SRU-based lithium battery SOE estimation method and system, terminal and storage medium
Through the SRU-based lithium battery SOE estimation method, combined with the fourth-order RC equivalent circuit model and adaptive dual-expanded Kalman filtering, the problem of difficult to accurately characterize the state of the lithium battery in the cascade is solved, the accuracy and adaptability of the battery state estimation are improved, and effective means for the safety control of the lithium battery in the cascade is provided.
Patent Information
- Application Number
- CN202510095134.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-21
- Publication Date
- 2025-06-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The prior art is difficult to accurately characterize the state of the lithium battery in the cascade, especially because the lithium battery has poor operating voltage consistency, and it is difficult to accurately reflect the actual state of the battery with only SOC.
Using the SRU-based SOE estimation method of lithium batteries, the ohmic internal resistance and actual energy are obtained by establishing a fourth-order RC equivalent circuit model, and the SRU prediction model is constructed based on historical data to obtain the optimal predicted ohmic internal resistance and actual energy, and the SOE estimation is performed using adaptive double-expanded Kalman filtering.
It improves the accuracy and adaptability of ohmic internal resistance, actual energy and SOE estimation of lithium batteries, and provides an effective means of safety control of lithium batteries in cascade utilization.
Smart Images
Figure CN120103150A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of lithium-ion battery safety control, and in particular relates to a lithium battery SOE estimation method, system, terminal and storage medium based on SRU. Background Art
[0002] Due to the energy crisis and environmental pollution, new energy sources have received more and more attention. Lithium-ion batteries are widely used in electric vehicles and energy storage systems due to their high energy, high power density, and environmental protection.
[0003] With the widespread application of lithium-ion batteries, second-life utilization has become a key method to solve the problem of battery retirement. Second-life lithium-ion batteries refer to power lithium batteries with a capacity of less than 80%. Second-life batteries still have a high discharge capacity and can be used for backup power and other occasions.
[0004] Due to the decline in performance of second-life lithium batteries and the poor consistency of lithium battery operating voltage, it is difficult to accurately characterize the status of second-life lithium batteries using SOC alone. Summary of the invention
[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a lithium battery SOE estimation method, system, terminal and storage medium based on SRU to solve the above-mentioned technical problems.
[0006] In a first aspect, the present invention provides a lithium battery SOE estimation method based on SRU, comprising: S1, establishing a fourth-order RC equivalent circuit model of the second-life lithium battery to be estimated, and obtaining the ohmic internal resistance and actual energy of the lithium battery based on the fourth-order RC equivalent circuit model; S2, obtaining the historical ohmic internal resistance sequence and the historical actual energy sequence, and building the SRU prediction model based on the historical ohmic internal resistance sequence and the historical actual energy sequence; S3, based on the current ohmic internal resistance and the current actual energy combined with the SRU prediction model, obtain the optimal predicted ohmic internal resistance and the optimal predicted actual energy; S4, based on the adaptive dual extended Kalman filter combined with the optimal predicted ohmic internal resistance and the optimal predicted actual energy, the lithium battery SOE is estimated.
[0007] In an optional implementation, the specific steps of step S2 include: The historical ohmic internal resistance series and the historical actual energy series are used as data sets, and the data sets are divided into training sets, validation sets, and test sets; The input vector of the model is the historical ohmic internal resistance and the historical actual energy and two input layer neuron numbers are set. The output vector of the model is the optimal predicted ohmic internal resistance and the optimal predicted actual energy and two output layer neuron numbers are set. Initialize the model’s weights and biases; The training set is input into the SRU prediction model, and the predicted output is calculated through forward propagation. The loss value between the actual output and the predicted output in the training set is calculated based on the loss function. The gradient of the loss function to the model weights and biases is calculated through back propagation, and the weights and biases of the model are updated using the gradient information. The preset number of training times per interval is used to calculate the evaluation index based on the validation set combined with the root mean square error function, and the model structure is adjusted based on the changing trend of the evaluation index; The trained SRU prediction model is finally evaluated based on the test set.
[0008] In an optional implementation, the forward propagation calculation specifically includes: Perform a linear transformation on the input vector, calculated as follows:
[0009] in, is a linear transformation of t time steps, is the weight matrix, is the input vector at time step t; The forget gate is calculated based on the input vector as follows:
[0010] in, is the forget gate output at time step t, is the Sigmoid function, is the weight matrix associated with the forget gate, is the bias vector of the forget gate; The internal state of the forget gate is calculated based on the forget gate output and the linear transformation, which is calculated as follows:
[0011] in, is the internal state of the forget gate at time step t, is the internal state of the forget gate at time step t-1, It is the operation between corresponding elements of the matrix; The reset gate is calculated based on the input vector as follows:
[0012] in, is the reset gate output at time step t, is the weight matrix associated with the reset gate, is the bias vector for the reset gate; The output vector is calculated based on the reset gate output and the internal state of the forget gate as follows:
[0013] in, is the output vector at time step t, is the activation function.
[0014] In an optional implementation, using gradient information to update the weights and biases of the model specifically includes: Initialize the first-order moment estimate used to track the gradient and update the first-order moment estimate, calculated as follows:
[0015] Where m is the first-order moment estimate, is a hyperparameter between 0 and 1, is the gradient of the weight matrix or the gradient of the bias vector; Initialize the second-order moment estimate used to track the gradient and update the second-order moment estimate, calculated as follows:
[0016] in, is the second-order moment estimate, is another hyperparameter between 0 and 1; The bias correction is performed on the first-order moment estimate and the second-order moment estimate respectively; The model parameters are updated based on the first-order moment estimate and the second-order moment estimate after bias correction under the same gradient, and the calculation is as follows:
[0017] Where A is the model parameter weight matrix or bias vector, yes, is the corrected first-order moment estimate, is the modified second-order moment estimate, This is a value that prevents the denominator from being zero.
[0018] In an optional implementation, step S4 specifically includes: After adding the ohmic internal resistance and actual energy, the state equation, observation equation and parameter equation formulas of the lithium battery SOE estimation system are:
[0019]
[0020]
[0021] in, is the system parameter variable at time k, ; is the system state variable at time k, is the system state variable at time k+1; is the input variable of the system, i.e., the lithium-ion battery current; is the system observation variable at time k+1, i.e., the operating voltage of the lithium-ion battery; is zero-mean Gaussian white noise; is the system parameter variable at time k+1, is the system parameter variable at time k, is the noise mean at time k, that is, zero-mean Gaussian white noise; is the noise associated with the parameter variable at time k, that is, zero-mean Gaussian white noise; is the system output variable at time k+1; The lithium battery SOE estimation based on adaptive dual extended Kalman filter combined with optimal predicted ohmic internal resistance and optimal predicted actual energy includes: S4-1, calculate the expected value of the initial system parameter variable and the initial error covariance matrix; S4-2, calculating the volume point product at the previous moment based on the error covariance matrix at the previous moment and the predicted value of the state variable at the previous moment, and obtaining the volume point at the current moment through the state transfer function; S4-3, calculating the average value of all volume points at the current moment, obtaining the state variable prediction value at the current moment, and calculating the state prediction covariance matrix based on the volume points and the state variable prediction value at the current moment; S4-4, based on the state prediction covariance and the state prediction value at the current moment, the volume point product in the state update is calculated, and based on the observation function, the predicted value of the observation value is calculated for another volume point at the current moment; the predicted values of all the observation values are averaged to obtain the predicted value of the observed state; based on the predicted value of the observation value and the predicted value of the observed state, the observation value covariance matrix is calculated, and the cross covariance matrix between another volume point and the state prediction value and between the observation value prediction value and the observed state prediction value is calculated, and the Kalman gain is calculated based on the observation value covariance matrix and the cross covariance matrix; S4-5, using the Kalman gain and the difference between the observed value and the observed state prediction value, the state prediction value at the current moment is updated to obtain the optimal state estimation value, and the optimal SOE estimation is obtained based on the optimal state estimation value.
[0022] In an optional embodiment, in step S3, after obtaining the optimal predicted ohmic internal resistance, the optimal predicted ohmic internal resistance is judged against the safety threshold. When the optimal predicted ohmic internal resistance is greater than the safety threshold, the difference between the optimal predicted ohmic internal resistance and the safety threshold is calculated, and the adjustment coefficient is determined according to the size of the difference, and the current charge and discharge current is adjusted based on the adjustment coefficient.
[0023] In an optional embodiment, in step S1, the SOE of the second-life lithium battery is the ratio of the remaining energy of the lithium battery to the maximum available energy, which is calculated as follows:
[0024] in, and They are the SOE values of the discrete states of the second-life lithium battery at time k and k+1 respectively; It is to utilize the energy of lithium batteries in stages; and They are the current and working voltage of the lithium battery in discrete states at time k; Sampling period.
[0025] In a second aspect, the present invention provides a lithium battery SOE estimation system based on SRU. When the system is implemented, the above-mentioned lithium battery SOE estimation method based on SRU is executed. The system includes: A parameter acquisition module is used to establish a fourth-order RC equivalent circuit model of the second-life lithium battery to be estimated, and to obtain the ohmic internal resistance and actual energy of the lithium battery based on the fourth-order RC equivalent circuit model; An SRU prediction model building module obtains the historical ohmic internal resistance sequence and the historical actual energy sequence, and builds an SRU prediction model based on the historical ohmic internal resistance sequence and the historical actual energy sequence; The parameter prediction module obtains the optimal predicted ohmic internal resistance and the optimal predicted actual energy based on the current ohmic internal resistance and the current actual energy combined with the SRU prediction model; The SOE estimation module estimates the SOE of lithium batteries based on an adaptive dual extended Kalman filter combined with the optimal predicted ohmic internal resistance and the optimal predicted actual energy.
[0026] In a third aspect, a terminal is provided, including: processor, memory, wherein: The memory is used to store computer programs. The processor is used to call and run the computer program from the memory, so that the terminal executes the above-mentioned terminal method.
[0027] In a fourth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer executes the methods described in the above aspects.
[0028] The beneficial effect of the present invention lies in that the SRU-based lithium battery SOE estimation method, system, terminal and storage medium provided by the present invention first establish a fourth-order RC equivalent circuit model of the cascade utilization lithium battery to be estimated to obtain the ohmic internal resistance and actual energy, then use historical data to build an SRU prediction model, and then combine the current data to obtain the optimal prediction value, and finally estimate the lithium battery SOE based on the adaptive double extended Kalman filter and the optimal prediction value. This method comprehensively considers the battery characteristics and historical data, improves the accuracy and adaptability of the lithium battery ohmic internal resistance, actual energy and SOE estimation, and provides an effective means for the safe control of cascade utilization lithium batteries.
[0029] In addition, the invention has a reliable design principle, a simple structure and a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] In order to more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the description will be briefly introduced below. Obviously, the accompanying drawings in the following description 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.
[0031] Figure 1 It is a schematic flow chart of a method for estimating the SOE of a lithium battery based on SRU according to an embodiment of the present invention.
[0032] Figure 2 It is a fourth-order RC equivalent circuit model diagram of an embodiment of the present invention.
[0033] Figure 3 It is a structural diagram of an SRU algorithm according to an embodiment of the present invention.
[0034] Figure 4 The figure is a flow chart of SOE estimation based on SRU optimization according to an embodiment of the present invention.
[0035] Figure 5 It is a schematic block diagram of a lithium battery SOE estimation system based on SRU according to an embodiment of the present invention.
[0036] Figure 6 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0037] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.
[0039] The SRU-based lithium battery SOE estimation method provided in the embodiment of the present invention is executed by a computer device, and accordingly, the SRU-based lithium battery SOE estimation system runs in the computer device.
[0040] Figure 1 FIG. 1 is a schematic flow chart of a method for estimating the SOE of a lithium battery based on SRU according to an embodiment of the present invention. Figure 1 The execution subject may be a lithium battery SOE estimation system based on SRU. According to different requirements, the order of the steps in the flow chart may be changed, and some may be omitted.
[0041] like Figure 1 As shown, the method includes: S1, establishing a fourth-order RC equivalent circuit model of the second-life lithium battery to be estimated, and obtaining the ohmic internal resistance and actual energy of the lithium battery based on the fourth-order RC equivalent circuit model; The complex electrical characteristics of the battery, such as the influence of polarization impedance and other factors on battery performance, are fully considered. Compared with the simple circuit model, the fourth-order RC equivalent circuit model not only fully considers the influence of polarization impedance of the second-life lithium battery on the performance, but also can timely and accurately reflect the status information of the second-life lithium-ion battery. The model has high accuracy and is easier to implement.
[0042] S2, obtaining the historical ohmic internal resistance sequence and the historical actual energy sequence, and building the SRU prediction model based on the historical ohmic internal resistance sequence and the historical actual energy sequence; The RU model itself can process the time series information in the sequence data. By learning the rules in the historical data, it can better capture the trend of the battery's ohmic internal resistance and actual energy changing over time. This learning method based on historical data makes the model adaptable and generalizable, and can make a relatively accurate prediction of the future battery status, providing valuable reference information for the battery management system, which helps to take measures in advance to deal with possible changes in battery performance.
[0043] S3, based on the current ohmic internal resistance and the current actual energy combined with the SRU prediction model, obtain the optimal predicted ohmic internal resistance and the optimal predicted actual energy; By combining current data, sudden changes or abnormal conditions in the battery status can be captured in a timely manner, while the predictive ability of the SRU model can be used to make reasonable estimates of future status, thereby providing strong support for the safe use and effective management of batteries.
[0044] S4, based on the adaptive dual extended Kalman filter combined with the optimal predicted ohmic internal resistance and the optimal predicted actual energy, the lithium battery SOE is estimated.
[0045] Combining the optimal predicted ohmic internal resistance and actual energy, the accurate information obtained in the previous steps can be fully utilized, so that the SOE estimation can comprehensively consider the various characteristics and state changes of the battery, provide key state evaluation indicators for the safe and efficient operation of second-life lithium batteries, and help optimize battery use and management strategies.
[0046] Optionally, as an embodiment of the present invention, in step S1, the SOE of the second-life lithium battery is the ratio of the remaining energy of the lithium battery to the maximum available energy, which is calculated as follows:
[0047] in, and They are the SOE values of the discrete states of the second-life lithium battery at time k and k+1 respectively; It is to utilize the energy of lithium batteries in stages; and They are the current and working voltage of the lithium battery in discrete states at time k; Sampling period.
[0048] Optionally, as an embodiment of the present invention, refer to Figure 2 , the construction of the fourth-order RC equivalent circuit model specifically includes: The discrete state equation of the fourth-order RC equivalent model of the second-life lithium-ion battery is as follows:
[0049] The discrete observation equation of the fourth-order RC equivalent model of the cascade utilization of lithium-ion batteries is as follows:
[0050] make
[0051]
[0052]
[0053]
[0054] It is concluded that
[0055]
[0056] in, Indicates the open circuit voltage of the second-life lithium-ion battery; The working voltage of the lithium-ion battery for cascade utilization; is the ohmic internal resistance; , , , is the polarization internal resistance; , , , is the polarization capacitance; , , , Capacitor , , , The voltage across the terminals; , , , is the time constant; is the charge and discharge current; It is the SOE of the lithium-ion battery used in the discrete state k, k+1 time; It is to utilize the energy of lithium-ion batteries in a cascade manner; is the charge and discharge current at time k in the discrete state; is the working voltage of the lithium-ion battery in the discrete state at time k; is the sampling period; is the k, k+1 time in the discrete state Estimated voltage value of the branch; is the k, k+1 time in discrete state Estimated voltage value of the branch; is the k, k+1 time in discrete state Estimated voltage value of the branch; is the k, k+1 time in discrete state Estimated voltage value of the branch; is the independent system noise, is a four-dimensional vector; is the open circuit voltage corresponding to the SOE value of the lithium-ion battery in the discrete state at time k. is the system status information at time k, is the input variable at time k, is the state transfer matrix at time k, is the input gain matrix at time k, is system status information, is the output matrix, is the dynamic ohmic internal resistance at the kth moment, is the ohmic internal resistance, It is the system output information.
[0057] Optionally, as an embodiment of the present invention, the specific steps of step S2 include: The historical ohmic internal resistance series and the historical actual energy series are used as data sets, and the data sets are divided into training sets, validation sets, and test sets; The input vector of the model is the historical ohmic internal resistance and the historical actual energy and two input layer neuron numbers are set. The output vector of the model is the optimal predicted ohmic internal resistance and the optimal predicted actual energy and two output layer neuron numbers are set. Initialize the model’s weights and biases; The training set is input into the SRU prediction model, and the predicted output is calculated through forward propagation. The loss value between the actual output and the predicted output in the training set is calculated based on the loss function. The gradient of the loss function to the model weights and biases is calculated through back propagation, and the weights and biases of the model are updated using the gradient information. The preset number of training times per interval is used to calculate the evaluation index based on the validation set combined with the root mean square error function, and the model structure is adjusted based on the changing trend of the evaluation index; The trained SRU prediction model is finally evaluated based on the test set.
[0058] Optionally, as an embodiment of the present invention, refer to Figure 3 , in the forward propagation calculation, it specifically includes: Perform a linear transformation on the input vector, calculated as follows:
[0059] in, is a linear transformation of t time steps, is the weight matrix, is the input vector at time step t; The forget gate is calculated based on the input vector as follows:
[0060] in, is the forget gate output at time step t, is the Sigmoid function, is the weight matrix associated with the forget gate, is the bias vector of the forget gate; The internal state of the forget gate is calculated based on the forget gate output and the linear transformation, which is calculated as follows:
[0061] in, is the internal state of the forget gate at time step t, is the internal state of the forget gate at time step t-1, It is the operation between corresponding elements of the matrix; The reset gate is calculated based on the input vector as follows:
[0062] in, is the reset gate output at time step t, is the weight matrix associated with the reset gate, is the bias vector for the reset gate; The output vector is calculated based on the reset gate output and the internal state of the forget gate as follows:
[0063] in, is the output vector at time step t, is the activation function.
[0064] Optionally, as an embodiment of the present invention, updating the weights and biases of the model using gradient information specifically includes: Initialize the first-order moment estimate used to track the gradient and update the first-order moment estimate, calculated as follows:
[0065] Where m is the first-order moment estimate, is a hyperparameter between 0 and 1, is the gradient of the weight matrix or the gradient of the bias vector; Initialize the second-order moment estimate used to track the gradient and update the second-order moment estimate, calculated as follows:
[0066] in, is the second-order moment estimate, is another hyperparameter between 0 and 1; The bias correction is performed on the first-order moment estimate and the second-order moment estimate respectively: , ; The model parameters are updated based on the first-order moment estimate and the second-order moment estimate after bias correction under the same gradient, and the calculation is as follows:
[0067] Where A is the model parameter weight matrix or bias vector, yes, is the corrected first-order moment estimate, is the modified second-order moment estimate, This is a value that prevents the denominator from being zero.
[0068] Optionally, as an embodiment of the present invention, in step S3, after obtaining the optimal predicted ohmic internal resistance, the optimal predicted ohmic internal resistance is judged against the safety threshold. When the optimal predicted ohmic internal resistance is greater than the safety threshold, the difference between the optimal predicted ohmic internal resistance and the safety threshold is calculated, and an adjustment coefficient is determined according to the size of the difference, and the current charge and discharge current is adjusted based on the adjustment coefficient.
[0069] Optionally, as an embodiment of the present invention, refer to Figure 4 , step S4 specifically includes: After adding the ohmic internal resistance and actual energy, the state equation, observation equation and parameter equation formulas of the lithium battery SOE estimation system are:
[0070]
[0071]
[0072] in, is the system parameter variable at time k, ; is the system state variable at time k, is the system state variable at time k+1; is the input variable of the system, i.e., the lithium-ion battery current; is the system observation variable at time k+1, i.e., the operating voltage of the lithium-ion battery; is zero-mean Gaussian white noise; is the system parameter variable at time k+1, is the system parameter variable at time k, is the noise mean at time k, that is, zero-mean Gaussian white noise; is the noise associated with the parameter variable at time k, that is, zero-mean Gaussian white noise; is the system output variable at time k+1; The lithium battery SOE estimation based on adaptive dual extended Kalman filter combined with optimal predicted ohmic internal resistance and optimal predicted actual energy includes: S4-1, calculate the expected value of the initial system parameter variable and the initial error covariance matrix;
[0073]
[0074] S4-2, calculating the volume point product at the previous moment based on the error covariance matrix at the previous moment and the predicted value of the state variable at the previous moment, and obtaining the volume point at the current moment through the state transfer function;
[0075]
[0076]
[0077] in, , ; [1] represents 𝑛 as the set of points in 𝑢 space, that is:
[0078] S4-3, calculating the average value of all volume points at the current moment, obtaining the state variable prediction value at the current moment, and calculating the state prediction covariance matrix based on the volume points and the state variable prediction value at the current moment;
[0079]
[0080] S4-4, based on the state prediction covariance and the state prediction value at the current moment, the volume point product in the state update is calculated, and based on the observation function, the predicted value of the observation value is calculated for another volume point at the current moment; the predicted values of all the observation values are averaged to obtain the predicted value of the observed state; based on the predicted value of the observation value and the predicted value of the observed state, the observation value covariance matrix is calculated, and the cross covariance matrix between another volume point and the state prediction value and between the observation value prediction value and the observed state prediction value is calculated, and the Kalman gain is calculated based on the observation value covariance matrix and the cross covariance matrix;
[0081]
[0082]
[0083] Observation status prediction:
[0084] Kalman gain:
[0085]
[0086]
[0087] S4-5, using the Kalman gain and the difference between the observed value and the observed state prediction value, the state prediction value at the current moment is updated to obtain the optimal state estimation value, and the optimal SOE estimation is obtained based on the optimal state estimation value.
[0088] Optimal state estimation:
[0089] Optimal estimated covariance:
[0090] State error and observation error:
[0091]
[0092] Process noise covariance:
[0093]
[0094] Observation noise covariance:
[0095]
[0096] in , b is the forgetting factor of x, 0 <b<1; ; and They are the estimate and optimal estimate of the state variables at time k; and are the estimated value and actual observed value at time k; and They are the estimate and optimal estimate of the error covariance at time k respectively.
[0097] Calculated from the above formula:
[0098] The best estimate of SOE is .
[0099] In some embodiments, the SRU-based lithium battery SOE estimation system may include a plurality of functional modules composed of computer program segments. The computer program of each program segment in the SRU-based lithium battery SOE estimation system may be stored in a memory of a computer device and executed by at least one processor to perform (see Figure 1 Description) Function to estimate SOE of lithium battery based on SRU.
[0100] In this embodiment, the SRU-based lithium battery SOE estimation system can be divided into multiple functional modules according to the functions it performs, such as Figure 5 As shown. The functional modules of the system may include: parameter acquisition module, SRU prediction model building module, parameter prediction module, SOE estimation module. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in the memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments. The system includes: A parameter acquisition module is used to establish a fourth-order RC equivalent circuit model of the second-life lithium battery to be estimated, and to obtain the ohmic internal resistance and actual energy of the lithium battery based on the fourth-order RC equivalent circuit model; An SRU prediction model building module obtains the historical ohmic internal resistance sequence and the historical actual energy sequence, and builds an SRU prediction model based on the historical ohmic internal resistance sequence and the historical actual energy sequence; The parameter prediction module obtains the optimal predicted ohmic internal resistance and the optimal predicted actual energy based on the current ohmic internal resistance and the current actual energy combined with the SRU prediction model; The SOE estimation module estimates the SOE of lithium batteries based on an adaptive dual extended Kalman filter combined with the optimal predicted ohmic internal resistance and the optimal predicted actual energy.
[0101] By establishing a fourth-order RC equivalent circuit model to obtain initial parameters, using historical data to build an SRU model, combining current data to predict parameters, and then performing SOE estimation based on an adaptive dual extended Kalman filter and predicted parameters, this architecture comprehensively considers battery characteristics and data information, improves the accuracy and reliability of lithium battery ohmic internal resistance, actual energy and SOE estimation, and provides strong support for the safe control and effective management of lithium battery recycling.
[0102] Figure 6A schematic diagram of the structure of a terminal provided in an embodiment of the present invention, wherein the terminal can be used to execute the method for estimating the SOE of a lithium battery based on SRU provided in an embodiment of the present invention.
[0103] The terminal may include: a processor, a memory and a communication unit. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention. It may be a bus structure or a star structure, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.
[0104] The memory can be used to store the execution instructions of the processor, and the memory can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory are executed by the processor, the terminal is enabled to execute some or all of the steps in the following method embodiments.
[0105] The processor is the control center of the storage terminal, which uses various interfaces and lines to connect various parts of the entire electronic terminal, and executes various functions and / or processes data of the electronic terminal by running or executing software programs and / or modules stored in the memory, and calling data stored in the memory. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor can only include a central processing unit (CPU). In the embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.
[0106] The communication unit is used to establish a communication channel so that the storage terminal can communicate with other terminals, receive user data sent by other terminals or send user data to other terminals.
[0107] The present invention also provides a computer-readable storage medium, wherein the computer storage medium may store a program, and when the program is executed, the program may include some or all of the steps in each embodiment provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).
[0108] Therefore, the technical effects that can be achieved by this embodiment can be found in the description above and will not be repeated here.
[0109] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes, including several instructions for enabling a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention.
[0110] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the terminal embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.
[0111] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or modules, which can be electrical, mechanical or other forms.
[0112] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0113] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.
[0114] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention.
Claims
1. A lithium battery SOE estimation method based on SRU, characterized in that: The following steps are involved: S1, establishing a fourth-order RC equivalent circuit model of the second-life lithium battery to be estimated, and obtaining the ohmic internal resistance and actual energy of the lithium battery based on the fourth-order RC equivalent circuit model; S2, obtaining the historical ohmic internal resistance sequence and the historical actual energy sequence, and building the SRU prediction model based on the historical ohmic internal resistance sequence and the historical actual energy sequence; S3, based on the current ohmic internal resistance and the current actual energy combined with the SRU prediction model, obtain the optimal predicted ohmic internal resistance and the optimal predicted actual energy; S4, based on the adaptive dual extended Kalman filter combined with the optimal predicted ohmic internal resistance and the optimal predicted actual energy, the lithium battery SOE is estimated.
2. The SRU-based lithium battery SOE estimation method according to claim 1, characterized in that: The specific steps of step S2 include: The historical ohmic internal resistance series and the historical actual energy series are used as data sets, and the data sets are divided into training sets, validation sets, and test sets; The input vector of the model is the historical ohmic internal resistance and the historical actual energy and two input layer neuron numbers are set. The output vector of the model is the optimal predicted ohmic internal resistance and the optimal predicted actual energy and two output layer neuron numbers are set. Initialize the model’s weights and biases; The training set is input into the SRU prediction model, and the predicted output is calculated through forward propagation. The loss value between the actual output and the predicted output in the training set is calculated based on the loss function. The gradient of the loss function to the model weights and biases is calculated through back propagation, and the weights and biases of the model are updated using the gradient information. The preset number of training times per interval is used to calculate the evaluation index based on the validation set combined with the root mean square error function, and the model structure is adjusted based on the changing trend of the evaluation index; The trained SRU prediction model is finally evaluated based on the test set.
3. The SRU-based lithium battery SOE estimation method according to claim 2, characterized in that: The forward propagation calculation specifically includes: Perform a linear transformation on the input vector, calculated as follows: in, is a linear transformation of t time steps, is the weight matrix, is the input vector at time step t; The forget gate is calculated based on the input vector as follows: in, is the forget gate output at time step t, is the Sigmoid function, is the weight matrix associated with the forget gate, is the bias vector of the forget gate; The internal state of the forget gate is calculated based on the forget gate output and the linear transformation, which is calculated as follows: in, is the internal state of the forget gate at time step t, is the internal state of the forget gate at time step t-1, It is the operation between corresponding elements of the matrix; The reset gate is calculated based on the input vector as follows: in, is the reset gate output at time step t, is the weight matrix associated with the reset gate, is the bias vector for the reset gate; The output vector is calculated based on the reset gate output and the internal state of the forget gate as follows: in, is the output vector at time step t, is the activation function.
4. The SRU-based lithium battery SOE estimation method according to claim 1, characterized in that: Using gradient information to update the model's weights and biases specifically includes: Initialize the first-order moment estimate used to track the gradient and update the first-order moment estimate, calculated as follows: Where m is the first-order moment estimate, is a hyperparameter between 0 and 1, is the gradient of the weight matrix or the gradient of the bias vector; Initialize the second-order moment estimate used to track the gradient and update the second-order moment estimate, calculated as follows: in, is the second-order moment estimate, is another hyperparameter between 0 and 1; The bias correction is performed on the first-order moment estimate and the second-order moment estimate respectively; The model parameters are updated based on the first-order moment estimate and the second-order moment estimate after bias correction under the same gradient, and the calculation is as follows: Where A is the model parameter weight matrix or bias vector, yes, is the corrected first-order moment estimate, is the modified second-order moment estimate, This is a value that prevents the denominator from being zero.
5. The SRU-based lithium battery SOE estimation method according to claim 1, characterized in that: Step S4 specifically includes: After adding the ohmic internal resistance and actual energy, the state equation, observation equation and parameter equation formulas of the lithium battery SOE estimation system are: in, is the system state variable at time k, is the system state variable at time k+1; is the input variable of the system, i.e., the lithium-ion battery current; is the system observation variable at time k+1, i.e., the operating voltage of the lithium-ion battery; is zero-mean Gaussian white noise; is the system parameter variable at time k+1, is the system parameter variable at time k, ; is the noise mean at time k, that is, zero-mean Gaussian white noise; The lithium battery SOE estimation based on adaptive dual extended Kalman filter combined with optimal predicted ohmic internal resistance and optimal predicted actual energy includes: S4-1, calculate the expected value of the initial system parameter variable and the initial error covariance matrix; S4-2, calculating the volume point product at the previous moment based on the error covariance matrix at the previous moment and the predicted value of the state variable at the previous moment, and obtaining the volume point at the current moment through the state transfer function; S4-3, calculating the average value of all volume points at the current moment, obtaining the state variable prediction value at the current moment, and calculating the state prediction covariance matrix based on the volume points and the state variable prediction value at the current moment; S4-4, based on the state prediction covariance and the state prediction value at the current moment, another volume point product in the state update is calculated, and based on the observation function, another volume point at the current moment is calculated to obtain the predicted value of the observation value; the predicted values of all the observation values are averaged to obtain the predicted value of the observed state; based on the predicted value of the observation value and the predicted value of the observed state, the observation value covariance matrix is calculated, the cross covariance matrix between another volume point and the state prediction value and between the observation value prediction value and the observed state prediction value is calculated, and the Kalman gain is calculated based on the observation value covariance matrix and the cross covariance matrix; S4-5, using the Kalman gain and the difference between the observed value and the observed state prediction value, the state prediction value at the current moment is updated to obtain the optimal state estimation value, and the optimal SOE estimation is obtained based on the optimal state estimation value.
6. The SRU-based lithium battery SOE estimation method according to claim 1, characterized in that: In step S3, after obtaining the optimal predicted ohmic internal resistance, the optimal predicted ohmic internal resistance is judged against the safety threshold. When the optimal predicted ohmic internal resistance is greater than the safety threshold, the difference between the optimal predicted ohmic internal resistance and the safety threshold is calculated, and the adjustment coefficient is determined according to the size of the difference, and the current charge and discharge current is adjusted based on the adjustment coefficient.
7. The SRU-based lithium battery SOE estimation method according to claim 1, characterized in that: In step S1, the SOE of the second-life lithium battery is the ratio of the remaining energy of the lithium battery to the maximum available energy, which is calculated as follows: in, and They are the SOE values of the discrete states of the second-life lithium battery at time k and k+1 respectively; It is to utilize the energy of lithium batteries in stages; and They are the current and working voltage of the lithium battery in discrete states at time k; Sampling period.
8. A lithium battery SOE estimation system based on SRU, characterized in that: When the system is implemented, the SRU-based lithium battery SOE estimation method according to any one of claims 1 to 7 is executed, and the system includes: A parameter acquisition module is used to establish a fourth-order RC equivalent circuit model of the second-life lithium battery to be estimated, and to obtain the ohmic internal resistance and actual energy of the lithium battery based on the fourth-order RC equivalent circuit model; An SRU prediction model building module obtains the historical ohmic internal resistance sequence and the historical actual energy sequence, and builds an SRU prediction model based on the historical ohmic internal resistance sequence and the historical actual energy sequence; The parameter prediction module obtains the optimal predicted ohmic internal resistance and the optimal predicted actual energy based on the current ohmic internal resistance and the current actual energy combined with the SRU prediction model; The SOE estimation module estimates the SOE of lithium batteries based on an adaptive dual extended Kalman filter combined with the optimal predicted ohmic internal resistance and the optimal predicted actual energy.
9. A terminal, characterized in that: include: A memory for storing a lithium battery SOE estimation program based on the SRU; A processor is used to implement the steps of the SRU-based lithium battery SOE estimation method as described in any one of claims 1 to 7 when executing the SRU-based lithium battery SOE estimation program.
10. A computer-readable storage medium, characterized in that: The readable storage medium stores a SRU-based lithium battery SOE estimation program, and when the SRU-based lithium battery SOE estimation program is executed by a processor, the steps of the SRU-based lithium battery SOE estimation method as described in any one of claims 1-7 are implemented.