(t) represents the initial current value input by the current sensor, represents the first-order derivative of z(t), Q represents the initial battery capacity, V(t) represents the output voltage value, f(z(t)) represents the OCV function of the open circuit voltage, z(t) represents the SOC value, R0 represents the second initial resistance value, and t represents the time.

[0010] Optionally, calculating the SOC value of the battery according to the first parameter and the second parameter includes: calculating the SOC value of the battery according to a second formula, wherein the second formula includes:

[0011]

[0012] Δz(t) represents the change in the SOC value, represents an estimate of the SOC value, represents the battery capacity in the first parameter, represents the bias function, represents the current value of the second parameter.

[0013] Optionally, when the update algorithm includes a least squares method, updating the first initial parameter using the update algorithm to obtain the first parameter includes: calculating a least squares solution using a third formula, wherein the third formula is:

[0014] in,

[0015] in, represents a vector containing parameters to be estimated, M represents a matrix consisting of accumulated current and time, M is used to form an input matrix in a linear system, T represents transpose, F represents a function representing the difference between the estimated battery state and the initial state, plus a correction term calculated based on the accumulated current and capacity; the impedance transfer function is calculated using the fourth formula, wherein the fourth formula is:

[0016]

[0017] Wherein, G(s) represents the transfer function, which is a mathematical description of the battery voltage response to the change of the current input, s represents a complex variable in the frequency domain, and m represents a coefficient related to the state of charge of the battery; the fifth formula is used to filter the electrical signal, wherein the fifth formula is:

[0018] ω1(t)=Λω1(t)+bI(t)

[0019] ω2(t)=Λω2(t)+bV(t),

[0020] Where, ω1(t)=[i1(t),i2(t),i3(t)] T ,ω2(t)=[v1(t),v2(t),v3(t)] T ,

[0021]

[0022] λ0=1 / τ 3 ,λ1=3 / τ 2 , λ2=3 / τ, ω1 represents the filtered current signal, ω2 represents the filtered voltage signal, λ represents the coefficient of the filter, and τ represents the time constant of the filter;

[0023] The error signal is calculated using the sixth formula, wherein the sixth formula is: e(t)=v3(t)-ω(t) T θ, e(t) represents the error signal between the filtered voltage signal and the model output. θ is a vector that contains a combination of model parameters.

[0024] ω(t)=[i1(t),i2(t),i3(t),v2(t)] T ;

[0025] The estimated parameters are calculated using the seventh formula, wherein the seventh formula is: in, represents the corrected voltage signal, X(t) represents a matrix including current and voltage signals, β represents the estimated parameter, and β represents the estimated parameter; the first initial parameter is updated according to the estimated parameter to obtain the first parameter.

[0026] Optionally, updating the first initial parameter according to the estimated parameter to obtain the first parameter includes: calculating the first parameter according to an eighth formula, wherein the eighth formula is:

[0027]

[0028] represents the first resistance value, represents the second resistance value, Indicates the capacitance value, express The second element in the vector, express The third element in the vector, express The fourth element in the vector.

[0029] Optionally, when the update algorithm includes an online gradient method, updating the first initial parameter using the update algorithm to obtain the first parameter includes: calculating the error according to the ninth formula and the tenth formula, wherein the ninth formula is: The tenth formula is: in,

[0030]

[0031] ψ(t)=[i2(t)i3(t)] T ,

[0032] e v3 (t) represents the error between the third voltage measurement value and the estimated voltage value, v3(t) represents the third voltage measurement value, represents the estimated value of the third voltage, e z (t) represents the difference between the actual battery state of charge and the estimated value of the state, represents the estimated value of the parameter vector, which mainly includes parameters related to the battery resistance and capacitance, ψ(t) represents a vector containing measurements related to the battery dynamics, v2(t) represents another voltage measurement, m(z(t)) represents a function related to the battery SoC, which affects the battery open circuit voltage, and α1 represents the parameter vector An element in, related to the battery capacity or current sensor deviation, M(t) T represents the transpose of the matrix, which consists of the accumulated current and time of the battery, represents a vector containing estimated values ​​related to the battery model parameters.

[0033] Optionally, when the update algorithm includes an online gradient method, the first initial parameter is updated using the update algorithm to obtain the first parameter, and further comprising: calculating a gradient update law according to an eleventh formula, wherein the eleventh formula is:

[0034]

[0035] in, express The rate of change over time t, γ 12 represents the design parameter used to adjust the speed and stability of the update rule, represents the partial derivative, φ represents a parameter vector, which contains the parameters related to the dynamic characteristics of the battery, express The rate of change over time t, γ 34represents the second design parameter, which is used to adjust the speed and stability of the update rule, α represents a parameter vector, which includes parameters related to the state of charge of the battery and the current sensor deviation, and M(t) represents a regression vector, which includes the measurement value related to the accumulated current time t; the error is updated according to the twelfth formula, wherein the twelfth formula is:

[0036]

[0037] Where m represents the coefficient related to the battery model, represents the deviation between the true value and the estimated value of the α parameter, represents the deviation between the true value and the estimated value of the φ parameter, represents the second deviation between the true value of the φ parameter and the estimated value, represents the deviation between the true value and the estimated value of the α parameter; updating the first initial parameter according to the thirteenth formula to obtain the first parameter, wherein the thirteenth formula is:

[0038]

[0039] According to another aspect of the present application, a device for determining the SOC of a battery is provided, comprising: an acquisition unit, used to acquire initial parameters of the battery, wherein the initial parameters include a first initial parameter and a second initial parameter, the first initial parameter includes at least one of an initial capacitance value, an initial voltage value, an initial resistance value and an initial battery capacity, and the second initial parameter includes an initial current value input by a current sensor; a first update unit, used to update the first initial parameter using an update algorithm to obtain a first parameter, wherein the update algorithm includes a least squares method and / or an online gradient method; a second update unit, used to update the second initial parameter using a bias function to obtain a second parameter; and a calculation unit, used to calculate the SOC value of the battery based on the first parameter and the second parameter.

[0040] According to another aspect of the present application, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any one of the methods for determining the SOC of a battery are implemented.

[0041] According to another aspect of the present application, a battery SOC detection system is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for determining the SOC of the battery as described.

[0042] By applying the technical solution of the present application, the least squares method and / or the online gradient method are used to update the parameters. The least squares method can reduce the influence of data errors on fitting and improve the reliability of parameter estimation. The online gradient method can easily jump out of the local optimal solution and find the global optimal solution. Therefore, the first parameter obtained by the present solution has a high accuracy. By correcting and updating the current value of the current sensor through the bias function, the influence of the measurement result can be reduced and the fluctuation of the error influence can be reduced. Therefore, the second parameter obtained by the present solution has a high accuracy. Therefore, SOC estimation is performed according to the updated parameters, which can reduce the estimation error and improve the accuracy of SOC estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0044] Figure 1 A hardware structure block diagram of a mobile terminal for executing a method for determining a battery SOC provided in an embodiment of the present application is shown;

[0045] Figure 2 A schematic flow chart of a method for determining the SOC of a battery provided in accordance with an embodiment of the present application is shown;

[0046] Figure 3 A schematic diagram showing a Thevenin equivalent circuit;

[0047] Figure 4 A block diagram of the SOC-SOH estimator of the present scheme is shown;

[0048] Figure 5 A schematic flow chart showing another method for determining the SOC of a battery;

[0049] Figure 6 A structural block diagram of a device for determining the SOC of a battery provided according to an embodiment of the present application is shown.

[0050] The above drawings include the following reference numerals:

[0051] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0052] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0053] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0054] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0055] As introduced in the background technology, the accuracy of SOC estimation in the prior art is low. To solve the above problem, the embodiments of the present application provide a method, device, computer program product and battery SOC detection system for determining the SOC of a battery.

[0056] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0057] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for determining the SOC of a battery according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0058] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the display method of device information in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0059] In this embodiment, a method for determining the SOC of a battery running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0060] Figure 2 FIG. 1 is a flow chart of a method for determining the SOC of a battery according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0061] Step S201, obtaining initial parameters of the battery, wherein the initial parameters include a first initial parameter and a second initial parameter, the first initial parameter includes at least one of an initial capacitance value, an initial voltage value, an initial resistance value and an initial battery capacity, and the second initial parameter includes an initial current value input by a current sensor;

[0062] Specifically, in this solution, an initial battery model can be constructed. The battery model can be a Thevenin equivalent circuit model, and the initial parameters of the battery can be obtained through the battery model.

[0063] Step S202, using an update algorithm to update the first initial parameter to obtain a first parameter, wherein the update algorithm includes a least squares method and / or an online gradient method;

[0064] Specifically, the least squares method is a commonly used statistical method for estimating parameters in linear models. Its basic idea is to determine the optimal parameter estimates by minimizing the residual sum of squares between the observed data and the model predictions. In the linear regression model, the least squares method fits the data by finding the parameter estimates that minimize the residual sum of squares. Specifically, the least squares method calculates the sum of squares of the residuals (the difference between the observed values ​​and the model predictions) and finds the parameter value that minimizes this sum of squares. When the least squares method is used to update the parameters, the model parameters are actually constantly adjusted so that the residual sum of squares gradually decreases and eventually reaches an optimal parameter estimate. Doing so can make the model fit the data better and improve the accuracy of the prediction. In general, updating parameters using the least squares method means constantly adjusting the model parameters by minimizing the residual sum of squares to obtain the best parameter estimates, thereby better fitting the data and improving the accuracy of the prediction.

[0065] The online gradient method is an optimization algorithm used to update parameters when training a model. Unlike the traditional batch gradient descent method, the online gradient method only uses one sample at a time to calculate the gradient and updates the model's parameters based on the gradient. This method can greatly reduce the amount of computation and is especially suitable for large-scale data sets. Updating parameters using the online gradient method means that in each iteration, only the information of one sample is used to adjust the parameters of the model, rather than using the information of the entire training set as in the batch gradient descent method. This allows for faster convergence and can work more effectively on large-scale data sets. However, since only one sample is considered in each iteration, it may lead to instability in parameter updates, and a certain degree of tuning is required to ensure the performance of the model.

[0066] Therefore, there are three schemes in this scheme to update the first initial parameter. The first method is to update using the least squares method; the second method is to update using the online gradient method; and the third method is to update using the least squares method and the online gradient method.

[0067] Step S203, using a bias function to update the second initial parameter to obtain a second parameter;

[0068] Specifically, the current collected by the current sensor may have errors, so it may be updated, and specifically, an offset function may be used for updating.

[0069] Step S204: Calculate the SOC value of the battery according to the first parameter and the second parameter.

[0070] Specifically, since the first parameter is an updated parameter and the second parameter is also an updated parameter, both the first parameter and the second parameter are relatively accurate and have basically no error, and therefore, the accuracy of the calculated SOC value is also relatively high.

[0071] Through this embodiment, the least squares method and / or the online gradient method are used to update the parameters. The least squares method can reduce the influence of data errors on fitting and improve the reliability of parameter estimation. The online gradient method can easily jump out of the local optimal solution and find the global optimal solution. Therefore, the first parameter obtained by this scheme has a high accuracy. The current value of the current sensor is corrected and updated by the bias function, which can reduce the influence of the measurement result and the fluctuation of the error influence. Therefore, the second parameter obtained by this scheme has a high accuracy. Therefore, SOC estimation based on the updated parameters can reduce the estimation error and improve the accuracy of SOC estimation.

[0072] Specifically, to overcome the situation of large errors, accurate SoC estimation requires good battery model parameter estimation (i.e., estimation of initial parameters, the battery model is the Thevenin equivalent circuit model), so most indirect methods are combined with SoH estimation.

[0073] SoH estimation can be done either offline or online. Particle swarm optimization and genetic algorithms accurately estimate model parameters from experimental data, but the computational cost is high. Therefore, nonlinear least squares (NLS) can be applied to estimate the parameters of the equivalent circuit model including the ohmic resistance, and recursive least squares can be used to estimate the parameters online from the discrete-time ECM. The use of least squares greatly saves computational cost and makes the calculation smoother.

[0074] Model-based coupled SoC-SoH estimation methods include sigma point KF, Unscented KF (UKF), double EKF and nonlinear observers. For high-dimensional state space and nonlinear systems, the computational complexity of SPKF is high, especially when a large amount of measurement data needs to be processed, the computational cost will increase significantly. UKF may be affected by nonlinear metric conversion when dealing with nonlinear systems, which may lead to inaccurate state estimation, especially in highly nonlinear cases. EKF relies on linearization of system dynamics, which may lead to accumulation of estimation errors in highly nonlinear systems, especially when parameter uncertainty is large or system dynamics change greatly. Nonlinear observers have high requirements on the accuracy of system model and observation model, and may lead to instability and deviation of estimation when there is large uncertainty in system model or observation model. Based on Lyapunov's observer, which is effective in estimating the open circuit voltage (OCV) with slow changes in SOH parameters, some schemes propose an adaptive estimator of dynamic OCV, assuming that the capacity is known, give the sensitivity of estimated SoC to current sensor deviation, and quantify the steady-state SoC error estimated by KF and RLS. Compared with the model-based coupled SoC-SoH estimation method, the dynamic OCV adaptive estimator has better real-time performance, adaptability and robustness in estimating SOH parameters. It also simplifies the implementation and calculation process and is more suitable for the estimation tasks of complex nonlinear systems and uncertain systems.

[0075] This solution proposes a coupled SoC-SoH estimator estimation and battery health monitoring based on the ECM model. The online SoC estimator based on Lyapunov (Lyapunov function) using real-time voltage and current data is combined with the SoH estimator based on batch least squares (BLS), which estimates the battery capacity, ohmic resistance and current sensor bias offline based on daily voltage and current data. In the model, a nonlinear OCV curve and a pair RC are included. The solution of this application proposes an online SoC-SoH bias estimator for real-time implementation and proves its stability. This innovation solves the problem of large errors and high computational costs.

[0076] Specifically, the solution of this application uses the battery voltage and current measured by a large battery pack to propose a method for real-time SoC and batch least squares estimation of SOH and current sensor bias. An online gradient-based SoH estimator and an online SoC estimator are used to achieve real-time onboard health monitoring. At the same time, the online and offline SoC-Soh algorithms are tested using data from battery electric locomotives.

[0077] In order to ensure the accuracy of SOC and SOH estimation, a more suitable battery model needs to be established, which requires identifying suitable model parameters and reflecting the fast and slow change characteristics of the battery. In the specific implementation process, the initial parameters of the battery can be obtained by the following steps: Calculate the above initial parameters according to the first calculation formula, where the above first calculation formula is:

[0078]

[0079] V(t)=f(z(t))-V c (t)-I(t)R0,

[0080] Indicates V c (t), R1 represents the first initial resistance value, C1 represents the above initial capacitance value, V c (t) represents the initial voltage value of the capacitor, I(t) represents the initial current value input by the current sensor, represents the first-order derivative of z(t), Q represents the above-mentioned initial battery capacity, V(t) represents the output voltage value, f(z(t)) represents the OCV function of the open circuit voltage, z(t) represents the above-mentioned SOC value, R0 represents the second initial resistance value, and t represents the time.

[0081] In this solution, a first-order RC model can be used to calculate the initial parameters. The first-order RC model has relatively high accuracy, relatively low complexity, and is convenient to calculate. Therefore, relatively accurate initial parameters can be obtained through this implementation.

[0082] Specifically, at present, the research on battery models has been extensive, and a variety of different battery models have been proposed, including electrochemical models, fractional equivalent circuit models, and equivalent circuit models. Among them, the electrochemical model and the fractional equivalent circuit model are complex; their computational efficiency is low and their online application is difficult. The first-order RC model has the characteristics of high accuracy, low complexity, and convenient calculation. It is the best among the integer-order equivalent circuit models and is the first choice for the joint estimation algorithm of this scheme. The electrodynamics of the battery is given by Figure 3 The Thevenin equivalent circuit shown is modeled. The ohmic resistance R0 simulates the ultra-high frequency impedance from the electrolyte, separator, current collector and active materials. The dynamics of the RC pair formed by R1 and C1 represents a slow diffusion process, and the open circuit voltage is a nonlinear function of the SoC. The open circuit voltage OCV function f(z) is obtained by polynomial fitting and has the form:

[0083] f(z)=V0+m(z)z (Formula 2.1), m(z)z is the slope with respect to the state variable z, and V0 represents the bias voltage.

[0084] The state space equation is shown in the first calculation formula, which is referred to as Formula 2.2 in this scheme.

[0085] Specifically, a state variable dynamic equation is established to estimate the rate of change of the voltage Vc and the state variable z. The formula is:

[0086]

[0087] in,

[0088] express The rate of change over time t, represents the estimated value of the capacitor voltage, I represents the current flowing through the battery, with a positive value representing discharge and a negative value representing charge, L1 represents the first gain of the observer selected to ensure stability, L2 represents the second gain of the observer selected to ensure stability, V represents the terminal voltage of the battery, represents the estimated value of the voltage, represents the estimated value of the state variable, Indicates estimated state of charge An estimate of the open circuit voltage, Represents an estimate of the capacitor voltage.

[0089] Specifically, the error dynamics is formulated as:

[0090]

[0091] Indicates the rate of change of the capacitor voltage error over time t, ΔV c represents the error of capacitor voltage, ΔV represents the error of terminal voltage, It represents the rate of change of the error of the state of charge SoC with time t.

[0092] The above, and Output voltage error:

[0093]

[0094] Specifically, in the solution of this application, a Lyapunov function is also defined:

[0095]

[0096] Where λ>0 is used to prove the stability of state estimation.

[0097] The derivative of U1 is:

[0098]

[0099] Using Young's inequality, we can get the upper limit of the derivative U1 of the Lyapunov function, thereby determining the conditions of the observer gains L1 and L2 to ensure that the state estimation error converges to zero over time, as follows:

[0100] Using the cross terms of Young's inequality,

[0101]

[0102] Using η>0, for

[0103]

[0104] Using 2η>m, we can simplify the above inequality to and The error dynamics are made asymptotically stable and the error converges to zero. The gain condition simplifies to L1 < 0 and L2 > 0, and since λ and η are arbitrary positive numbers, the gain can be selected using pole placement.

[0105] In a specific implementation process, the SOC value of the battery is calculated according to the first parameter and the second parameter, which can be achieved by the following steps: the SOC value of the battery is calculated according to a second formula, wherein the second formula includes:

[0106]

[0107] Δz(t) represents the change in the above SOC value, represents the estimated value of the above SOC value, represents the battery capacity in the first parameter above, represents the above bias function, Indicates the current value of the second parameter mentioned above.

[0108] In this scheme, the accumulated current is used to update the battery capacity estimate and sensor bias. The accumulated current is closely related to the battery capacity and SoC, because the battery SoC can be calculated by accumulating the current flowing in or out. Since the current sensor may have a bias, if it is not corrected, this bias will accumulate as the accumulated current increases, resulting in errors in SoC and capacity estimation. This can correct the accumulated error caused by the current sensor bias, thereby improving the accuracy of SoC and capacity estimation. This is crucial for the Battery Management System (BMS) because it directly affects the battery life and safety. Accurate SoC and capacity estimation can avoid overcharging or overdischarging, both of which shorten the battery life. By accurately controlling the battery charging and discharging process, the number of cycles and total service life of the battery can be maximized. It is not only used for SoC estimation, but also for real-time monitoring of the battery's state of health (SoH). By regularly updating the estimated values ​​of capacity and sensor bias, the system can detect the degradation of battery performance and warn of possible problems in advance.

[0109] Specifically, the capacity estimation depends largely on the current sensor bias, so the capacity and current sensor bias are estimated by solving the least squares problem, and the SoC is estimated by solving (hereinafter referred to as z(t)). The second formula of this scheme can also be referred to as formula 2.10 in this scheme.

[0110] In a specific implementation process, when the above-mentioned update algorithm includes the least squares method, the update algorithm is used to update the above-mentioned first initial parameter to obtain the first parameter, which can be achieved by the following steps: the least squares solution is calculated using the third formula, wherein the above-mentioned third formula is:

[0111] in,

[0112] in, represents a vector containing the parameters to be estimated, M represents a matrix consisting of accumulated current and time, M is used to form the input matrix in the linear system, T represents transpose, F represents a function representing the difference between the estimated battery state and the initial state, plus a correction term calculated based on the accumulated current and capacity; the fourth formula is used to calculate the impedance transfer function, wherein the fourth formula is:

[0113]

[0114] Wherein, G(s) represents the transfer function, which is a mathematical description of the battery voltage response to the change of the current input, s represents a complex variable in the frequency domain, and m represents a coefficient related to the state of charge of the battery; the fifth formula is used to filter the electrical signal, wherein the fifth formula is:

[0115] ω1(t)=Λω1(t)+bI(t)

[0116] ω2(t)=Λω2(t)+bV(t),

[0117] Where, ω1(t)=[i1(t),i2(t),i3(t)] T ,ω2(t)=[v1(t),v2(t),v3(t)] T ,

[0118]

[0119] λ0=1 / τ 3 ,λ1=3 / τ 2 , λ2=3 / τ, ω1 represents the filtered current signal, ω2 represents the filtered voltage signal, λ represents the coefficient of the filter, and τ represents the time constant of the filter;

[0120] The error signal is calculated using the sixth formula, wherein the sixth formula is: e(t) = v3(t) - ω(t) T θ, e(t) represents the error signal between the filtered voltage signal and the model output. θ is a vector that contains a combination of model parameters.

[0121] ω(t)=[i1(t),i2(t),i3(t),v2(t)] T ;

[0122] The seventh formula is used to calculate the estimated parameters, wherein the seventh formula is: in, represents the corrected voltage signal, X(t) represents a matrix including current and voltage signals, β represents the above-mentioned estimated parameters, and β represents the above-mentioned estimated parameters; the above-mentioned first initial parameters are updated according to the above-mentioned estimated parameters to obtain the above-mentioned first parameters.

[0123] In this solution, the current and voltage data of the battery over a period of time are used to find the most appropriate parameters through an optimization algorithm to match the dynamic behavior of the battery. By designing appropriate observer gains, it is possible to ensure that these errors converge to zero over time, thereby improving the accuracy of state estimation. By continuously adjusting parameters to minimize the errors in voltage and state estimation, real-time adaptability and accuracy of battery parameters are ensured, and good performance can be maintained even when the battery ages or environmental conditions change. In this way, through the combined use of offline and online algorithms, the battery management system (BMS) can more accurately estimate the battery's state of charge (SoC), state of health (SoH) and other key parameters, which is crucial for the efficient and safe operation of the battery. Online parameter updates allow the system to adapt to changes in battery characteristics in real time, such as capacity decay and sensor deviations, ensuring performance consistency throughout the battery's life cycle.

[0124] By solving Can get and The estimated value of is the offset of the current sensor, and is the capacity of the battery. This process is a critical part of battery health monitoring and condition assessment as it allows real-time updates of the battery's state of health and remaining energy.

[0125] Specifically, the third formula In represents the pseudo-inverse, i.e., the pseudo-inverse matrix.

[0126] Specifically, the least squares solution of the equation is the third formula, which may also be referred to as Formula 2.11 in this solution.

[0127] Specifically, using the discretized version of

[0128] The other parameters are estimated using the impedance transfer function of the linearized equation, namely the fourth formula, which may also be referred to as formula 2.12 in this solution.

[0129] Two third-order filters are used to filter the bias-corrected current and voltage signals, that is, the fifth formula is used for filtering, which may also be referred to as formula 2.13 in this solution. The filtering time constant τ is selected to provide the required bandwidth.

[0130] The outputs of the two filters form an error model, that is, the error signal is calculated according to the sixth formula, which can also be called formula 2.14 in this scheme.

[0131] Specifically, the best fitting parameters satisfy The slope is decomposed from formula (2.14) for the optimization problem, that is, the seventh formula is used to calculate the error, which can also be called formula 2.15 in this scheme.

[0132] in, and

[0133] In some embodiments, the first initial parameter is updated according to the estimated parameter to obtain the first parameter, which can be specifically achieved by the following steps: the first parameter is calculated according to the eighth formula, wherein the eighth formula is:

[0134]

[0135] represents the first resistance value, represents the second resistance value, Indicates the capacitance value, express The second element in the vector, express The third element in the vector, express The fourth element in the vector.

[0136] In this solution, key parameters in the battery equivalent circuit model are obtained, which are crucial for understanding and optimizing battery performance. This enhances the capabilities of the battery management system by providing accurate battery parameter estimation, enabling accurate monitoring of battery status, performance optimization, and safe operation, thereby significantly extending battery life and improving overall system efficiency.

[0137] Specifically, the optimization problem is discretized using N samples with a fixed sampling rate. 3×N and Instead of estimating the capacity, in the pseudo-inverse Calculate parameter estimates in The estimated values ​​of the physical parameters R0, R1 and C1 can be obtained by the eighth formula, which may also be referred to as formula 2.16 in this solution.

[0138] Estimated parameters and Update the model and estimate the bias The measured current is corrected. These updated parameters are fed to the SoC estimator to obtain a more accurate SoC estimate. Additionally, in the case of noisy regression matrices (M and χ), total least squares can be performed instead of BLS to have accurate parameter estimates.

[0139] In some embodiments, when the update algorithm includes an online gradient method, the update algorithm is used to update the first initial parameter to obtain the first parameter, which can be specifically implemented by the following steps: when the update algorithm includes an online gradient method, the update algorithm is used to update the first initial parameter to obtain the first parameter, including: calculating the error according to the ninth formula and the tenth formula, wherein the ninth formula is: The tenth formula above is: in,

[0140]

[0141] ψ(t)=[i2(t)i3(t)] T ,

[0142] e v3 (t) represents the error between the third voltage measurement value and the estimated voltage value, v3(t) represents the third voltage measurement value, represents the estimated value of the third voltage, e z (t) represents the difference between the actual battery state of charge and the estimated value of the state, represents the estimated value of the parameter vector, mainly including parameters related to the battery resistance and capacitance, ψ(t) represents a vector containing measurements related to the battery dynamics, such as secondary current and tertiary current, v2(t) represents another voltage measurement, m(z(t)) represents a function related to the battery SoC, which affects the battery open circuit voltage, and α1 represents the parameter vector An element in, related to the battery capacity or current sensor deviation, M(t) T represents the transpose of the matrix, which consists of the accumulated current and time of the battery, represents a vector containing estimated values ​​related to the battery model parameters.

[0143] In this solution, based on the gradient descent method, which is an iterative optimization algorithm, it is used to minimize error functions, such as the error of battery voltage and state estimation. By continuously adjusting the parameter vector, it is attempted to gradually reduce these errors until convergence, so that real-time adaptive adjustment of battery parameters and accurate state estimation can be achieved, which not only significantly enhances the performance and safety of the battery management system, but also optimizes the battery charging and discharging strategy, and warns of potential faults, thereby effectively extending the battery life and improving the overall operating efficiency.

[0144] Specifically, the recursive estimation of SoC, R0, Q, and b provides a real-time SOH metric. The SoC estimator has a relatively fast dynamic speed (seconds / minutes). However, since the model parameters are expected to change slowly, the SOH estimation can be performed very slowly (days). Figure 4The block diagram of the online coupled SoC-SoH estimator is shown. A gradient-based parameter estimator is proposed to estimate:

[0145] and Minimize errors.

[0146] Specifically, the ninth formula in this solution can also be referred to as formula 2.17.

[0147] Specifically, the tenth formula in this solution can also be referred to as formula 2.18.

[0148] In some embodiments, when the update algorithm includes an online gradient method, the update algorithm is used to update the first initial parameter to obtain the first parameter, which can be specifically implemented by the following steps: calculating the gradient update law according to the eleventh formula, wherein the eleventh formula is:

[0149]

[0150] in, express The rate of change over time t, γ 12 represents the design parameter, also known as gain, used to adjust the speed and stability of the update rule, represents the partial derivative, φ represents a parameter vector, which contains the parameters related to the dynamic characteristics of the battery, express The rate of change over time t, γ 34 represents the second design parameter, also known as gain, which is used to adjust the speed and stability of the update rule, α represents a parameter vector, which includes parameters related to the state of charge of the battery and the current sensor deviation, and M(t) represents a regression vector, which includes the measurement values ​​related to the accumulated current time t; the error is updated according to the twelfth formula, where the above twelfth formula is:

[0151]

[0152] Where m represents the coefficient related to the battery model, represents the deviation between the true value and the estimated value of the α parameter, represents the deviation between the true value and the estimated value of the φ parameter, represents the second deviation between the true value of the φ parameter and the estimated value, represents the deviation between the true value and the estimated value of the α parameter; the first initial parameter is updated according to the thirteenth formula to obtain the first parameter, wherein the thirteenth formula is:

[0153]

[0154] In this solution, an online update mechanism is described for real-time adjustment of dynamic parameter estimates of the battery, including but not limited to the battery's state of health (SoH) parameters. This allows an online update mechanism, and the battery management system is able to adjust its estimate of the battery state in real time to adapt to changes in battery parameters over time and usage conditions. This real-time adaptability improves the accuracy of the state estimate and is particularly important for large battery packs, as they may experience more complex temperature, current, and aging effects.

[0155] Specifically, the eleventh formula of the present application may also be referred to as formula 2.19. To prove stability, it is assumed that the time constant R1C1 is known and the battery parameters change slowly.

[0156] The error change is calculated by the twelfth formula, which in this scheme may also be referred to as formula 2.20.

[0157] Specifically, in and if And the gain matrix The gradient update formula can be written as:

[0158]

[0159] Specifically, the thirteenth formula is a semi-positive definite matrix with eigenvalues Exponential stability of the system For λ min (A+A T ) ≥ 0, which is guaranteed under the continuous excitation of ψ(t) and M(t). In the dynamic operation of electric vehicles or locomotives, the continuous excitation condition is usually valid.

[0160] When the parameters (φ, α) and the state (z, V c ) are known. The stability of the coupled system can be proven by demonstrating the stability of the inputs to the state of both systems, following the discussed small gain theorem. A least squares based offline SoH estimator uses a batch of experimental voltage and current data to periodically update the model parameters and current sensor bias, potentially improving the online SoC estimation of each cell. An online estimator is implemented to monitor the module-level health parameters of Q and R in real time.

[0161] In summary, the main contents of the solution of this application are as follows:

[0162] The problem of large errors is addressed by adopting a least squares based offline SoH estimator. This estimator uses a batch of experimental voltage and current data to periodically update the model parameters and current sensor bias, potentially improving the online SoC estimation of each cell. In order to achieve accurate SoC estimation, good battery model parameter estimation is required, so an indirect method combined with SoH estimation is adopted to minimize the error.

[0163] By combining the offline and online methods, the data of battery electric locomotives were experimentally tested. The offline state of health (SoH) algorithm showed a lower open-loop voltage prediction, indicating that the accuracy of SoC and parameter estimation was verified. This combined method overcomes the limitations that may exist in a single method. Traditional offline methods may not be able to capture the characteristics of real-time changes, while online methods may be limited by the stability and real-time nature of the data. By combining the two methods, their respective limitations can be compensated and the accuracy and reliability of battery health assessment can be improved. At the same time, a more comprehensive and reliable battery health assessment is provided.

[0164] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the method for determining the SOC of a battery of the present application will be described in detail below in conjunction with specific embodiments.

[0165] This embodiment relates to a specific method for determining the SOC of a battery. Figure 5 As shown, the following steps are included:

[0166] Collect measurement data, initialize battery model parameters, update battery model parameters using batch least squares and online gradient methods, correct current sensor bias by estimating and applying bias values, and estimate SOC using updated battery model parameters.

[0167] For offline SOH batch least squares estimation, parameter estimation is performed based on historical data. For online SOH real-time estimation, online gradient method is used to monitor battery monitoring in real time.

[0168] Determine whether the parameter estimation is stable, and output the estimated values ​​of SOC and SOH if stable.

[0169] The embodiments of the present application also provide a device for determining the SOC of a battery. It should be noted that the device for determining the SOC of a battery in the embodiments of the present application can be used to execute the method for determining the SOC of a battery provided in the embodiments of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and those that have been described will not be repeated here. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0170] The following is an introduction to the device for determining the SOC of a battery provided in an embodiment of the present application.

[0171] Figure 6 is a structural block diagram of a device for determining the SOC of a battery according to an embodiment of the present application. Figure 6 As shown, the device comprises:

[0172] An acquisition unit 10 is used to acquire initial parameters of the battery, wherein the initial parameters include a first initial parameter and a second initial parameter, the first initial parameter includes at least one of an initial capacitance value, an initial voltage value, an initial resistance value and an initial battery capacity, and the second initial parameter includes an initial current value input by a current sensor;

[0173] A first updating unit 20, configured to update the first initial parameter by adopting an updating algorithm to obtain a first parameter, wherein the updating algorithm includes a least square method and / or an online gradient method;

[0174] A second updating unit 30, configured to update the second initial parameter using a bias function to obtain a second parameter;

[0175] The calculation unit 40 is used to calculate the SOC value of the battery according to the first parameter and the second parameter.

[0176] Through this embodiment, the least squares method and / or the online gradient method are used to update the parameters. The least squares method can reduce the influence of data errors on fitting and improve the reliability of parameter estimation. The online gradient method can easily jump out of the local optimal solution and find the global optimal solution. Therefore, the first parameter obtained by this scheme has a high accuracy. The current value of the current sensor is corrected and updated by the bias function, which can reduce the influence of the measurement result and the fluctuation of the error influence. Therefore, the second parameter obtained by this scheme has a high accuracy. Therefore, SOC estimation based on the updated parameters can reduce the estimation error and improve the accuracy of SOC estimation.

[0177] In order to ensure the accuracy of SOC and SOH estimation, it is necessary to establish a more suitable battery model, which requires identifying appropriate model parameters and reflecting the fast and slow change characteristics of the battery. In the specific implementation process, the acquisition unit includes a first calculation module, which is used to calculate the above initial parameters according to a first calculation formula, wherein the above first calculation formula is:

[0178]

[0179] V(t)=f(z(t))-V c (t)-I(t)R0,

[0180] Indicates V c (t), R1 represents the first initial resistance value, C1 represents the above initial capacitance value, V c (t) represents the initial voltage value of the capacitor, I(t) represents the initial current value input by the current sensor, represents the first-order derivative of z(t), Q represents the above-mentioned initial battery capacity, V(t) represents the output voltage value, f(z(t)) represents the OCV function of the open circuit voltage, z(t) represents the above-mentioned SOC value, R0 represents the second initial resistance value, and t represents the time.

[0181] In this solution, a first-order RC model can be used to calculate the initial parameters. The first-order RC model has relatively high accuracy, relatively low complexity, and is convenient to calculate. Therefore, relatively accurate initial parameters can be obtained through this implementation.

[0182] In a specific implementation process, the calculation unit includes a second calculation module, and the second calculation module is used to calculate the SOC value of the battery according to a second formula, wherein the second formula includes:

[0183]

[0184] Δz(t) represents the change in the above SOC value, represents the estimated value of the above SOC value, represents the battery capacity in the first parameter above, represents the above bias function, Indicates the current value of the second parameter mentioned above.

[0185] In this scheme, the accumulated current is used to update the battery capacity estimate and sensor bias. The accumulated current is closely related to the battery capacity and SoC, because the battery SoC can be calculated by accumulating the current flowing in or out. Since the current sensor may have a bias, if it is not corrected, this bias will accumulate as the accumulated current increases, resulting in errors in SoC and capacity estimation. This can correct the accumulated error caused by the current sensor bias, thereby improving the accuracy of SoC and capacity estimation. This is crucial for the Battery Management System (BMS) because it directly affects the battery life and safety. Accurate SoC and capacity estimation can avoid overcharging or overdischarging, both of which shorten the battery life. By accurately controlling the battery charging and discharging process, the number of cycles and total service life of the battery can be maximized. It is not only used for SoC estimation, but also for real-time monitoring of the battery's state of health (SoH). By regularly updating the estimated values ​​of capacity and sensor bias, the system can detect the degradation of battery performance and warn of possible problems in advance.

[0186] In a specific implementation process, when the above-mentioned update algorithm includes the least squares method, the first update unit includes a third calculation module, a fourth calculation module, a fifth calculation module, a sixth calculation module, a seventh calculation module and a first update module, and the third calculation module is used to calculate the least squares solution using a third formula, wherein the above-mentioned third formula is:

[0187] in,

[0188] in, represents a vector containing parameters to be estimated, M represents a matrix consisting of accumulated current and time, M is used to form an input matrix in a linear system, T represents transpose, F represents a function representing the difference between the estimated battery state and the initial state, plus a correction term calculated based on the accumulated current and capacity; the fourth calculation module is used to calculate the impedance transfer function using the fourth formula, wherein the fourth formula is:

[0189]

[0190] Wherein, G(s) represents a transfer function, which is a mathematical description of the battery voltage response to the change of the current input, s represents a complex variable in the frequency domain, and m represents a coefficient related to the state of charge of the battery; the fifth calculation module is used to filter the electrical signal using the fifth formula, wherein the fifth formula is:

[0191] ω1(t)=Λω1(t)+bI(t)

[0192] ω2(t)=Λω2(t)+bV(t),

[0193] Where, ω1(t)=[i1(t),i2(t),i3(t)] T ,ω2(t)=[v1(t),v2(t),v3(t)] T ,

[0194]

[0195] λ0=1 / τ 3 ,λ1=3 / τ 2 , λ2=3 / τ, ω1 represents the filtered current signal, ω2 represents the filtered voltage signal, λ represents the coefficient of the filter, and τ represents the time constant of the filter;

[0196] The sixth calculation module is used to calculate the error signal using the sixth formula, wherein the sixth formula is: e(t)=v3(t)-ω(t) T θ, e(t) represents the error signal between the filtered voltage signal and the model output. θ is a vector that contains a combination of model parameters.

[0197] ω(t)=[i1(t),i2(t),i3(t),v2(t)] T ;

[0198] The seventh calculation module is used to calculate the estimated parameter using the seventh formula, wherein the seventh formula is: in, represents the corrected voltage signal, X(t) represents a matrix containing current and voltage signals, β represents the above-mentioned estimated parameters, and β represents the above-mentioned estimated parameters; the first updating module is used to update the above-mentioned first initial parameters according to the above-mentioned estimated parameters to obtain the above-mentioned first parameters.

[0199] In this solution, the current and voltage data of the battery over a period of time are used to find the most appropriate parameters through an optimization algorithm to match the dynamic behavior of the battery. By designing appropriate observer gains, it is possible to ensure that these errors converge to zero over time, thereby improving the accuracy of state estimation. By continuously adjusting parameters to minimize the errors in voltage and state estimation, real-time adaptability and accuracy of battery parameters are ensured, and good performance can be maintained even when the battery ages or environmental conditions change. In this way, through the combined use of offline and online algorithms, the battery management system (BMS) can more accurately estimate the battery's state of charge (SoC), state of health (SoH) and other key parameters, which is crucial for the efficient and safe operation of the battery. Online parameter updates allow the system to adapt to changes in battery characteristics in real time, such as capacity decay and sensor deviations, ensuring performance consistency throughout the battery's life cycle.

[0200] In some embodiments, the update module includes a calculation submodule, and the calculation submodule is used to calculate the first parameter according to an eighth formula, wherein the eighth formula is:

[0201]

[0202] represents the first resistance value, represents the second resistance value, Indicates the capacitance value, express The second element in the vector, express The third element in the vector, express The fourth element in the vector.

[0203] In this solution, key parameters in the battery equivalent circuit model are obtained, which are crucial for understanding and optimizing battery performance. This enhances the capabilities of the battery management system by providing accurate battery parameter estimation, enabling accurate monitoring of battery status, performance optimization, and safe operation, thereby significantly extending battery life and improving overall system efficiency.

[0204] In some embodiments, when the update algorithm includes an online gradient method, the first calculation unit includes an eighth calculation module, and the eighth calculation module is used to update the first initial parameter using the update algorithm to obtain the first parameter, including: calculating the error according to the ninth formula and the tenth formula, wherein the ninth formula is: The tenth formula above is: in,

[0205]

[0206] ψ(t)=[i2(t)i3(t)] T ,

[0207] e v3 (t) represents the error between the third voltage measurement value and the estimated voltage value, v3(t) represents the third voltage measurement value, represents the estimated value of the third voltage, e z (t) represents the difference between the actual battery state of charge and the estimated value of the state, represents the estimated value of the parameter vector, which mainly includes parameters related to the battery resistance and capacitance, ψ(t) represents a vector containing measurements related to the battery dynamics, such as secondary current and tertiary current, v2(t) represents another voltage measurement, m(z(t)) represents a function related to the battery's SoC, which affects the battery's open circuit voltage, and α1 represents the parameter vector An element in, related to the battery capacity or current sensor deviation, M(t) T represents the transpose of the matrix, which consists of the accumulated current and time of the battery, represents a vector containing estimated values ​​related to the battery model parameters.

[0208] In this solution, based on the gradient descent method, which is an iterative optimization algorithm, it is used to minimize error functions, such as the error of battery voltage and state estimation. By continuously adjusting the parameter vector, it is attempted to gradually reduce these errors until convergence, so that real-time adaptive adjustment of battery parameters and accurate state estimation can be achieved, which not only significantly enhances the performance and safety of the battery management system, but also optimizes the battery charging and discharging strategy, and warns of potential faults, thereby effectively extending the battery life and improving the overall operating efficiency.

[0209] In some embodiments, when the update algorithm includes an online gradient method, the first computing unit further includes a ninth computing module, a second updating module, and a third updating module, and the ninth computing module is used to calculate the gradient update law according to the eleventh formula, wherein the eleventh formula is:

[0210]

[0211] in, express The rate of change over time t, γ 12 represents the design parameter, also known as gain, used to adjust the speed and stability of the update rule, represents the partial derivative, φ represents a parameter vector, which contains the parameters related to the dynamic characteristics of the battery, express The rate of change over time t, γ 34 represents the second design parameter, also known as gain, which is used to adjust the speed and stability of the update rule, α represents a parameter vector, which includes parameters related to the state of charge of the battery and the current sensor deviation, and M(t) represents a regression vector, which includes the measurement value related to the accumulated current time t; the second update module is used to update the error according to the twelfth formula, wherein the above twelfth formula is:

[0212]

[0213] Where m represents the coefficient related to the battery model, represents the deviation between the true value and the estimated value of the α parameter, represents the deviation between the true value and the estimated value of the φ parameter, represents the second deviation between the true value of the φ parameter and the estimated value, represents the deviation between the true value and the estimated value of the α parameter; the third updating module is used to update the first initial parameter according to the thirteenth formula to obtain the first parameter, wherein the thirteenth formula is:

[0214]

[0215] In this solution, an online update mechanism is described for real-time adjustment of dynamic parameter estimates of the battery, including but not limited to the battery's state of health (SoH) parameters. This allows an online update mechanism, and the battery management system is able to adjust its estimate of the battery state in real time to adapt to changes in battery parameters over time and usage conditions. This real-time adaptability improves the accuracy of the state estimate and is particularly important for large battery packs, as they may experience more complex temperature, current, and aging effects.

[0216] The battery SOC determination device includes a processor and a memory, wherein the acquisition unit, the first update unit, the second update unit and the calculation unit are stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions. The modules are all located in the same processor; or, the modules are located in different processors in any combination.

[0217] The processor includes a kernel, and the kernel calls the corresponding program unit from the memory. One or more kernels can be set, and the problem of low accuracy of SOC estimation in the prior art can be solved by adjusting kernel parameters.

[0218] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0219] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for determining the SOC of the battery.

[0220] An embodiment of the present invention provides a processor, and the processor is used to run a program, wherein the method for determining the SOC of the battery is executed when the program is run.

[0221] An embodiment of the present invention provides a device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, at least the steps of a method for determining the SOC of a battery are implemented.

[0222] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0223] A computer program product includes a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for determining the SOC of the battery in each embodiment of the present application are implemented.

[0224] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0225] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0226] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0227] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0228] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0229] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0230] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0231] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0232] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0233] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0234] 1) The method for determining the SOC of the battery of the present application adopts the least squares method and / or the online gradient method to update the parameters. The least squares method can reduce the influence of data errors on fitting and improve the reliability of parameter estimation. The online gradient method can easily jump out of the local optimal solution and find the global optimal solution. Therefore, the accuracy of the first parameter obtained by this scheme is relatively high. The current value of the current sensor is corrected and updated by the bias function, which can reduce the influence of the measurement result and reduce the fluctuation of the error influence. Therefore, the accuracy of the second parameter obtained by this scheme is relatively high. Therefore, SOC estimation is performed according to the updated parameters, which can reduce the estimation error and improve the accuracy of SOC estimation.

[0235] 2) The device for determining the SOC of the battery of the present application adopts the least squares method and / or the online gradient method to update the parameters. The least squares method can reduce the influence of data errors on fitting and improve the reliability of parameter estimation. The online gradient method can easily jump out of the local optimal solution and find the global optimal solution. Therefore, the accuracy of the first parameter obtained by this scheme is relatively high. The current value of the current sensor is corrected and updated by the bias function, which can reduce the influence of the measurement result and the fluctuation of the error influence. Therefore, the accuracy of the second parameter obtained by this scheme is relatively high. Therefore, SOC estimation is performed according to the updated parameters, which can reduce the estimation error and improve the accuracy of SOC estimation.

[0236] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for determining the SOC of a battery, characterized in that: include: Acquire initial parameters of the battery, wherein the initial parameters include a first initial parameter and a second initial parameter, the first initial parameter includes at least one of an initial capacitance value, an initial voltage value, an initial resistance value, and an initial battery capacity, and the second initial parameter includes an initial current value input by a current sensor; Using an update algorithm to update the first initial parameter to obtain a first parameter, wherein the update algorithm includes a least squares method and / or an online gradient method; Updating the second initial parameter using a bias function to obtain a second parameter; The SOC value of the battery is calculated according to the first parameter and the second parameter.

2. The method according to claim 1, characterized in that Get the initial parameters of the battery, including: The initial parameters are calculated according to a first calculation formula, wherein the first calculation formula is: V(t)=f(z(t))-V c (t)-I(t)R0, Indicates V c (t), R1 represents the first initial resistance value, C1 represents the initial capacitance value, V c I(t) represents the initial voltage value of the capacitor, I(t) represents the initial current value input by the current sensor, represents the first-order derivative of z(t), Q represents the initial battery capacity, V(t) represents the output voltage value, f(z(t)) represents the OCV function of the open circuit voltage, z(t) represents the SOC value, R0 represents the second initial resistance value, and t represents the time.

3. The method according to claim 1, characterized in that Calculating the SOC value of the battery according to the first parameter and the second parameter includes: The SOC value of the battery is calculated according to a second formula, wherein the second formula includes: Δz(t) represents the change in the SOC value, represents an estimate of the SOC value, represents the battery capacity in the first parameter, represents the bias function, represents the current value of the second parameter.

4. The method according to claim 1, characterized in that: In the case where the updating algorithm includes the least squares method, the updating algorithm is used to update the first initial parameter to obtain the first parameter, including: The least squares solution is calculated using the third formula, where the third formula is: in, in, represents a vector containing the parameters to be estimated, M represents a matrix consisting of accumulated current and time, M is used to form the input matrix in the linear system, T represents the transpose, F represents a function representing the difference between the estimated battery state and the initial state, plus a correction term calculated based on the accumulated current and capacity; The impedance transfer function is calculated using the fourth formula, wherein the fourth formula is: Where G(s) is the transfer function, a mathematical description of the battery voltage response to changes in current input, s is a complex variable in the frequency domain, and m is a coefficient related to the battery's state of charge. The fifth formula is used to filter the electrical signal, wherein the fifth formula is: ω1(t)=Λω1(t)+bI(t) ω2(t)=Λω2(t)+bV(t), where ω1(t) = [i1(t), i2(t), i3(t)] T , ω2(t) = [v1(t), v2(t), v3(t)] T , λ0=1 / τ 3 ,λ1=3 / τ 2 , λ2=3 / τ, ω1 represents the filtered current signal, ω2 represents the filtered voltage signal, λ represents the coefficient of the filter, and τ represents the time constant of the filter; The error signal is calculated using the sixth formula, wherein the sixth formula is: e(t)=v3(t)-ω(t) T θ, e(t) represents the error signal between the filtered voltage signal and the model output. θ is a vector that contains a combination of model parameters. ω(t)=[i1(t),i2(t),i3(t),v2(t)] T ; The estimated parameters are calculated using the seventh formula, wherein the seventh formula is: in, represents the corrected voltage signal, X(t) represents a matrix containing current and voltage signals, and β represents the estimated parameter; The first initial parameter is updated according to the estimated parameter to obtain the first parameter.

5. The method according to claim 4, characterized in that Updating the first initial parameter according to the estimated parameter to obtain the first parameter includes: The first parameter is calculated according to an eighth formula, wherein the eighth formula is: represents the first resistance value, represents the second resistance value, Indicates the capacitance value, express The second element in the vector, express The third element in the vector, express The fourth element in the vector.

6. The method according to claim 1, characterized in that In the case where the update algorithm includes an online gradient method, the first initial parameter is updated using the update algorithm to obtain the first parameter, including: The error is calculated according to the ninth formula and the tenth formula, wherein the ninth formula is: The tenth formula is: in, e v3 (t) represents the error between the third voltage measurement value and the estimated voltage value, v3(t) represents the third voltage measurement value, represents the estimated value of the third voltage, e z (t) represents the difference between the actual battery state of charge and the estimated value of the state, represents the estimated value of the parameter vector, which mainly includes parameters related to the battery resistance and capacitance, ψ(t) represents a vector containing measurements related to the battery dynamics, v2(t) represents another voltage measurement, m(z(t)) represents a function related to the battery SoC, which affects the battery open circuit voltage, and α1 represents the parameter vector An element in, related to the battery capacity or current sensor deviation, M(t) T represents the transpose of the matrix, which consists of the accumulated current and time of the battery, represents a vector containing estimated values ​​related to the battery model parameters.

7. The method according to claim 6, characterized in that In the case where the update algorithm includes an online gradient method, the first initial parameter is updated using the update algorithm to obtain the first parameter, further comprising: The gradient update law is calculated according to the eleventh formula, wherein the eleventh formula is: in, express The rate of change over time t, γ 12 represents the design parameter used to adjust the speed and stability of the update rule, represents the partial derivative, φ represents a parameter vector, which contains the parameters related to the dynamic characteristics of the battery, express The rate of change over time t, γ 34 represents the second design parameter used to adjust the speed and stability of the update rule, α represents a parameter vector containing parameters related to the battery state of charge and current sensor bias, M(t) represents a regression vector containing the measurement values ​​related to the accumulated current time t; The error is updated according to the twelfth formula, wherein the twelfth formula is: Where m represents the coefficient related to the battery model, represents the deviation between the true value and the estimated value of the α parameter, represents the deviation between the true value and the estimated value of the φ parameter, represents the second deviation between the true value of the φ parameter and the estimated value, Represents the deviation between the true value and the estimated value of the α parameter; The first initial parameter is updated according to the thirteenth formula to obtain the first parameter, wherein the thirteenth formula is:

8. A device for determining the SOC of a battery, characterized in that: include: an acquisition unit, configured to acquire initial parameters of the battery, wherein the initial parameters include a first initial parameter and a second initial parameter, the first initial parameter includes at least one of an initial capacitance value, an initial voltage value, an initial resistance value, and an initial battery capacity, and the second initial parameter includes an initial current value input by a current sensor; A first updating unit, configured to update the first initial parameter by adopting an updating algorithm to obtain a first parameter, wherein the updating algorithm includes a least square method and / or an online gradient method; A second updating unit, configured to update the second initial parameter by using a bias function to obtain a second parameter; A calculation unit is used to calculate the SOC value of the battery according to the first parameter and the second parameter.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for determining the SOC of a battery according to any one of claims 1 to 7 are implemented.

10. A battery SOC detection system, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for determining the SOC of a battery as described in any one of claims 1 to 7.