Battery State of Charge (SOC) Estimation Method, Device, Equipment and Storage Medium
By acquiring multiple ultrasound features and combining neural network models, the problem of inaccurate SOC estimation caused by single ultrasound features is solved, and the accurate estimation of battery charge state is achieved.
Patent Information
- Application Number
- CN202210545225.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-19
- Publication Date
- 2025-07-04
- Estimated Expiration
- 2042-05-19
AI Technical Summary
In the existing battery state of charge SOC estimation methods, single ultrasound characteristics lead to inaccurate detection results and insufficient applicability.
By obtaining the time domain peak, frequency domain peak, time domain center, frequency domain center, energy integral, waveform index, kurtosis coefficient, skewness coefficient, discrete coefficient, shape coefficient, offset frequency and other characteristics of the ultrasonic response signal at different frequencies, combined with the fully connected neural network and gradient enhancement model, a battery SOC estimation model is generated and SOC estimation is performed.
Improves the accuracy and reliability of battery state of charge SOC estimation, and can provide accurate SOC value estimation in different SOC intervals.
Smart Images
Figure CN115015761B_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the technical field of battery management, and particularly relates to a method, device, equipment and storage medium for estimating the state of charge (SOC) of a battery. Background Art
[0002] The evaluation of the state of charge (SOC) of power batteries for electric vehicles is very important for both the real-time state detection of in-service power batteries and the cascade utilization of retired batteries.
[0003] Most power battery detection methods evaluate the battery state by measuring information such as the current, voltage, and temperature of the battery. However, the battery voltage, current, and temperature at a single moment cannot provide reliable battery state information. It is necessary to use the time-series data generated during a period of time, or even the complete battery charge and discharge process, and combine battery model technology to estimate the battery's state of charge, health state, etc., and the testing takes a long time.
[0004] Currently, due to the advantages of ultrasonic detection technology such as fast speed, accuracy, and non-destructiveness, ultrasonic detection technology has begun to be applied in the evaluation of the SOC of power batteries.
[0005] However, when detecting the battery SOC, the selected ultrasonic features are single, resulting in inaccurate battery SOC estimation results and poor applicability. Summary of the Invention
[0006] This application provides a method, device, equipment and storage medium for estimating the state of charge (SOC) of a battery, which avoids the problem of inaccurate detection results caused by single ultrasonic features and improves the accuracy of the SOC estimation value.
[0007] In a first aspect, this application provides a method for estimating the state of charge (SOC) of a battery, including:
[0008] During the charge and discharge process of the battery to be measured, according to a preset SOC period, obtain the ultrasonic response signals corresponding to the ultrasonic excitation signals received by the battery to be measured at different frequencies;
[0009] Extract the ultrasonic response signals to obtain ultrasonic features at different frequencies. The ultrasonic features include at least three features among the time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency;
[0010] Input the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery to be measured.
[0011] In a possible implementation of the first aspect, obtaining the ultrasonic response signal corresponding to the ultrasonic excitation signal received by the battery under test according to the preset SOC period includes:
[0012] Send an ultrasonic excitation signal within a preset frequency range to the battery under test every preset SOC period;
[0013] Collect the ultrasonic response signal corresponding to the ultrasonic excitation signal.
[0014] In a possible implementation of the first aspect, extracting the ultrasonic response signal to obtain ultrasonic features includes:
[0015] Extract the ultrasonic response signal to obtain the first ultrasonic feature at each frequency. The first ultrasonic feature includes at least three features among the time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency;
[0016] Determine the optimal excitation frequency for each feature in the first ultrasonic feature. The optimal excitation frequency is the frequency corresponding to the highest feature intensity;
[0017] Determine each feature corresponding to the optimal excitation frequency as the second ultrasonic feature;
[0018] Analyze and screen each feature in the second ultrasonic feature to determine the ultrasonic feature. The ultrasonic feature includes at least two features in the first ultrasonic feature that meet the preset conditions. The preset condition is that the range of change of the feature with the battery SOC is greater than the first preset range and fluctuates within the second preset range.
[0019] In a possible implementation of the first aspect, the process of generating the battery SOC estimation model includes:
[0020] During the charging and discharging process of the sample battery, obtain the sample ultrasonic features corresponding to the ultrasonic response signals of the sample battery within a preset frequency range at different SOCs. The sample battery is at least one;
[0021] Train the original fully connected neural network FNN model according to the sample ultrasonic features to obtain the FNN model. The FNN model is used to estimate the SOC value in the first SOC interval;
[0022] Train the original gradient boosting XGBoost model according to the sample ultrasonic features to obtain the XGBoost model. The XGBoost model is used to estimate the SOC value in the second SOC interval, and the second SOC interval is lower than the first SOC interval;
[0023] Fuse the FNN model and the XGBoost model to obtain the battery SOC estimation model.
[0024] In a possible implementation manner of the first aspect, training the original gradient boosting XGBoost model according to the sample ultrasonic features to obtain the XGBoost model includes:
[0025] Training the original gradient boosting XGBoost model according to the sample ultrasonic features and the failure features corresponding to the sample ultrasonic features with labels, where the failure features represent features that do not change with the battery SOC.
[0026] In a possible implementation manner of the first aspect, the battery SOC estimation model is further used to accurately estimate the SOC value of the battery when the ultrasonic features include missing features;
[0027] Inputting the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery to be measured includes:
[0028] Performing failure discrimination on the ultrasonic features, obtaining the failure features and non-failure features in the ultrasonic features, and marking the failure features;
[0029] Inputting the marked failure features and the non-failure features other than the failure features into the battery SOC estimation model to obtain the SOC estimation value of the battery.
[0030] In a possible implementation manner of the first aspect, during the charging and discharging process of the battery to be measured, after obtaining the ultrasonic response signal corresponding to the ultrasonic excitation signal of the battery to be measured in the preset frequency range according to the preset SOC period, it further includes:
[0031] When it is determined that the intensity of the ultrasonic response signal decays, it is determined that the battery to be measured is in an overcharge state or an over-discharge state.
[0032] A method for estimating the state of charge (SOC) of a battery according to the present application selects at least three features from the time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency to estimate the SOC of the battery to be measured, and inputs multiple features into the battery SOC estimation model for SOC estimation, ensuring the accuracy of SOC estimation.
[0033] In a second aspect, the present application provides a device for estimating the state of charge (SOC) of a battery. This device is used to execute the method in the first aspect or any possible implementation manner of the first aspect. Specifically, the device may include:
[0034] A collection module, configured to obtain ultrasonic response signals corresponding to ultrasonic excitation signals received by the battery under test at different frequencies according to a preset SOC period during the charge and discharge process of the battery under test;
[0035] An acquisition module, configured to extract the ultrasonic response signals to obtain ultrasonic features at different frequencies, where the ultrasonic features include at least three features of a time-domain peak value, a frequency-domain peak value, a time-domain centroid, a frequency-domain centroid, an energy integral, a waveform index, a kurtosis coefficient, a skewness coefficient, a dispersion coefficient, a shape coefficient, and an offset frequency;
[0036] An estimation module, configured to input the ultrasonic features into a battery SOC estimation model to obtain an SOC estimation value of the battery under test, where the battery SOC estimation model is used to accurately estimate the SOC value of each SOC of the battery.
[0037] In a third aspect, the present application provides an electronic device, which includes a memory and a processor. The memory is used to store instructions; the processor executes the instructions stored in the memory, so that the device executes the method for estimating the state of charge (SOC) of the battery in the first aspect or any possible implementation manner of the first aspect.
[0038] In a fourth aspect, a computer-readable storage medium is provided, in which instructions are stored. When the instructions are run on a computer, the computer is caused to execute the method for estimating the state of charge (SOC) of the battery in the first aspect or any possible implementation manner of the first aspect.
[0039] In a fifth aspect, a computer program product containing instructions is provided. When the instructions are run on a device, the device is caused to execute the method for estimating the state of charge (SOC) of the battery in the first aspect or any possible implementation manner of the first aspect.
[0040] It can be understood that the beneficial effects of the above second aspect to fifth aspect can refer to the relevant descriptions in the above first aspect, and will not be elaborated here. Description of the Drawings
[0041] In order to more clearly illustrate the technical solutions in the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0042] Figure 1 It is a schematic flowchart of a method for estimating the SOC of a battery provided by an embodiment of the present application;
[0043] Figure 2aIt is a schematic flow chart of a battery SOC estimation method provided by an embodiment of the present application;
[0044] Figure 2b It is a schematic diagram of normalization analysis provided by an embodiment of the present application;
[0045] Figure 2c It is a schematic diagram of sensitivity analysis provided by an embodiment of the present application;
[0046] Figure 2d It is a schematic diagram of correlation analysis provided by an embodiment of the present application;
[0047] Figure 3 It is a schematic flow chart of a battery SOC estimation method provided by an embodiment of the present application;
[0048] Figure 4a It is a schematic flow chart of a battery SOC estimation method provided by an embodiment of the present application;
[0049] Figure 4b It is a schematic flow chart of a battery SOC estimation method provided by an embodiment of the present application;
[0050] Figure 4c It is a schematic flow chart of a battery SOC estimation method provided by an embodiment of the present application;
[0051] Figure 5a It is a schematic diagram of sending and collecting signals provided by an embodiment of the present application;
[0052] Figure 5b It is a schematic flow chart of a battery SOC estimation method provided by an embodiment of the present application;
[0053] Figure 5c It is a schematic diagram of testing a battery SOC estimation method provided by an embodiment of the present application;
[0054] Figure 6 It is a schematic structural diagram of a battery SOC estimation device provided by the present application;
[0055] Figure 7 It is a schematic structural diagram of an electronic device provided by the present application. Detailed implementation manners
[0056] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures, technologies, etc. are presented to thoroughly understand the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.
[0057] It should be understood that when used in the specification and appended claims of this application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.
[0058] It should also be understood that the term "and / or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0059] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" depending on the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" depending on the context.
[0060] In addition, in the description of the specification and appended claims of this application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.
[0061] Reference to "one embodiment" or "some embodiments" or the like described in the specification of this application means that a specific feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.
[0062] The embodiments of this application provide a method, device, equipment and readable storage medium for estimating the state of charge (SOC) of a battery. This method can be implemented through a battery management system and is applied to the scenario of battery SOC estimation.
[0063] Among them, the state of charge (SOC) of the battery is one of the main parameters of the battery state, and its value is defined as the ratio of the remaining capacity of the battery to the total capacity of the battery.
[0064] The above-mentioned battery includes, but is not limited to, lithium batteries. For example, it may also include lithium metal-air batteries, lead-acid batteries, nickel-metal hydride batteries, nickel-cadmium batteries, etc., and specific types are not limited here.
[0065] In addition, the battery management system is connected to the electrical system including the above-mentioned battery. For example, the battery management system can communicate with the electrical system in forms such as an application (APP), a web page, a public account in an application, and a mini-program, enabling the battery management system to send ultrasonic excitation signals to the battery in the electrical system, collect ultrasonic response signals from the battery, and then estimate the SOC value based on the ultrasonic response signals.
[0066] It can be understood that the ultrasonic response signal can detect the acoustic performance difference caused by the change in the mechanical properties of the battery electrode material structure. Thus, the state of charge of the battery to be measured can be detected using the ultrasonic response signal.
[0067] Among them, the electrical system is applied to a power device. The electrical system can generate propulsion torque and transmit the propulsion torque to the drive wheels to drive the power device to move.
[0068] The power device can be a hybrid electric vehicle or a battery electric vehicle, and can also be, for example, an airplane, a ship, or a rail vehicle, as well as a fixed or mobile power device, platform, robot, etc., but is not limited thereto.
[0069] Among them, the battery management device includes a battery management system, and the battery management device can be a smart phone, a tablet computer, a desktop computer, a server, etc. At the same time, the battery management device includes a display screen, or is externally connected to a display screen.
[0070] Based on the above scenario description, below, taking the battery management system as an example, in combination with the accompanying drawings and application scenarios, the battery SOC estimation method provided by the embodiments of the present application will be described in detail.
[0071] Please refer to Figure 1 , Figure 1 which shows a schematic flow chart of the battery SOC estimation method provided by an embodiment of the present application.
[0072] As Figure 1 shown, the battery state of charge SOC estimation method provided by the present application may include:
[0073] S101. During the charging and discharging process of the battery to be measured, according to a preset SOC period, obtain the ultrasonic response signals corresponding to the ultrasonic excitation signals received by the battery to be measured at different frequencies.
[0074] The battery management system can preset a preset SOC period in advance and store the preset SOC period in the battery management system and / or a storage device.
[0075] Among them, the storage device can communicate with the battery management system, enabling the battery management system to obtain the preset SOC period from the storage device. The storage method and specific type of the storage device are not limited in this application.
[0076] The battery management system can send an ultrasonic excitation signal to the battery under test at the previous SOC of the preset SOC period and collect the ultrasonic response signal corresponding to the ultrasonic excitation signal at the next SOC.
[0077] Among them, the battery management system can set the preset SOC period according to actual needs. For example, the preset SOC period can be 1%, 5%, 10%, etc., and specific values are not limited here.
[0078] In some embodiments, when the preset SOC period is 5%, the battery management system sends an ultrasonic excitation signal to the battery under test every 5% SOC.
[0079] For example, the battery management system, based on the preset SOC period of 5% stored in advance, sends an ultrasonic excitation signal to the battery under test at the previous 5% SOC period, and collects the ultrasonic response signal when the battery under test is in the next 5% SOC period. At the same time, when the battery under test is in the next 5% SOC, an ultrasonic excitation signal is sent to the battery under test again.
[0080] S102. Extract the ultrasonic response signal to obtain ultrasonic features at different frequencies.
[0081] Among them, the ultrasonic features include at least three of the time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency.
[0082] The offset frequency refers to the fact that when the battery is at different SOCs, the optimal excitation frequency of the ultrasonic features will change to a certain extent.
[0083] The characteristic frequencies corresponding to the battery at different SOCs are different, and this offset of the characteristic frequencies caused by different battery SOCs can also be used to estimate the SOC of the battery conversely.
[0084] Among them, the calculation formulas for the time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency are as follows:
[0085] Time-domain peak value (Pt), Pt = max(A)
[0086] Among them, A represents the time-domain signal in the ultrasonic response signal, that is, the time-domain data.
[0087] Frequency domain peak value (Pf), Pf = max(S)
[0088] Wherein, S represents the frequency domain signal in the ultrasonic response signal, that is, the frequency domain data.
[0089] Time domain centroid (Ct)
[0090] Wherein, A represents the time domain signal in the ultrasonic response signal, that is, the time domain data.
[0091] Frequency domain centroid (Cf)
[0092] Wherein, S represents the frequency domain signal in the ultrasonic response signal, that is, the frequency domain data.
[0093] Energy integral (E)
[0094] Wherein, A represents the time domain signal in the ultrasonic response signal, that is, the time domain data.
[0095] Waveform index (W)
[0096] Wherein, A represents the time domain signal in the ultrasonic response signal, that is, the time domain data, and N represents the number of times of collecting the ultrasonic response signal.
[0097] Kurtosis coefficient (Ku)
[0098] Wherein, A represents the time domain signal in the ultrasonic response signal, that is, the time domain data, and N represents the number of times of collecting the ultrasonic response signal.
[0099] Skewness coefficient (Sk)
[0100] Wherein, A represents the time domain signal in the ultrasonic response signal, that is, the time domain data, and N represents the number of times of collecting the ultrasonic response signal.
[0101] Coefficient of dispersion (D)
[0102] Wherein, A represents the time domain signal in the ultrasonic response signal, that is, the time domain data, and N represents the number of times of collecting the ultrasonic response signal.
[0103] Shape coefficient (Sc)
[0104] Wherein, A represents the time domain signal in the ultrasonic response signal, that is, the time domain data, and N represents the number of times of collecting the ultrasonic response signal.
[0105] Offset frequency (Of)
[0106] Among them, A represents the time-domain signal in the ultrasonic response signal, i.e., time-domain data, and N represents the number of times of collecting the ultrasonic response signal.
[0107] For example, after the battery management system collects the ultrasonic response signal, it extracts 9 features from the ultrasonic response signal, including the time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency.
[0108] S103. Input the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery to be measured.
[0109] The battery SOC estimation model is used to accurately estimate the SOC value of each SOC of the battery.
[0110] Among them, the battery SOC estimation model is used to estimate all SOCs of the battery to be measured.
[0111] The model training system can pre-generate the battery SOC estimation model and store the battery SOC estimation model in the battery management system and / or storage device.
[0112] Based on S102, after the battery management system collects the ultrasonic features, it inputs the ultrasonic features into the battery SOC estimation model, and then the SOC estimation value of the battery to be measured can be output.
[0113] For example, after the battery management system collects the ultrasonic response signal and extracts 9 features from the ultrasonic response signal, including the time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency, it inputs the 9 collected features into the battery SOC estimation model, and then the SOC estimation value of the battery to be measured can be output.
[0114] For the battery state of charge SOC estimation method provided in this application, the battery management system obtains the ultrasonic response signal corresponding to the ultrasonic excitation signal of the battery to be measured according to the preset SOC period, extracts the ultrasonic response signal to obtain ultrasonic features, and the ultrasonic features include at least three features. By inputting the ultrasonic features into the battery SOC estimation model, the SOC estimation value of the battery to be measured can be output. By inputting the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery to be measured, the accuracy of the SOC estimation value of the battery to be measured is ensured.
[0115] Based on the above Figure 1 description of the embodiments shown, when the battery management system obtains the ultrasonic features, it can obtain them in various ways.
[0116] Next, in combination with Figure 2a, a specific implementation process of the state of charge (SOC) estimation method for the battery of this application will be introduced in detail.
[0117] Based on Figure 1 the description of S102, when the battery management system acquires the ultrasonic characteristics, it needs to first acquire the first ultrasonic characteristics, then determine the optimal excitation frequency of each characteristic in the first ultrasonic characteristics, and finally determine the ultrasonic characteristics according to the second ultrasonic characteristics corresponding to the optimal excitation frequency.
[0118] Please refer to Figure 2a , Figure 2a which shows a schematic flowchart of the battery SOC estimation method provided by an embodiment of this application.
[0119] As Figure 2a shown, the battery SOC estimation method provided by this application may include:
[0120] S201. During the charging and discharging process of the battery under test, at every preset SOC period, an ultrasonic excitation signal within a preset frequency range is sent to the battery under test.
[0121] The battery management system has a data setting page. Among them, the data setting page is used to set the SOC period and the frequency range.
[0122] In some embodiments, the data setting page may include an SOC period setting window, a frequency range setting window, and a sending control. The SOC period setting window is used to set the SOC period, the frequency range setting window is used to set the frequency range, and the determination control is used to trigger sending an ultrasonic excitation signal within a preset frequency range at every SOC period.
[0123] The user can set the preset SOC period and the preset frequency range on the data setting page of the battery management system, and send an ultrasonic excitation signal within the preset frequency range to the battery under test.
[0124] The battery management system can preset the preset frequency range and store the preset frequency range in the battery management system and / or the storage device.
[0125] In some embodiments, the battery management system includes an ultrasonic generating device located on the outer surface of the battery under test, and the ultrasonic generating device is used to emit an ultrasonic excitation signal to the battery through the battery management system.
[0126] Among them, the ultrasonic generating device can be arranged on the upper surface of the battery under test through a probe.
[0127] In some embodiments, a specific illustration is made with the preset SOC period being 5%. When the preset SOC period is 5%, the battery management system sends an ultrasonic excitation signal with a frequency range of 25 - 200 kHz to the battery under test at every 5% SOC.
[0128] For example, the battery management system, according to a preset SOC cycle stored in advance with a value of 5% and a frequency range of 25 - 200 kHz, when the battery under test is in the previous 5% SOC cycle, sends an ultrasonic excitation signal with a frequency range of 25 - 200 kHz to the battery under test, and when the battery under test is in the next 5% SOC cycle, collects an ultrasonic response signal with a frequency range of 25 - 200 kHz. At the same time, when the battery under test is in the next 5% SOC cycle, it sends an ultrasonic excitation signal with a frequency range of 25 - 200 kHz to the battery under test again. S202: Collect the ultrasonic response signal corresponding to the ultrasonic excitation signal.
[0129] The battery management system includes an ultrasonic data acquisition device located on the outer surface of the battery under test, and the ultrasonic data acquisition device is used to collect the ultrasonic response signal propagated through the battery under test.
[0130] Among them, the ultrasonic data acquisition device collects the ultrasonic response signal emitted by the ultrasonic generating device and propagated through the battery under test, providing data preparation for extracting the first ultrasonic feature in the ultrasonic response signal.
[0131] S203: Extract the ultrasonic response signal to obtain the first ultrasonic feature of each frequency.
[0132] Among them, the first ultrasonic feature includes at least three features of time - domain peak value, frequency - domain peak value, time - domain centroid, frequency - domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, offset frequency.
[0133] S204: Determine the optimal excitation frequency of each feature in the first ultrasonic feature, and the optimal excitation frequency is the frequency corresponding to the highest feature intensity.
[0134] Under multi - frequency excitation in the preset frequency range, the battery under test can reflect the response conditions at different frequencies.
[0135] Based on S203, when the battery management system obtains the first ultrasonic feature of each frequency, according to the relationship curve between the intensity of each feature in the first ultrasonic feature and the frequency, it selects the sensitive frequency of each feature, that is, the optimal excitation frequency.
[0136] For example, the optimal frequencies of time-domain peak, frequency-domain peak, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency are time-domain peak (100 kHz), frequency-domain peak (100 kHz), time-domain centroid (50 kHz), frequency-domain centroid (40 kHz), energy integral (100 kHz), waveform index (125 kHz), kurtosis coefficient (140 kHz), skewness coefficient (85 kHz), shape coefficient (50 kHz), respectively.
[0137] Among them, the dispersion coefficient is not very sensitive in the experiment, so it is not used.
[0138] The offset frequency means that when the battery is at different SOCs, the sensitive frequency of the first ultrasonic feature will change to a certain extent. Therefore, there is no sensitive frequency for the offset frequency.
[0139] S205. Determine each feature corresponding to the optimal excitation frequency as the second ultrasonic feature.
[0140] For example, determine the time-domain peak (100 kHz), frequency-domain peak (100 kHz), time-domain centroid (50 kHz), frequency-domain centroid (40 kHz), energy integral (100 kHz), waveform index (125 kHz), kurtosis coefficient (140 kHz), skewness coefficient (85 kHz), shape coefficient (50 kHz), and offset frequency as the second ultrasonic features.
[0141] S206. Analyze and screen each feature in the second ultrasonic features to determine the ultrasonic features, where the ultrasonic features include at least two features in the first ultrasonic features that meet the preset conditions.
[0142] Among them, the preset condition is that the change range of the feature with the change of the battery SOC is greater than the first preset range and fluctuates within the second preset range, that is, it changes significantly and fluctuates stably with the change of the SOC of the battery to be measured.
[0143] Among them, the analysis includes normalization processing, correlation analysis, and sensitivity analysis.
[0144] Sensitivity analysis means that some of the second ultrasonic features may not be applicable to battery detection. Therefore, it is necessary to analyze the second ultrasonic features through sensitivity analysis to determine the second ultrasonic features that are sensitive to the internal changes of the battery for state estimation of the battery.
[0145] Correlation analysis means that some of the second ultrasonic features may not be applicable to battery detection. Therefore, it is necessary to analyze the second ultrasonic features through correlation analysis to determine the second ultrasonic features that are sensitive to the internal changes of the battery for state estimation of the battery.
[0146] Screening refers to passing the second ultrasonic features sensitive to internal battery changes through normalization processing, correlation analysis, and sensitivity analysis, and taking the second ultrasonic features sensitive to internal battery changes as ultrasonic features.
[0147] Illustrated by way of example, the process of the battery management system analyzing and screening each feature in the second ultrasonic features to determine the ultrasonic features is described as follows:
[0148] As Figure 2b shown, it is the range of change of each second ultrasonic feature with the change of battery SOC after normalization. Among them, the larger the range of change, the more applicable the second ultrasonic feature is.
[0149] It can be seen from the figure that the ranges of change of the three second ultrasonic features, namely kurtosis coefficient, skewness coefficient, and dispersion coefficient, are relatively narrow, which is not conducive to the estimation of the SOC of the battery under test.
[0150] As Figure 2c shown, it is the process of calculating the sensitivity based on the least significant bit (LSB) of each second ultrasonic feature. Among them, the greater the sensitivity, the more applicable the second ultrasonic feature is.
[0151] The least significant bit interval of the change of the second ultrasonic feature with the change of battery SOC. It can be seen from Figure 2c that the accuracies of the three second ultrasonic features, namely dispersion coefficient, shape coefficient, and offset frequency, are relatively low, which is not conducive to the estimation of the SOC of the battery under test.
[0152] As Figure 2d shown, based on the data of multiple tests of the battery under test, calculate the fluctuation degree of the second ultrasonic feature during multiple charge and discharge processes of the battery under test. The smaller the fluctuation degree, the more stable the second ultrasonic feature is.
[0153] The fluctuation degree of the second ultrasonic feature in multiple tests, aiming to test the stability of the test index. It can be seen from Figure 2d that the stability of the first six indexes is significantly greater than that of the last five indexes.
[0154] Based on the above analysis, after screening, the determined ultrasonic features include: time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, waveform index, and offset frequency.
[0155] S207. Input the ultrasonic features into the battery SOC estimation model to obtain the SOC estimated value of the battery under test.
[0156] Illustrated by way of example, input the determined ultrasonic features after screening, namely time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, energy integral, and waveform index, into the battery SOC estimation model to obtain the SOC estimated value of the battery under test.
[0157] In this application, the battery management system first obtains the first ultrasonic feature from the ultrasonic response signal, then determines the optimal excitation frequency of each feature in the first ultrasonic feature, and finally determines the ultrasonic feature according to the second ultrasonic feature corresponding to the optimal excitation frequency. By means of the optimal excitation frequency, by inputting the ultrasonic features obtained through multiple screenings into the battery SOC estimation model to estimate the value of SOC, the accuracy of the estimated SOC value can be guaranteed.
[0158] Based on Figure 1 the description of the illustrated embodiment, this application also provides the generation process of a pre-designed battery SOC estimation model.
[0159] Next, in combination with Figure 3 , the specific implementation manners of this application for executing the above process will be introduced in detail.
[0160] Based on Figure 1 the description of S103, the battery management system uses the battery SOC estimation model to accurately estimate the entire SOC of the battery under test.
[0161] The generation process of the battery SOC estimation model can be generated by the SOC estimation model generation system, or can be generated by other feasible model generation systems, which will not be elaborated here.
[0162] Please refer to Figure 3 , Figure 3 which shows the schematic flow chart of the battery state of charge SOC estimation method provided by an embodiment of this application.
[0163] As Figure 3 shown, the process of generating the battery SOC estimation model provided by this application may include:
[0164] S301. During the charging and discharging process of the sample battery, at different SOCs, obtain the sample ultrasonic features corresponding to the ultrasonic response signals of the sample battery within a preset frequency range, where the sample battery is at least one.
[0165] In some embodiments, there are multiple sample batteries.
[0166] Among them, the sample battery may be a pre-prepared experimental battery. For example, it may be a lithium battery pack.
[0167] S302. According to the sample ultrasonic features, train the original fully connected neural network FNN model to obtain the FNN model. The fully connected neural network (FNN) model is a 7-input network, with two hidden layers, the number of neurons in the hidden layer is 32, and the output is a single output.
[0168] Among them, the FNN model is used to accurately estimate the SOC value in the first SOC interval. For example, the range of the first SOC interval can be 50-100%.
[0169] S303. Train the original extreme gradient boosting (XGBoost) model according to the sample ultrasonic features to obtain the XGBoost model.
[0170] Among them, the original extreme gradient boosting (XGBoost) model, that is, the XGBoost model is used to accurately estimate the SOC value in the second SOC interval, and the second SOC interval is lower than the first SOC interval. For example, the range of the second SOC interval can be 0-50%.
[0171] It should be noted that the execution order of S303 and S304 is not in sequence, and they can be executed sequentially or simultaneously.
[0172] In some embodiments, the XGBoost model is trained according to the sample ultrasonic features and failure features.
[0173] Due to the interference of external environmental factors, such as external noise, vibration of the test bench and other factors, it may cause some ultrasonic features to be unavailable for estimating the SOC of the battery under test, that is, failure features.
[0174] Among them, the failure feature is a feature that does not change with the change of the battery SOC.
[0175] Therefore, in some embodiments, when generating the XGBoost model, the XGBoost model is trained according to the sample ultrasonic features and failure features. The finally obtained XGBoost model can also estimate the battery SOC when there are failure features in the ultrasonic features collected by the battery management system.
[0176] S304. Integrate the FNN model and the XGBoost model to obtain the battery SOC estimation model.
[0177] When integrating the two models, different weights are assigned to the prediction results of the two models in different SOC intervals to obtain the final battery SOC estimation value.
[0178] In some embodiments, integrating the FNN model and the XGBoost model to obtain the battery SOC estimation model can also estimate the SOC of the battery under test when there are missing features in the ultrasonic features of the battery under test. In other embodiments, when the battery management system determines that there are missing features in the ultrasonic features, the missing features are set to -1, and the SOC of the battery under test is estimated through the battery SOC estimation model.
[0179] For example, after the battery management system obtains the ultrasonic characteristics, it discriminates the failure of the ultrasonic characteristics. When it is determined that there are failure characteristics in the ultrasonic characteristics, the marked failure characteristics and the non-failure characteristics other than the failure characteristics are input into the SOC estimation model to obtain the SOC estimation value of the battery.
[0180] In this application, when the SOC estimation model generation system generates the battery SOC estimation model, it generates an FNN model with higher estimation accuracy for the first SOC interval of the battery to be measured. At the same time, it generates an XGBoost model with higher estimation accuracy for the second SOC interval of the battery to be measured, and then fuses the two models. The obtained battery SOC estimation model can estimate the accurate SOC value whether the battery to be measured is in the high SOC interval or the low SOC interval.
[0181] Based on Figure 1 the description of the embodiments shown, the battery management system can also detect the faults of the battery to be measured in various ways, that is, detect whether the battery to be measured is in an overcharged state or an over-discharged state.
[0182] Next, in combination with Figure 4a , the specific implementation manners of the battery management system for executing the above process will be introduced in detail.
[0183] Please refer to Figure 4a , Figure 4a which shows a schematic flow chart of a method for estimating the state of charge (SOC) of a battery provided by an embodiment of this application.
[0184] As Figure 4a shown, the method for estimating the state of charge (SOC) of a battery provided by this application may include:
[0185] S401. During the charging and discharging process of the battery to be measured, according to a preset SOC period, obtain the ultrasonic response signal corresponding to the ultrasonic excitation signal received by the battery to be measured.
[0186] S402. When it is determined that the intensity of the ultrasonic response signal has attenuated, determine that the battery to be measured is in an overcharged state or an over-discharged state.
[0187] Since the intensity of the ultrasonic signal will attenuate significantly when passing through gas. When the battery is overcharged and over-discharged, the electrochemical side reactions inside the battery will intensify, resulting in the generation of gas, and the fault condition of the battery is judged by detecting the gas inside the battery.
[0188] For example, as Figure 4b shown is the change of the time-domain peak value of the ultrasonic signal during the overcharging process of the battery to be measured. The object of this experiment is a 10 Ah battery to be measured, and the upper charging limit voltage is 4.25.
[0189] As can be seen in Figure 4b when the battery is over-discharged to about 4.6V, the ultrasonic response signal will have a very obvious attenuation. Subsequent battery detection can be based on the significant attenuation of the signal to detect the overcharge state of the battery.
[0190] As Figure 4c shown, the schematic diagram of the change in the ultrasonic signal peak during the over-discharge process of the battery under test is similar to overcharge, and there will be a very obvious attenuation at a critical point.
[0191] S403. Extract the ultrasonic response signal to obtain ultrasonic features.
[0192] It should be noted that the execution order of S402 and S403 is not sequential. They can be executed sequentially or simultaneously.
[0193] S404. Input the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery under test.
[0194] Among them, S401, S403 and S404 are respectively similar to the implementation manners of S101, S102 and S103 in the Figure 1 embodiment shown, and will not be elaborated here in this application.
[0195] In this application, when the battery management system determines that the intensity of the ultrasonic response signal has a significant attenuation, it determines that the battery under test is in an overcharge state or an over-discharge state. By means of whether the intensity of the ultrasonic response signal has a significant attenuation, the overcharge state or the over-discharge state of the battery under test can be quickly determined.
[0196] Based on the above description, in a specific embodiment, as Figure 5a shown in the schematic diagram of the transmitted signal and the acquired signal, the following assumptions are made:
[0197] 1. The battery management system sends an ultrasonic excitation signal to the battery under test through an ultrasonic generator, and collects the ultrasonic response signal through an ultrasonic data collector;
[0198] 2. The ultrasonic generator is connected to probe 1, and the ultrasonic data collector is connected to probe 2. Probe 1 and probe 2 are connected to the same side of the battery under test and are located on both sides of the battery;
[0199] 3. Both probe 1 and probe 2 are inclined on the battery under test, and the directions of probe 1 and probe 2 close to the battery under test are away from each other, which can attenuate the intensity of the reflected signal;
[0200] 4. The battery under test is a power battery.
[0201] As Figure 5bThe schematic flowchart of the SOC estimation method shown above. Based on the above assumptions, the battery management system can execute the battery SOC estimation method provided in this application.
[0202] The battery management system executing the battery SOC estimation method may include:
[0203] Step 11: During the charge and discharge process of the battery under test, when the battery management system determines that the power battery under test is in the previous 5% SOC cycle, it sends an excitation signal with a frequency range of 25 - 200 kHz to the power battery under test through probe 1 connected to the ultrasonic generator.
[0204] Step 12: When the battery management system determines that the power battery under test is in the next 5% SOC cycle, it collects the ultrasonic response signal through probe 2 connected to the ultrasonic data collector.
[0205] Step 13: Extract the ultrasonic response signal to obtain the first ultrasonic feature at each frequency. The first ultrasonic feature includes the time-domain peak, frequency-domain peak, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, dispersion coefficient, shape coefficient, and offset frequency at each frequency.
[0206] Step 14: Determine the optimal excitation frequency for the time-domain peak, frequency-domain peak, time-domain centroid, frequency-domain centroid, energy integral, waveform index, kurtosis coefficient, skewness coefficient, and shape coefficient. Among them, the time-domain peak (100 kHz), frequency-domain peak (100 kHz), time-domain centroid (50 kHz), frequency-domain centroid (40 kHz), energy integral (100 kHz), waveform index (125 kHz), kurtosis coefficient (140 kHz), skewness coefficient (85 kHz), shape coefficient (50 kHz).
[0207] Step 15: Determine the time-domain peak (100 kHz), frequency-domain peak (100 kHz), time-domain centroid (50 kHz), frequency-domain centroid (40 kHz), energy integral (100 kHz), waveform index (125 kHz), kurtosis coefficient (140 kHz), skewness coefficient (85 kHz), shape coefficient (50 kHz), dispersion coefficient, and offset frequency as the second ultrasonic feature.
[0208] Step 16: Analyze and screen the time-domain peak (100 kHz), frequency-domain peak (100 kHz), time-domain centroid (50 kHz), frequency-domain centroid (40 kHz), energy integral (100 kHz), waveform index (125 kHz), kurtosis coefficient (140 kHz), skewness coefficient (85 kHz), shape coefficient (50 kHz), dispersion coefficient, and offset frequency to obtain ultrasonic features: time-domain peak (100 kHz), frequency-domain peak (100 kHz), time-domain centroid (50 kHz), frequency-domain centroid (40 kHz), energy integral (100 kHz), waveform index (125 kHz), and offset frequency.
[0209] Step 17: Perform failure discrimination on the time-domain peak (100 kHz), frequency-domain peak (100 kHz), time-domain centroid (50 kHz), frequency-domain centroid (40 kHz), energy integral (100 kHz), waveform index (125 kHz), and offset frequency. When it is determined that the energy integral (100 kHz) is a failure feature, set the energy integral (100 kHz) to -1, splice it with the non-failure features other than the failure feature, and then input it into the battery SOC estimation model to obtain the SOC estimation value of the battery.
[0210] In summary, by testing a normal battery under test, an overcharged battery under test, and an over-discharged battery under test, the SOC and fault conditions of the battery under test can be obtained. The test results Figure 5c are shown as follows.
[0211] Corresponding to Figure 1 the battery SOC estimation method described in the above
[0212] shown embodiment, the present application further provides a battery state of charge SOC estimation device. Figure 6 Next, in conjunction with
[0213] please refer to Figure 6 , Figure 6 which shows a schematic block diagram of a battery SOC estimation device provided by an embodiment of the present application.
[0214] As Figure 6 shown, a battery state of charge SOC estimation device 600 provided by an embodiment of the present application includes an acquisition module 601, an acquisition module 602, and an estimation module 603.
[0215] The acquisition module 601 is configured to obtain ultrasonic response signals corresponding to ultrasonic excitation signals received by the battery under test at different frequencies according to a preset SOC period during the charging and discharging process of the battery under test;
[0216] An acquisition module 602, configured to extract the ultrasonic response signal to obtain ultrasonic features at different frequencies, where the ultrasonic features include at least three of a time-domain peak value, a frequency-domain peak value, a time-domain centroid, a frequency-domain centroid, an energy integral, a waveform index, a kurtosis coefficient, a skewness coefficient, a dispersion coefficient, a shape coefficient, and an offset frequency;
[0217] An estimation module 603, configured to input the ultrasonic features into a battery SOC estimation model to obtain an SOC estimation value of the battery to be measured, where the battery SOC estimation model is used to accurately estimate the SOC value of each SOC of the battery.
[0218] In some embodiments, the battery management system further includes a sending module ( Figure 6 not shown in the figure).
[0219] The sending module is configured to:
[0220] Every preset SOC period, send an ultrasonic excitation signal within a preset frequency range to the battery to be measured.
[0221] The acquisition module 601 is specifically configured to:
[0222] Acquire the ultrasonic response signal corresponding to the ultrasonic excitation signal.
[0223] In some embodiments, the acquisition module 602 is specifically configured to:
[0224] Extract the ultrasonic response signal to obtain first ultrasonic features at each frequency, where the first ultrasonic features include at least three of a time-domain peak value, a frequency-domain peak value, a time-domain centroid, a frequency-domain centroid, an energy integral, a waveform index, a kurtosis coefficient, a skewness coefficient, a dispersion coefficient, a shape coefficient, and an offset frequency;
[0225] Determine the optimal excitation frequency of each feature in the first ultrasonic features, where the optimal excitation frequency is the frequency corresponding to the highest feature intensity;
[0226] Determine each feature corresponding to the optimal excitation frequency as second ultrasonic features;
[0227] Analyze and screen each feature in the second ultrasonic features to determine ultrasonic features, where the ultrasonic features include at least two features in the first ultrasonic features that meet a preset condition, and the preset condition is that the range of change of the feature with the battery SOC is greater than a first preset range and fluctuates within a second preset range.
[0228] In some embodiments, the battery SOC estimation model system is used to:
[0229] During the charge and discharge process of the sample battery, at different SOCs, obtain the sample ultrasonic characteristics corresponding to the ultrasonic response signals of the sample battery within a preset frequency range, where the sample battery is at least one;
[0230] Train the original fully connected neural network FNN model according to the sample ultrasonic characteristics to obtain the FNN model, where the FNN model is used to estimate the SOC value in the first SOC interval;
[0231] Train the original gradient boosting XGBoost model according to the sample ultrasonic characteristics to obtain the XGBoost model, where the XGBoost model is used to estimate the SOC value in the second SOC interval, and the second SOC interval is lower than the first SOC interval;
[0232] Fuse the FNN model and the XGBoost model to obtain the battery SOC estimation model.
[0233] In some embodiments, the battery SOC estimation model system is specifically used for:
[0234] Train the original gradient boosting XGBoost model according to the sample ultrasonic characteristics and the failure characteristics corresponding to the sample ultrasonic characteristics with labels, where the failure characteristics are characteristics that do not change with the battery SOC.
[0235] In some embodiments, the estimation module 603 is specifically used for:
[0236] Perform failure discrimination on the ultrasonic characteristics, obtain the failure characteristics and non-failure characteristics in the ultrasonic characteristics, and label the failure characteristics;
[0237] Input the labeled failure characteristics and the non-failure characteristics other than the failure characteristics into the battery SOC estimation model to obtain the SOC estimation value of the battery.
[0238] In some embodiments, the battery management device further includes a determination module ( Figure 6 not shown in the figure).
[0239] The determination module is used for:
[0240] When it is determined that the intensity of the ultrasonic response signal has attenuated, determine that the battery under test is in an overcharged state or an over-discharged state.
[0241] It should be understood that the battery SOC estimation device 600 of the present application can be implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The above PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof. It can also be implemented by software. Figure 1 For the battery SOC estimation method shown, when implemented by software Figure 1 For the battery SOC estimation method shown, the device 600 and its various modules can also be software modules.
[0242] Figure 7 It is a schematic structural diagram of an electronic device provided by the present application. As Figure 7 shown, for the specific implementation manner of the electronic device, reference can be made to the description of the above battery management system, and it can execute the above battery SOC estimation method.
[0243] Among them, the electronic device 700 includes a processor 701, a memory 702, a communication interface 703, and a bus 706. Among them, the processor 701, the memory 702, and the communication interface 703 communicate through the bus 706, and communication can also be achieved by other means such as wireless transmission. The memory 702 is used to store instructions, and the processor 701 is used to execute the instructions stored in the memory 702. The memory 702 stores program code 7021, and the processor 701 can call the program code 7021 stored in the memory 702 to execute Figure 1 the battery SOC estimation method shown.
[0244] It should be understood that in the present application, the processor 701 can be a CPU, and the processor 701 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.
[0245] The memory 702 may include a read-only memory and a random access memory, and provide instructions and data to the processor 701. The memory 702 may also include a non-volatile random access memory. The memory 702 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchlink dynamic random access memory (SLDRAM), and direct rambus random access memory (DR RAM).
[0246] In addition to the data bus, the bus 706 may also include a power bus, a control bus, a status signal bus, etc. However, for the sake of clarity, in Figure 7 all kinds of buses are labeled as the bus 706.
[0247] It should be understood that the device 700 according to the present application may correspond to the device 600 in the present application, and may correspond to the device in the method shown in the present application Figure 1 When the device 700 corresponds to the device in the method shown in Figure 1 the above and other operations and / or functions of each module in the device 700 are respectively for implementing the operation steps of the method executed by the device in Figure 1 For the sake of brevity, they will not be repeated here.
[0248] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned various method embodiments can be implemented.
[0249] The present application provides a computer program product. When the computer program product runs on an electronic device, it enables the electronic device to execute steps in the above-mentioned method embodiments when implemented.
[0250] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not imply the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the present application.
[0251] It should be noted that the content such as information interaction and execution process between the above-mentioned device / units, due to being based on the same concept as the method embodiments of the present application, for its specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details are not described herein again.
[0252] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is used as an example for illustration. In actual applications, the above-mentioned functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the above-mentioned device can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working process of the units and modules in the above system can refer to the corresponding process in the foregoing method embodiments, and details are not described herein again.
[0253] In the above embodiments, the descriptions of the various embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0254] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.
[0255] In the embodiments provided in the present application, it should be understood that the disclosed device / network device and method can be implemented in other ways. For example, the device / network device embodiments described above are merely illustrative. For example, the division of the above-mentioned modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.
[0256] The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the present application.
[0257] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for estimating the state of charge (SOC) of a battery, characterized in that, Including: During the charge and discharge process of the battery under test, according to a preset SOC period, obtain the ultrasonic response signals corresponding to the ultrasonic excitation signals received by the battery under test at different frequencies; Extract the ultrasonic response signals to obtain ultrasonic features at different frequencies. The ultrasonic features include at least three features among time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, time-domain energy integral, time-domain waveform index, time-domain kurtosis coefficient, time-domain skewness coefficient, time-domain dispersion coefficient, time-domain shape coefficient, and time-domain offset frequency; Input the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery under test. The battery SOC estimation model is used to accurately estimate the SOC value of each SOC of the battery; Among them, the process of generating the battery SOC estimation model includes: During the charge and discharge process of the sample battery, at different SOCs, obtain the sample ultrasonic features corresponding to the ultrasonic response signals of the sample battery in a preset frequency range. The sample battery is at least one; Train the original fully connected neural network FNN model according to the sample ultrasonic features to obtain the FNN model. The FNN model is used to estimate the SOC value in the first SOC interval; Train the original gradient boosting XGBoost model according to the sample ultrasonic features and the failure features corresponding to the labeled sample ultrasonic features. The failure features are features that do not change with the battery SOC. The XGBoost model is used to estimate the SOC value in the second SOC interval, and the second SOC interval is lower than the first SOC interval; Fuse the FNN model and the XGBoost model to obtain the battery SOC estimation model.
2. The method according to claim 1, characterized in that The step of obtaining the ultrasonic response signals corresponding to the ultrasonic excitation signals received by the battery under test according to a preset SOC period includes: Send ultrasonic excitation signals in a preset frequency range to the battery under test every preset SOC period; Collect the ultrasonic response signals corresponding to the ultrasonic excitation signals.
3. The method according to claim 2, wherein The step of extracting the ultrasonic response signals to obtain ultrasonic features includes: Extract the ultrasonic response signals to obtain the first ultrasonic features at each frequency. The first ultrasonic features include at least three features among time-domain peak value, frequency-domain peak value, time-domain centroid, frequency-domain centroid, time-domain energy integral, time-domain waveform index, time-domain kurtosis coefficient, time-domain skewness coefficient, time-domain dispersion coefficient, time-domain shape coefficient, and time-domain offset frequency; Determine the optimal excitation frequency of each feature in the first ultrasonic features. The optimal excitation frequency is the frequency corresponding to the highest feature intensity; Determine each feature corresponding to the optimal excitation frequency as the second ultrasonic feature; Analyze and screen each feature in the second ultrasonic features to determine the ultrasonic features. The ultrasonic features include at least two features in the first ultrasonic features that meet the preset conditions. The preset conditions are that the range of change of the feature with the battery SOC is greater than the first preset range and fluctuates within the second preset range.
4. The method according to claim 3, characterized in that The battery SOC estimation model is also used to accurately estimate the SOC value of the battery when missing features are included in the ultrasonic features; The inputting the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery to be measured includes: Performing failure discrimination on the ultrasonic features, obtaining the failure features and non-failure features in the ultrasonic features, and marking the failure features; Inputting the marked failure features and the non-failure features other than the failure features into the battery SOC estimation model to obtain the SOC estimation value of the battery.
5. The method according to claim 1, wherein After obtaining the ultrasonic response signal corresponding to the ultrasonic excitation signal received by the battery to be measured according to a preset SOC period during the charge and discharge process of the battery to be measured, it further includes: When it is determined that the intensity of the ultrasonic response signal decays, it is determined that the battery to be measured is in an overcharge state or an over-discharge state.
6. A state of charge (SOC) estimation device for a battery, characterized in that, It includes: An acquisition module, configured to obtain, according to a preset SOC period during the charge and discharge process of the battery to be measured, the ultrasonic response signal corresponding to the ultrasonic excitation signal received by the battery to be measured at different frequencies; An acquisition module, configured to extract the ultrasonic response signal to obtain ultrasonic features at different frequencies, where the ultrasonic features include at least three features of a time-domain peak value, a frequency-domain peak value, a time-domain centroid, a frequency-domain centroid, a time-domain energy integral, a time-domain waveform index, a time-domain kurtosis coefficient, a time-domain skewness coefficient, a time-domain dispersion coefficient, a time-domain shape coefficient, and a time-domain offset frequency; An estimation module, configured to input the ultrasonic features into the battery SOC estimation model to obtain the SOC estimation value of the battery to be measured, where the battery SOC estimation model is used to accurately estimate the SOC value of each SOC interval of the battery; Among them, the process of generating the battery SOC estimation model includes: During the charge and discharge process of the sample battery, at different SOCs, obtaining the sample ultrasonic features corresponding to the ultrasonic response signal of the sample battery in a preset frequency range, where the sample battery is at least one; Training the original fully connected neural network FNN model according to the sample ultrasonic features to obtain an FNN model, where the FNN model is used to estimate the SOC value of the first SOC interval; Training the original gradient boosting XGBoost model according to the sample ultrasonic features and the failure features corresponding to the marked sample ultrasonic features, where the failure features are features that do not change with the battery SOC, and the XGBoost model is used to estimate the SOC value of the second SOC interval, and the second SOC interval is lower than the first SOC interval; Fusing the FNN model and the XGBoost model to obtain the battery SOC estimation model.
7. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 6.
8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method according to any one of claims 1 to 6.