Non-invasive lithium ion battery charge state detection method, equipment, medium and product
By acquiring the time domain data of the second-order harmonic signal of the guided wave on the surface of the lithium-ion battery and constructing a quantitative SOC mapping model, combined with in-situ characterization technology, the problems of insufficient interpretability and reliability of existing detection methods are solved, and high-precision non-destructive detection of the state of charge of the lithium-ion battery and analysis of the microscopic properties of the material are achieved.
Patent Information
- Application Number
- CN202510783285.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-12
- Publication Date
- 2025-09-05
AI Technical Summary
Existing lithium-ion battery state of charge detection methods have difficulty achieving non-invasive, real-time detection of multi-scale dynamic responses, and lack in-depth analysis of the relationship between the microscopic properties of internal battery materials and guided wave propagation characteristics, resulting in insufficient interpretability and reliability of detection results.
By acquiring the time domain data of the second-order harmonic wave signal of the guided wave, converting it into frequency domain data, extracting characteristic parameters, and constructing a SOC quantitative mapping model, combined with in-situ characterization methods to obtain the microscopic properties of the material, and using the CNN-LSTM model to optimize the model parameters, non-destructive detection of the state of charge of lithium-ion batteries can be achieved.
The accuracy and interpretability of lithium-ion battery state of charge detection are improved, non-destructive detection of the microscopic properties of lithium-ion battery materials is achieved, and the reliability of the detection results is enhanced.
Smart Images
Figure CN120595128A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of non-destructive testing of lithium-ion batteries, and in particular to a non-invasive method, device, medium, and product for detecting the state of charge of a lithium-ion battery. Background Art
[0002] Lithium-ion batteries are the core energy storage units of new energy vehicles, energy storage systems, and smart grids. Accurate detection of their SOC (State of Charge) is a key technology to ensure efficient operation, extend battery life, and improve safety. However, existing SOC detection methods (such as open circuit voltage method, ampere-hour integration method, and equivalent circuit model) are difficult to achieve non-invasive, multi-scale dynamic response real-time detection. In addition, existing detection technologies lack cross-scale correlation analysis of the microscopic properties of internal battery materials (such as electrode phase transitions, interface side reactions, and crystal structure degradation), resulting in insufficient reliability and interpretability of SOC estimation models.
[0003] In recent years, ultrasonic guided wave detection technology has gradually become a research hotspot in the field of battery state of charge monitoring due to its non-destructive, high sensitivity and real-time advantages. In particular, guided wave technology can capture the dynamic changes of material properties and electrochemical states inside the battery by exciting and receiving acoustic wave signals. However, the existing guided wave detection technology still faces the following key problems in application: (1) Complex calculation of dispersion curves. When guided waves propagate in the multi-layer structure of the battery, a dispersion effect will occur. A complex theoretical model needs to be constructed to calculate the dispersion curve to determine the propagation mode of the guided waves and the selection of frequency. However, the multi-layer and porous characteristics of the internal materials of the battery increase the difficulty of modeling, making it difficult to balance calculation accuracy and practicality. (2) The contradiction between frequency selection and signal attenuation. Although high-frequency guided waves can provide richer material microscopic information, they are prone to severe attenuation when propagating in the multi-layer structure of the battery, making it difficult to effectively extract useful signals. Although low-frequency guided waves have less attenuation, they are not sensitive enough to changes in microscopic properties and cannot meet the needs of high-precision detection. (3) Insufficient interpretability. Most existing technologies directly associate waveguide characteristic parameters (such as sound velocity, amplitude, and attenuation coefficient) with SOC, lacking in-depth analysis of the interaction mechanism between the microscopic properties of internal battery materials and waveguide propagation characteristics, resulting in limited interpretability and reliability of detection results. Summary of the Invention
[0004] The purpose of this application is to provide a non-invasive lithium-ion battery state of charge detection method, equipment, medium and product, which can achieve non-destructive detection of the charge state of the lithium-ion battery and obtain the material microscopic properties of the lithium-ion battery under the charge state, thereby improving the accuracy of lithium-ion battery charge state detection and improving the interpretability and reliability of the detection results.
[0005] To achieve the above objectives, this application provides the following solutions:
[0006] In a first aspect, the present application provides a non-invasive lithium-ion battery state of charge detection method, comprising:
[0007] Acquiring real-time time domain data of a second-order harmonic wave guided wave signal generated by applying a sinusoidal signal to the surface of a lithium-ion battery;
[0008] Converting the time domain data into frequency domain data;
[0009] Extracting characteristic parameters based on the time domain data and the frequency domain data;
[0010] Construct a SOC quantitative mapping model;
[0011] The characteristic parameters are input into the SOC quantitative mapping model to obtain the state of charge of the lithium-ion battery and the microscopic properties of the material under the state of charge.
[0012] Optionally, the process of constructing the SOC quantitative mapping model includes:
[0013] Obtain historical characteristic parameters of lithium-ion batteries at different states of charge;
[0014] In-situ characterization is used to obtain the material microscopic properties of lithium-ion batteries at different states of charge, and to correspond them one-to-one with the historical characteristic parameters of lithium-ion batteries at different states of charge;
[0015] An SOC quantitative mapping model is constructed based on historical characteristic parameters of the lithium-ion battery at different states of charge and material microscopic properties corresponding to the historical characteristic parameters.
[0016] Optionally, the time domain data of the waveguide second-order harmonic signal is collected by setting the sampling frequency.
[0017] Optionally, the devices for implementing the in-situ characterization method include: an atomic force microscope, a Raman spectrometer and a scanning electron microscope.
[0018] Optionally, in-situ characterization techniques are used to obtain the material microscopic properties of lithium-ion batteries at different states of charge, including:
[0019] Lithium-ion batteries at different states of charge are disassembled, and the in-situ characterization method is used to perform electrode testing on the disassembled lithium-ion batteries to obtain the microscopic properties of the material.
[0020] Optionally, constructing an SOC quantitative mapping model based on historical characteristic parameters of the lithium-ion battery at different states of charge and material microscopic properties corresponding to the historical characteristic parameters includes:
[0021] Construct an initial SOC quantitative mapping model based on the CNN-LSTM model;
[0022] Inputting the historical characteristic parameters into the initial SOC quantitative mapping model to obtain an initial predicted state of charge and an initial predicted material microstate;
[0023] The parameters of the initial SOC quantitative mapping model are optimized based on the initial predicted state of charge, the state of charge of the lithium-ion battery corresponding to the historical characteristic parameters, the initial predicted material microstate, and the material microproperties corresponding to the historical characteristic parameters, until the output of the initial SOC quantitative mapping model meets the set requirements, and the optimized initial SOC quantitative mapping model is used as the SOC quantitative mapping model.
[0024] Optionally, the mean absolute error and the mean variance error are used as the loss function of the initial SOC quantitative mapping model, and the Huber loss function is used to balance the mean absolute error and the mean variance error; the loss function of the initial SOC quantitative mapping model is used to determine whether the output of the initial SOC quantitative mapping model meets the set requirements.
[0025] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any one of the above-described non-invasive lithium-ion battery state of charge detection methods.
[0026] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the above-mentioned non-invasive lithium-ion battery state of charge detection methods.
[0027] In a fourth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned non-invasive lithium-ion battery state of charge detection methods.
[0028] According to the specific embodiments provided in this application, this application has the following technical effects:
[0029] This application provides a non-invasive lithium-ion battery state of charge detection method, equipment, medium and product. By acquiring the time domain data of the guided wave second-order harmonic signal generated by applying a sinusoidal signal to the surface of the lithium-ion battery, converting the time domain data into frequency domain data, and extracting the characteristic parameters of the time domain data and frequency domain data. The characteristic parameters are input into the SOC quantitative mapping model to obtain the state of charge of the lithium-ion battery and the microscopic properties of the material under this state of charge, thereby achieving non-destructive detection of the lithium-ion battery, improving the accuracy of lithium-ion battery state of charge detection, and improving the interpretability and reliability of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0031] Figure 1 This is a flow chart of a non-invasive lithium-ion battery state of charge detection method according to an embodiment of the present application;
[0032] Figure 2 A schematic diagram of a test system structure for a non-invasive lithium-ion battery state of charge detection method provided in one embodiment of the present application;
[0033] Figure 3 A schematic diagram of a guided wave second-order harmonic signal at different charge states during a cycle provided by an embodiment of the present application;
[0034] Figure 4 The scanning electron microscopy results of lithium-ion battery positive electrode materials at different states of charge provided in one embodiment of the present application;
[0035] Figure 5 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION
[0036] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.
[0037] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0038] In an exemplary embodiment, Figure 1 As shown, a non-invasive lithium-ion battery state of charge detection method is provided. The method is executed by a computer device, and can be specifically executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is used to perform non-destructive detection of the state of charge of a lithium-ion battery as an example for description, including:
[0039] Step 100: Acquire time domain data of a second-order harmonic wave guided wave signal in real time. The second-order harmonic wave guided wave signal is generated after a sinusoidal signal is applied to the surface of the lithium-ion battery.
[0040] Step 200: Convert the time domain data into frequency domain data.
[0041] Step 300: extracting characteristic parameters based on time domain data and frequency domain data.
[0042] Step 400: constructing a SOC quantitative mapping model.
[0043] Step 500: Input the characteristic parameters into the SOC quantitative mapping model to obtain the state of charge of the lithium-ion battery and the microscopic properties of the material under the state of charge.
[0044] By implementing the above steps 100 to 500, the present application can achieve non-destructive detection of the state of charge of a lithium-ion battery and obtain the material microscopic properties of the lithium-ion battery under the state of charge, thereby improving the accuracy of the lithium-ion battery state of charge detection and improving the interpretability and reliability of the detection results.
[0045] As an optional implementation, in order to improve the accuracy of lithium-ion battery state of charge detection and the interpretability and reliability of the detection results, step 400 includes:
[0046] Step 421 : Obtain historical characteristic parameters of the lithium-ion battery at different states of charge.
[0047] Step 422 : Using in-situ characterization methods, obtain the material microscopic properties of the lithium-ion battery at different states of charge, and make a one-to-one correspondence with the historical characteristic parameters of the lithium-ion battery at different states of charge.
[0048] Step 423 : constructing an SOC quantitative mapping model based on historical characteristic parameters of the lithium-ion battery at different states of charge and the material microscopic properties corresponding to the historical characteristic parameters.
[0049] The devices used to implement the in-situ characterization method may include: an atomic force microscope, a Raman spectrometer, and a scanning electron microscope. The process of step 422 may include: disassembling lithium-ion batteries at different states of charge, and performing electrode testing on the disassembled lithium-ion batteries using the in-situ characterization method to obtain the microscopic properties of the material.
[0050] Step 423 may include:
[0051] Step 4231: construct an initial SOC quantitative mapping model based on the CNN-LSTM model.
[0052] In step 4232, the historical characteristic parameters are input into the initial SOC quantitative mapping model to obtain the initial predicted state of charge and the initial predicted material microstate.
[0053] Step 4234: Optimize the parameters of the initial SOC quantitative mapping model based on the initial predicted state of charge, the lithium-ion battery state of charge corresponding to the historical characteristic parameters, the initial predicted material microstate, and the material microproperties corresponding to the historical characteristic parameters until the output of the initial SOC quantitative mapping model meets the set requirements. The optimized initial SOC quantitative mapping model is used as the SOC quantitative mapping model. The mean absolute error and mean variance error can be used as the loss function of the initial SOC quantitative mapping model, and the Huber loss function is used to balance the mean absolute error and the mean variance error. The loss function of the initial SOC quantitative mapping model is used to determine whether the output of the initial SOC quantitative mapping model meets the set requirements.
[0054] For example, a CNN-LSTM model with an attention mechanism was constructed as the initial SOC quantitative mapping model to capture the evolution of the second-order harmonic waveguide signal. MAE (Mean Absolute Error) and MSE (Mean Squared Error) were used as the model's loss function. The Huber loss function was used to balance MAE and MSE, and an L2 regularization term was added. δ in the Huber loss function was set to 1.0. When the prediction deviation was less than 1.0, the squared error (MSE) was used to maintain sensitivity to small errors. When the prediction deviation was greater than 1.0, the linear error (MAE) was used to reduce the impact of outliers on the model. The AdamW optimizer was used as the optimizer in the early stages of model optimization. In the later stages of rapid model convergence, the Lion optimizer was used for fine-tuning the model, and a cosine annealing learning rate schedule was introduced. The AdamW optimizer's learning rate was Ir = 3e-4, β1 = 0.9, and β2 = 0.999. The Lion optimizer has a learning rate of Ir = 1e-5 and a momentum of 0.9 (increase the momentum to 0.95 if oscillations occur during the optimization process). The cosine annealing learning rate schedule has a period of T = 50.
[0055] The acquired historical characteristic parameters and the corresponding material microscopic properties were divided into a sample training set (70%), a sample validation set (15%), and a sample test set (15%). A nested 5-fold cross-validation method was used, with standardized mean absolute error and prediction interval coverage probability used for evaluation. Gradient-weighted Class Activation Mapping (Grad-CAM) was used to visualize key signal intervals, and the Shapley value was used to quantify the contribution of each feature. The optimized initial SOC quantitative mapping model was retained as the SOC quantitative mapping model until the output of the initial SOC quantitative mapping model met the specified requirements.
[0056] The CNN-LSTM model with an attention mechanism can include a CNN feature encoder, a channel attention module, a bidirectional LSTM layer, a temporal attention layer, and a multi-task output layer. The CNN feature encoder consists of four layers of 1D convolutional filters with 64 / 128 / 256 / 512 kernel sizes of 5 / 3 / 3 / 3, respectively, and a Reluctant Unified Unit (ReLU) activation function. The channel attention module enhances key frequency band features. The bidirectional LSTM layer (LSTM time series modeling, 256 units) captures the dynamics of the waveguide second-order harmonic signal evolution. The temporal attention layer calculates the weights for each time step. The multi-task output layer connects branches in parallel to predict the state of charge and material microscopic properties.
[0057] In an exemplary embodiment, the Figure 2The test system shown includes an oscilloscope, a signal excitation receiving device (RAM-5000SNAP can be selected), a soft-pack lithium-ion battery, a low-frequency piezoelectric sensor, a high-frequency piezoelectric sensor, a charge and discharge device (Xinwei CT-4008 charge and discharge instrument can be selected), a host computer, a high-energy amplifier module, and an attenuator. In combination with the steps of the non-invasive lithium-ion battery state of charge detection method provided in the above embodiment, a charge and discharge cycle test is performed on the lithium-ion battery, and the guided wave second-order harmonic signal of the lithium-ion battery at different states of charge is analyzed and the material microscopic properties of the lithium-ion battery at this state of charge are analyzed. The signal excitation receiving device is combined with a 250kHz low-frequency piezoelectric sensor to excite a 12-cycle Hanning window modulated sinusoidal signal on the surface of the lithium-ion battery. A 500kHz high-frequency piezoelectric sensor is used to receive the guided wave second-order harmonic signal generated by the lithium-ion battery at different states of charge during the propagation process. At this time, the attenuation on the attenuator is 1 to 6dB, and the signal gain is 80 to 100dB. The lithium-ion battery was then subjected to a charge-discharge cycle using a charging and discharging device. The selected cycle steps were constant current and constant voltage charging (with an upper voltage of 4.35V and a current of 4A) and constant current discharge (with a current of 4A and a cutoff current of 0.05A). The second-order harmonic signal received by the high-frequency piezoelectric sensor was displayed on an oscilloscope. Given the strong directivity of second-order harmonics, the oscilloscope's sampling frequency was adjusted to 250MS / s. The time-domain waveform data of the guided wave second-order harmonic signal displayed on the oscilloscope was then collected and processed via a host computer connection.
[0058] Specifically include:
[0059] S1: A low-frequency piezoelectric sensor is used to apply a Hanning window-modulated sinusoidal signal to the surface of the lithium-ion battery. The sinusoidal signal is generated by a signal excitation and receiving device connected to the low-frequency sensor. After the sinusoidal signal is applied to the low-frequency sensor, it propagates through the lithium-ion battery. The high-frequency sensor receives the guided second-order harmonic signal, and the lithium-ion battery is subjected to a charge-discharge cycle test using a charge-discharge device. The center frequency range of the receiving probe of the signal excitation and receiving device can be 500kHz to 650kHz, and the frequency range of the excitation signal of the signal excitation and receiving device can be 250kHz to 325kHz. The number of cycles of the sinusoidal signal modulated by the excitation signal of the signal excitation and receiving device is preferably 10 to 15, more preferably 10 to 12; the signal gain is preferably 60dB to 100dB, more preferably 90dB to 100dB.
[0060] S2: Receive the guided wave second-order harmonic signal via a high-frequency piezoelectric sensor. Simultaneously, collect the time-domain waveform data of the guided wave second-order harmonic signal, i.e., the time-domain data of the guided wave second-order harmonic signal. This data is then transmitted to the signal excitation and receiving device for subsequent processing by the host computer. (The frequency range of the guided wave second-order harmonic signal is 500kHz to 650kHz). The meaning of the guided wave second-order harmonic signal is that when the guided wave propagates through the material, the material's nonlinear elastic response (caused by lattice distortion, dislocation motion, etc.) causes changes in the waveform, triggering the generation of second-order harmonics. The frequency of the guided wave is twice that of the excited linear acoustic wave, and its nonlinear coefficient can effectively reflect changes in the material's microstructure. The spacing between the low-frequency piezoelectric sensor and the high-frequency piezoelectric sensor can be 25 to 35 times the thickness of the lithium-ion battery, preferably 30 to 32 times. The time-domain data of the guided wave second-order harmonic signal is collected using a set sampling frequency, which can be 125MS / s to 1GS / s, preferably 250MS / s to 625MS / s. During the charge and discharge cycle, according to the time domain signal of the lithium-ion battery at SOC = 100% and SOC = 0%, the mapping relationship between the state of charge and the transit time and amplitude at different states of charge can be obtained, such as Figure 3 The normalized amplitude and transit time at the peak of the time domain signal of the second-order harmonic of the guided wave under different charge states are as follows: Figure 3 As shown in (b).
[0061] S3, uses Fourier transform method to convert time domain data into frequency domain data.
[0062] S4. Extract characteristic parameters of the guided wave second-order harmonic signal based on the time-domain and frequency-domain data. These characteristic parameters may include amplitude and transit time in the time domain, and amplitude and frequency offset in the frequency domain. These characteristic parameters are then correlated with the state of charge (SOC) at the time of receiving the guided wave second-order harmonic.
[0063] S5, combined with devices such as atomic force microscope (AFM), Raman spectrometer (Raman Spectrometer) and scanning electron microscope (SEM), the soft-pack lithium-ion batteries at different states of charge are disassembled, and the electrode test is carried out by in-situ characterization means, and the surface morphology and crystal structure evolution of the electrode materials of lithium-ion batteries at different states of charge are analyzed to obtain the material microscopic characteristics of lithium-ion batteries at different states of charge. In-situ characterization refers to the technology of real-time and dynamic monitoring of the evolution of the physical and chemical properties of the material under the actual working state of the material (such as charging and discharging, temperature change, stress loading, etc.). In this application, the guided wave second-order harmonic signal is used to realize non-destructive detection of the battery state of charge. Through the in-situ detection method, the battery is disassembled to explore the electrode morphology, and the non-destructive detection is combined with the in-situ characterization to verify the reliability and accuracy of the signal evolution law of the non-destructive detection. For the elastic modulus, the Box-Cox transformation is used to eliminate the dimensional difference.
[0064] AFM: Quantitatively characterizes the surface roughness, modulus, and ion diffusion dynamics of electrode materials through probe-surface interaction forces. The electrode is integrated into an AFM electrochemical cell. The surface topography is scanned during charge and discharge (tapping mode, resolution <1nm). Roughness parameters (Ra, Rq) and Young's modulus (via force-displacement curves) are calculated. Results include changes in electrode surface roughness, elastic modulus, mechanical property degradation, and lithium-ion diffusion pathways.
[0065] Raman spectroscopy: Analyzes the local structure of materials (such as bond lengths and coordination environments) through changes in molecular vibration modes. Analysis process: Use a transparent electrochemical cell (such as a quartz window); collect Raman spectra during charge and discharge (laser wavelength 532nm, power <1mW); fit characteristic peak positions (such as the A1g mode of NMC ≈ 590cm -1 The results obtained include: local structural distortion of the battery, phase transition characteristics, and SEI composition evolution.
[0066] SEM: Use scanning electron microscopy to observe the surface morphology, crack propagation and interface reaction of electrode materials in real time. Analysis process: Construct a micro in-situ battery (such as solid electrolyte packaging); take high-resolution images (resolution <1nm) during the charge and discharge process; combine EDS / EELS to analyze component distribution (such as Li and O element migration). The results obtained include: electrode surface morphology evolution: such as particle cracking, SEI film growth (thickness ≈10~50nm), crack propagation rate and interface side reactions. Figure 4 As shown, Figure 4Part a is the scanning result of the battery positive electrode material at 100% state of charge. Figure 4 Part b is the scanning result of the battery positive electrode material at 80% state of charge. Figure 4 Part c is the scanning result of the battery positive electrode material at 60% state of charge. Figure 4 Part d is the scanning result of the battery positive electrode material at 40% state of charge. Figure 4 Part e is the scanning result of the battery positive electrode material at 20% state of charge. Figure 4 Part f is the scanning result of the battery positive electrode material at 0% charge state. As the charge state increases, lithium ions are continuously released from the positive electrode material, the particle spacing on the surface of the positive electrode material gradually decreases, and the structural uniformity gradually improves.
[0067] S6, construct an initial SOC quantitative mapping model based on the CNN-LSTM model. Input the characteristic parameters obtained in step S4 into the initial SOC quantitative mapping model to obtain the initial predicted state of charge and the initial predicted material microstate. Optimize the parameters of the initial SOC quantitative mapping model based on the initial predicted state of charge, the lithium-ion battery state of charge corresponding to the historical characteristic parameters, the initial predicted material microstate and the material microproperties corresponding to the historical characteristic parameters, until the output of the initial SOC quantitative mapping model meets the set requirements, and use the optimized initial SOC quantitative mapping model as the SOC quantitative mapping model. This model can reflect the intrinsic correlation between the characteristic parameters of the second-order harmonic signal of the guided wave and the state of charge of the lithium-ion battery and the microscopic properties of the material (such as elastic modulus, surface roughness, and Raman peak).
[0068] In the embodiment provided in this application, the time domain data of the second-order harmonic signal of the waveguide is preprocessed before extracting the characteristic parameters, including: using a 4th-order Butterworth bandpass filter (50-300kHz) for filtering to eliminate high-frequency electromagnetic interference and low-frequency mechanical vibration noise. Then, timing alignment is performed, and the dynamic time warping (DTW) algorithm is used to compensate for the time shift deviation of the sensor signal caused by battery expansion during charging and discharging. The processed signal data is subjected to Min-Max normalization (scaling to the [-1,1] interval) processing.
[0069] This application uses a low-frequency-high-frequency piezoelectric sensor collaborative architecture to achieve rapid, non-destructive detection of the state of charge of lithium-ion batteries, avoiding the reliability risks of traditional destructive detection methods and the difficulty of frequency selection of waveguide detection technology. Combined with in-situ characterization technology, the dynamic coupling mechanism of waveguide characteristic parameters and the microevolution of electrode materials is revealed, and the interpretability of the characteristic parameters of the waveguide second-order harmonic signal is improved. This application is suitable for online health management of new energy vehicle power batteries, lithium-ion battery packs in energy storage power stations, and batteries for consumer electronic devices, especially in high dynamic loads (such as fast charging and discharging), extreme temperature environments, and long-cycle aging scenarios.
[0070] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data related to a non-invasive lithium-ion battery state of charge detection method. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a non-invasive lithium-ion battery state of charge detection method is implemented.
[0071] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.
[0072] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.
[0073] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0074] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0075] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0076] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.
[0077] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0078] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A non-invasive lithium-ion battery state of charge detection method, characterized in that: include: Acquire the time domain data of the guided wave second-order harmonic signal in real time; The guided wave second-order harmonic signal is generated after a sinusoidal signal is applied to the surface of the lithium-ion battery; Converting the time domain data into frequency domain data; Extracting characteristic parameters based on the time domain data and the frequency domain data; Construct a SOC quantitative mapping model; The characteristic parameters are input into the SOC quantitative mapping model to obtain the state of charge of the lithium-ion battery and the microscopic properties of the material under the state of charge.
2. The non-invasive lithium-ion battery state of charge detection method according to claim 1, characterized in that: The process of constructing the SOC quantitative mapping model includes: Obtain historical characteristic parameters of lithium-ion batteries at different states of charge; In-situ characterization is used to obtain the material microscopic properties of lithium-ion batteries at different states of charge, and to correspond them one-to-one with the historical characteristic parameters of lithium-ion batteries at different states of charge; An SOC quantitative mapping model is constructed based on historical characteristic parameters of the lithium-ion battery at different states of charge and material microscopic properties corresponding to the historical characteristic parameters.
3. The non-invasive lithium-ion battery state of charge detection method according to claim 1, wherein: The time domain data of the waveguide second-order harmonic signal is collected using a set sampling frequency.
4. The non-invasive lithium-ion battery state of charge detection method according to claim 2, wherein: The devices for realizing the in-situ characterization method include: atomic force microscope, Raman spectrometer and scanning electron microscope.
5. The non-invasive lithium-ion battery state of charge detection method according to claim 2, wherein: In-situ characterization is used to obtain the microscopic properties of lithium-ion battery materials at different states of charge, including: Lithium-ion batteries at different states of charge are disassembled, and the in-situ characterization method is used to perform electrode testing on the disassembled lithium-ion batteries to obtain the microscopic properties of the material.
6. The non-invasive lithium-ion battery state of charge detection method according to claim 2, characterized in that: A SOC quantitative mapping model is constructed based on historical characteristic parameters of the lithium-ion battery at different states of charge and material microscopic properties corresponding to the historical characteristic parameters, including: Construct an initial SOC quantitative mapping model based on the CNN-LSTM model; Inputting the historical characteristic parameters into the initial SOC quantitative mapping model to obtain an initial predicted state of charge and an initial predicted material microstate; The parameters of the initial SOC quantitative mapping model are optimized based on the initial predicted state of charge, the state of charge of the lithium-ion battery corresponding to the historical characteristic parameters, the initial predicted material microstate, and the material microproperties corresponding to the historical characteristic parameters, until the output of the initial SOC quantitative mapping model meets the set requirements, and the optimized initial SOC quantitative mapping model is used as the SOC quantitative mapping model.
7. The non-invasive lithium-ion battery state of charge detection method according to claim 6, characterized in that: The mean absolute error and the mean variance error are used as the loss function of the initial SOC quantitative mapping model, and the Huber loss function is used to balance the mean absolute error and the mean variance error; the loss function of the initial SOC quantitative mapping model is used to determine whether the output of the initial SOC quantitative mapping model meets the set requirements.
8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the non-invasive lithium-ion battery state of charge detection method according to any one of claims 1 to 7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the non-invasive lithium-ion battery state of charge detection method according to any one of claims 1 to 7 is implemented.
10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the non-invasive lithium-ion battery state of charge detection method according to any one of claims 1 to 7 is implemented.