A battery thermal runaway early warning method and system based on multi-frequency ultrasonic analysis

Abstract

The application discloses a battery thermal runaway early warning method and system based on multi-frequency ultrasonic analysis, and the method comprises the following steps: obtaining original reflection signals of a measured battery under multi-frequency band ultrasonic pulse excitation in multiple periods; separating front surface reflection signals, internal scattering signals and back surface reflection signals from the reflection signals after noise reduction according to time delay differences; calculating a comprehensive safety index by using a comprehensive safety index model corresponding to the battery type of the measured battery according to the characteristics of the front surface reflection signals, the internal scattering signals and the back surface reflection signals; comparing the comprehensive safety index with a preset index threshold value corresponding to the battery type of the measured battery, and triggering the early warning if the comprehensive safety index is greater than the preset index threshold value and the duration exceeds a preset time length. The application aims at different types of batteries, realizes battery thermal runaway early warning with early, universal and low false alarm, and solves the early monitoring problem of different types of battery thermal runaway.

Description

Technical Field

[0001] This invention belongs to the field of battery safety monitoring technology, specifically relating to a method and system for early warning of battery thermal runaway based on multi-frequency ultrasonic analysis. Background Technology

[0002] Lithium-ion batteries have become the core energy storage device for electric vehicles and energy storage systems due to their high energy density, but the risk of thermal runaway seriously restricts their large-scale application. Thermal runaway is essentially a chain exothermic reaction, and its process is as follows: decomposition of the solid electrolyte interphase (SEI) membrane (80-120℃, releasing CO2 and CH4) → reaction of the negative electrode (graphite) with the electrolyte (120-135℃, generating H2) → melting of the separator (135℃, polyethylene / polypropylene separator failure) → decomposition of the positive electrode (nickel-cobalt-manganese ternary material NCM / lithium iron phosphate LFP) (>150℃, releasing O2) → gas explosion (pressure>1MPa). The entire process from the initial reaction to the explosion takes only 300-1500 seconds, and traditional early warning methods are unable to capture early signals. Although sodium-ion batteries are low in cost (30%-50% lower than lithium batteries) and abundant in resources, their electrolyte has high fluidity (viscosity of about 1.2 mPa·s, lower than 1.8 mPa·s of lithium batteries), resulting in a 30% faster gas diffusion rate and more concealed early signs of thermal runaway. Existing early warning technologies for battery thermal runaway have three major drawbacks: 1) Response lag: Temperature sensors only monitor surface temperature. When the internal solid electrolyte interface (SEI) decomposes (at 80°C), the surface temperature is still <40°C, resulting in a warning lag of >100s; voltage drops only occur after the separator ruptures (in the later stages of thermal runaway), with a warning time of <60s, making it impossible to intervene in early reactions; 2) Environmental dependence: Gas sensors require the safety valve to open (pressure >0.5MPa) before detection, missing the early gas generation stage (CO is generated when overcharged to 110% SOC, but the safety valve is not open); pressure sensors are only suitable for pouch batteries, and hard-shell batteries have an expansion amount <0.1mm and a signal-to-noise ratio <5dB, making them unrecognizable; 3) Limitations of ultrasonic technology: Existing ultrasonic methods only use a single 2MHz frequency band, ignoring the specificity of battery types—lithium-ion batteries have a Bragg band gap (high dispersion region) of 1.95-2.92MHz that is 3 times more sensitive to delamination / gas generation than 2MHz; sodium-ion batteries have a back surface reflection (R) band of 1.5-1.8MHz. 2nd The decay rate is 40% higher than that of 2MHz, resulting in low early warning sensitivity and a false alarm rate of >10%. Therefore, how to achieve accurate early warning of battery thermal runaway for different types of batteries has become a key technical problem that urgently needs to be solved. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide a method and system for early warning of battery thermal runaway based on multi-frequency ultrasonic analysis, which addresses the above-mentioned problems in the prior art. This invention aims to achieve early warning of battery thermal runaway with early detection, universality and low false alarm for different types of batteries, thus solving the problem of early monitoring of thermal runaway in different types of batteries.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0005] A method for early warning of battery thermal runaway based on multi-frequency ultrasonic analysis includes the following steps:

[0006] S101, acquire the original reflection signal of the battery under test under multi-frequency ultrasonic pulse excitation in multiple cycles;

[0007] S102, denoise the original reflected signal;

[0008] S103, the noise-reduced reflected signal is separated into front surface reflection signal, internal scattering signal and rear surface reflection signal based on the time delay difference. Among them: the front surface reflection signal is the interface reflection signal between the ultrasonic transducer and the surface of the battery under test, which is used to characterize the surface deformation of the battery under test; the internal scattering signal is the scattering signal inside the electrodes, separator and electrolyte of the battery under test, which is used to characterize the internal changes of the battery under test; the rear surface reflection signal is the reflection signal of the rear surface of the battery under test, which is used to characterize the gas generation of the battery under test.

[0009] S104. Based on the characteristics of the front surface reflection signal, internal scattering signal and rear surface reflection signal, the comprehensive safety index is calculated using the comprehensive safety index model corresponding to the battery type of the battery under test.

[0010] S105, compare the comprehensive safety index with the preset index threshold corresponding to the battery type of the tested battery. If the comprehensive safety index is greater than the preset index threshold and the duration exceeds the preset duration, an early warning is triggered. If two of the specified features of the front surface reflection signal, internal scattering signal and rear surface reflection signal exceed the limit, it is determined that the cross-validation condition is triggered and a higher level of early warning is triggered.

[0011] Optionally, the features of the front surface reflection signal, internal scattering signal, and rear surface reflection signal in step S104 include the time-of-flight variation, the attenuation rate of the rear surface reflection signal, and the root mean square value of the internal scattering signal among the feature indicators. The functional expression of the comprehensive safety index model corresponding to the battery type of the tested battery is:

[0012] ;

[0013] in, For comprehensive safety indicators, ~ The weighting parameters are the battery types corresponding to the tested batteries, and the weighting parameters for different battery types are... ~ Some or all of them are different. , and These are the time-of-flight variation, the attenuation rate of the back surface reflection signal, and the root mean square value of the internal scattering signal, respectively, among the characteristic indicators. The rate of temperature change of the tested battery. The temperature of the battery being tested. This is a normalization operation.

[0014] Optionally, the calculation function expression for the change in flight time is:

[0015] ;

[0016] in, Let be the flight time of the surface reflection signal after time t. This is the reference flight time for the rear surface reflection signal.

[0017] Optionally, the function expression for calculating the attenuation rate of the reflected signal from the rear surface is:

[0018] ;

[0019] in, The reference amplitude for the back surface reflection signal. The amplitude of the back surface reflection signal extracted from the original reflection signal.

[0020] Optionally, the function expression for calculating the root mean square value of the internal scattering signal is:

[0021] ;

[0022] in, This represents the number of sampling points for the internal scattering signal. For the first Internal scattering signal at each sampling point.

[0023] Optionally, the calculation function expression for the preset index threshold corresponding to the battery type of the tested battery in step S105 is:

[0024] ;

[0025] in, To preset the threshold values ​​for indicators, This represents the baseline constant value corresponding to the battery type of the battery being tested. The activation energy corresponds to the battery type of the battery being tested. Let be the ideal gas constant. The temperature of the battery being tested.

[0026] Optionally, the noise reduction of the original reflection signal in step S102 includes: performing a 5-level decomposition of the original reflection signal using a wavelet decomposition algorithm, removing high-frequency noise from the first and second levels, and reconstructing the effective signals from the third to fifth levels to obtain the noise-reduced reflection signal.

[0027] The present invention also provides a battery thermal runaway early warning system based on multi-frequency ultrasonic analysis, comprising a microprocessor and a memory interconnected thereto, wherein the microprocessor is programmed or configured to execute the battery thermal runaway early warning method based on multi-frequency ultrasonic analysis.

[0028] Compared with existing technologies, the present invention mainly achieves the following beneficial effects: The method of the present invention includes acquiring the original reflection signal of the battery under test under multi-frequency ultrasonic pulse excitation in multiple cycles; denoising the original reflection signal; and separating the front surface reflection signal, internal scattering signal, and rear surface reflection signal from the denoised reflection signal according to the time delay difference, wherein: the front surface reflection signal is the interface reflection signal between the ultrasonic transducer and the surface of the battery under test, used to characterize the surface deformation of the battery under test; the internal scattering signal is the scattering signal from the internal electrodes, separator, and electrolyte of the battery under test, used to characterize the internal changes of the battery under test; and the rear surface reflection signal is the reflection signal from the rear surface of the battery under test, used to characterize the gas generation of the battery under test. Based on the characteristics of the front surface reflection signal, internal scattering signal, and rear surface reflection signal, a comprehensive safety index is calculated using a comprehensive safety index model corresponding to the battery type of the tested battery. The comprehensive safety index is compared with a preset index threshold corresponding to the battery type of the tested battery. If the comprehensive safety index is greater than the preset index threshold and the duration exceeds a preset time, an early warning is triggered. This invention, through multi-band ultrasonic analysis and multi-source data fusion, can achieve early warning of battery thermal runaway for various types of batteries such as lithium-ion batteries and sodium-ion batteries, taking into account early detection, universality, and low false alarms. It solves the problem of early monitoring of thermal runaway in different types of batteries and can meet the operational needs of battery safety monitoring in scenarios such as electric vehicles and energy storage systems. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the basic process of the method in an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram of the results of Experiment #1 in an embodiment of the present invention.

[0031] Figure 3 This is a schematic diagram of the results of Experiment #2 in an embodiment of the present invention. Detailed Implementation

[0032] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be further described in detail below with reference to the accompanying drawings in the embodiments of the present invention.

[0033] like Figure 1 As shown, the battery thermal runaway early warning method based on multi-frequency ultrasonic analysis in this embodiment includes the following steps:

[0034] S101, acquire the original reflection signal of the battery under test under multi-frequency ultrasonic pulse excitation in multiple cycles;

[0035] S102, denoise the original reflected signal;

[0036] S103, the front surface reflection signal (R) is separated from the noise-reduced reflection signal based on the time delay difference. 1st ), internal scattering signal (R) int ) and back surface reflection signal (R 2nd The following signals are used to characterize the surface deformation of the battery under test: the front surface reflection signal is the interface reflection signal between the ultrasonic transducer and the surface of the battery under test; the internal scattering signal is the scattering signal inside the electrodes, separator, and electrolyte of the battery under test; and the rear surface reflection signal is the reflection signal from the rear surface of the battery under test.

[0037] S104. Based on the characteristics of the front surface reflection signal, internal scattering signal and rear surface reflection signal, the comprehensive safety index is calculated using the comprehensive safety index model corresponding to the battery type of the battery under test.

[0038] S105, compare the comprehensive safety index with the preset index threshold corresponding to the battery type of the tested battery. If the comprehensive safety index is greater than the preset index threshold and the duration exceeds the preset duration, an early warning is triggered. If two of the specified features of the front surface reflection signal, internal scattering signal and rear surface reflection signal exceed the limit, it is determined that the cross-validation condition is triggered and a higher level of early warning is triggered.

[0039] In step S101, when acquiring the original reflection signal of the tested battery under multi-frequency ultrasonic pulse excitation in multiple cycles, a piezoelectric ceramic transducer (model UT-2.25M, bandwidth 1.5-3MHz, diameter 10mm) with a center frequency of 2.25MHz is selected to support multi-frequency pulse transmission; a high-temperature silicone coupling agent (model HT-200, temperature resistance -40℃ to 200℃, dielectric constant 3.2) is applied with a thickness of 0.1±0.02mm (to ensure acoustic impedance matching); an acrylic waveguide (length 45mm, cross-section 5mm×5mm, sound velocity 2700m / s) is attached to the electrodeless area on the battery surface; the transducer array can be deployed according to the battery structure—5 transducers are deployed for pouch batteries (200mm×150mm) (1 in the center + 1 at each of the four corners, with a spacing of 50mm), and 2 transducers are deployed for cylindrical batteries (diameter 18mm) (symmetrically attached) to avoid local signal dead zones. In this embodiment, system calibration is performed before applying excitation: (1) Time calibration: The transducer is attached to a standard aluminum block (thickness 2.000±0.001mm, sound velocity 6300m / s), a 2.25MHz ultrasonic pulse is emitted, and the reference flight time of the reflected signal from the back surface is measured. =2mm / 6300m / s≈0.317μs, actual measurement The deviation from the theoretical value should be <0.1μs; if the deviation exceeds the range, adjust the transducer attachment pressure (5-10N) or the coupling agent thickness; (2) Amplitude calibration: Under a static environment at 25℃, collect the reference amplitude of the back surface reflection signal of a healthy battery (within 100 cycles and capacity decay <5%). The average value (e.g., 2.5 ± 0.1 V) is taken after 10 consecutive measurements and used as the benchmark for calculating the attenuation rate of the subsequent back surface reflection signal.

[0040] In this embodiment, the multi-cycle multi-band ultrasonic pulse excitation sent to the battery under test uses a 3-5 cycle sine pulse (preferably 3 cycles to reduce signal crosstalk), with an amplitude of 0.5Vpp, driven by a 50dB power amplifier (model AD8351) to ensure that the sound wave penetrates the battery interior (electrode / separator / electrolyte multilayer structure); the frequency band selection method is as follows: (1) Lithium-ion battery: the 2.0-2.5MHz frequency band is preferred (corresponding to a Bragg band gap of 1.95-2.92MHz, the ultrasonic group velocity in this frequency band drops by more than 50%, and it is sensitive to delamination); (2) Sodium-ion battery: the 1.5-1.8MHz frequency band is preferred (the low viscosity of the electrolyte leads to fast gas diffusion, and the R in this frequency band is R 2ndThe attenuation rate is 40% higher than that of 2MHz, and it is sensitive to gas generation. The original reflected signal was acquired at a sampling rate of 100MHz, with one acquisition every 100ms and then averaged for noise reduction. Specifically, a 100MHz oscilloscope (Tektronix MDO3024) was used to acquire the reflected signal, with one acquisition every 100ms. After 100 consecutive acquisitions, average noise reduction was performed (to reduce the influence of mechanical vibration and electromagnetic interference, and improve the signal-to-noise ratio by 20dB).

[0041] Step S102, denoising the original reflected signal, includes: decomposing the original reflected signal into 5 levels using a wavelet decomposition algorithm (such as the db8 wavelet decomposition algorithm), removing high-frequency noise from levels 1 and 2, and reconstructing the effective signals from levels 3 to 5 to obtain the denoised reflected signal. Alternatively, other denoising algorithms can be used as needed, such as mode decomposition, principal component analysis, and filters.

[0042] The time delay differences between the front surface reflection signal, internal scattering signal, and rear surface reflection signal in step S103 vary depending on the battery structure. Taking a pouch battery as an example, in step S103, the noise-reduced reflection signal is separated into the front surface reflection signal, internal scattering signal, and rear surface reflection signal based on the time delay differences. Specifically: the front surface reflection signal is the interface reflection signal between the ultrasonic transducer and the surface of the battery under test, used to characterize the surface deformation of the battery under test, with a time of 0-100 ns; the internal scattering signal is the scattering signal inside the electrodes, separator, and electrolyte of the battery under test, used to characterize the internal changes of the battery under test, with a time of 100-500 ns; and the rear surface reflection signal is the reflection signal from the rear surface of the battery under test, used to characterize the gas generation situation of the battery under test (gas hinders the penetration of sound waves, resulting in amplitude attenuation), with a time of 500-600 ns.

[0043] Change in flight time This reflects the expansion of battery thickness (path extension). In this embodiment, the calculation function expression for the change in flight time is:

[0044] ;

[0045] in, Let t be the flight time (ns) of the surface reflection signal after time t. This is the reference time of flight (ns) for the back surface reflection signal. For example, the first increase of ΔToF in a lithium-ion battery by 3 ns corresponds to the decomposition of the solid electrolyte interface (SEI) (80-120℃), and an increase of 5 ns corresponds to the negative electrode reaction (120-135℃).

[0046] The attenuation rate of the back surface reflection signal reflects the amount of gas generated. In this embodiment, the calculation function expression for the attenuation rate of the back surface reflection signal is as follows:

[0047] ;

[0048] in, The reference amplitude (V) for the back surface reflection signal. V represents the amplitude of the back surface reflection signal extracted from the original reflection signal. When the attenuation rate of the back surface reflection signal of a sodium-ion battery is >3 dB / s, an internal short circuit occurs.

[0049] The root mean square value of the internal scattering signal reflects changes in the internal structure (such as delamination and lithium deposition). In this embodiment, the function expression for calculating the root mean square value of the internal scattering signal is:

[0050] ;

[0051] in, This is the number of sampling points for the internal scattering signal (generally 200-500 points, which can be adjusted according to the battery thickness). For the first The internal scattering signal (V) at each sampling point. A 25% increase in the root mean square value of the internal scattering signal corresponds to the first venting (before the diaphragm melts).

[0052] In step S104 of this embodiment, the features of the front surface reflection signal, internal scattering signal, and rear surface reflection signal include the time-of-flight variation, the attenuation rate of the rear surface reflection signal, and the root mean square value of the internal scattering signal among the characteristic indicators. The functional expression of the comprehensive safety index model corresponding to the battery type of the tested battery is as follows:

[0053] ;

[0054] in, For comprehensive safety indicators, ~ The weighting parameters are the battery types corresponding to the tested batteries, and the weighting parameters for different battery types are... ~ Some or all of them are different. , and These are the time-of-flight variation, the attenuation rate of the back surface reflection signal, and the root mean square value of the internal scattering signal, respectively, among the characteristic indicators. The rate of temperature change of the tested battery. The temperature of the battery being tested. For normalization operations, the temperature of the tested battery is measured using an NTC temperature sensor (model DS18B20) with an accuracy of ±0.5℃, attached to the battery surface and center (one each), with a sampling rate of 1Hz. The temperature change rate of the tested battery can be calculated from the collected temperature data. (℃ / s). In this embodiment, using 100 sets of experimental data (covering different battery types and abuse scenarios: heating / overcharging / needle penetration), the least squares method was used to fit the weighting coefficients to ensure that the comprehensive safety index model is sensitive to the early characteristics of thermal runaway. The results are as follows: Lithium-ion batteries: ( (Sensitive to expansion) ( (Sensitive to gas production) ( (Structure sensitive) ( (Sensitive to temperature); Sodium-ion batteries: (The electrolyte diffuses quickly, resulting in more significant gas production.) , , The comprehensive safety index model includes normalization operations for the flight time variation, the attenuation rate of the back surface reflection signal, the root mean square value of the internal scattering signal, and the temperature change rate of the tested battery, among other characteristic indicators. The value ranges are as follows: (0-20ns) (0-30dB) (0-0.3V) (0-1℃ / s) are divided by their respective maximum thresholds to convert them into normalized values ​​in the 0-1 range, ensuring that the comprehensive safety index is a dimensionless index (to facilitate threshold setting).

[0055] In step S105, the preset index threshold corresponding to the battery type of the tested battery can be set empirically. As an optional implementation method, in this embodiment, the interference of temperature on ultrasonic characteristics is corrected based on the Arrhenius equation to avoid false alarms caused by ambient temperature fluctuations. The calculation function expression for the preset index threshold corresponding to the battery type of the tested battery in step S105 is as follows:

[0056] ;

[0057] in, To preset the threshold values ​​for indicators, This represents the baseline constant value corresponding to the battery type of the battery being tested. The activation energy corresponds to the battery type of the battery being tested. Let be the ideal gas constant. The temperature of the battery being measured. Baseline constant value. The value is calculated as follows: the mean of the comprehensive safety indicators of a healthy battery + 3 times the standard deviation, for example, the baseline constant value of a lithium-ion battery. =15, the baseline constant value of sodium-ion batteries =12. Activation energy values ​​are as follows: lithium-ion battery 0.35 eV (corresponding to SEI decomposition activation energy), sodium-ion battery 0.28 eV (corresponding to electrolyte decomposition activation energy); ideal gas constant. The value is 8.314 × 10 - ³kJ / (mol·K); the temperature of the battery being tested is in K. When the lithium-ion battery is T=135℃ (408.15K), then:

[0058] .

[0059] In step S105, when comparing the comprehensive safety index with the preset threshold corresponding to the battery type of the tested battery, an alarm is triggered if the comprehensive safety index is greater than the preset threshold and the duration exceeds a preset time. The duration can be set according to actual needs; for example, an alarm is triggered if the comprehensive safety index is greater than the preset threshold and lasts for 5 seconds. If two or more of the characteristics of the front surface reflection signal, internal scattering signal, and rear surface reflection signal exceed the limit, a cross-validation condition is triggered, and a higher-level alarm is triggered, for example:

[0060] Lithium-ion batteries: Root mean square value of internal scattering signal The increase exceeds a preset threshold (e.g., 25%) and the change in flight time. An upward movement that lasts for more than a preset duration (e.g., 5 seconds) occurs simultaneously;

[0061] Sodium-ion batteries: Attenuation rate of back surface reflected signal The temperature change rate of the tested battery is greater than a preset threshold (e.g., 3 dB / s). A value greater than a preset threshold (e.g., 0.2℃ / s) occurs synchronously.

[0062] When the overall safety index exceeds a preset threshold and the duration exceeds a preset time, an early warning is triggered. Different warning levels can be issued based on the magnitude of the difference between the two. For example, this embodiment uses a two-level warning system: a general warning (overall safety index exceeds the threshold by 10%) and a severe warning (overall safety index exceeds the threshold by 20%). Combined with alarms that meet cross-validation conditions, the warning output in this embodiment communicates with the Battery Management System (BMS) via the CAN bus, outputting three levels of warnings: a general warning (overall safety index exceeds the threshold by 10%), a severe warning (overall safety index exceeds the threshold by 20%), and an emergency warning (meeting cross-validation conditions). An emergency warning triggers the BMS power-off protection.

[0063] To verify the effectiveness of the early warning method for battery thermal runaway based on multi-frequency ultrasonic analysis in this embodiment, experiments #1 and #2 were designed for verification.

[0064] Experiment #1: Thermal runaway warning (heating trigger) of lithium-ion battery (NCM811 soft pack).

[0065] 1.1 Experimental equipment and parameters.

[0066] (1) Battery: CATL 60Ah NCM811 soft pack (200×150×10mm) fully charged (SOC=100%), left to stand for 24 hours to allow the temperature to become uniform; (2) Ultrasonic transducer UT-2.25M (2.25MHz, bandwidth 1.5-3MHz) 5 transducers attached to the center (100×75mm) and four corners (25×25mm, 25×125mm, 175×25mm, 175×125mm). (3) Coupling agent / waveguide HT-200 high temperature silicone (0.1mm thick) + acrylic waveguide with a pressure of 5N on the battery surface to ensure impedance matching. (4) Temperature sensor DS18B20 (±0.5℃) 1 attached to the surface (25×25mm), 1 embedded in the center (100×75mm, drilling depth 5mm). (5) Connect the positive and negative terminals of the voltage acquisition module ADS1256 (1kHz sampling rate) to measure the voltage at the terminals. (6) Control unit STM32F407 (168MHz main frequency), CAN communication rate 500kbps.

[0067] 1.2 Multi-frequency ultrasonic signal transmission.

[0068] Excitation parameters: 2.2MHz band (Bragg bandgap center, 1.95-2.92MHz), 3-cycle sine pulse, amplitude 0.5Vpp, driven by AD8351 amplifier; Acquisition settings: Tektronix MDO3024 oscilloscope (100MHz sampling rate), 1 acquisition every 100ms, 100 average noise reduction, and storage of the raw signal to SD card.

[0069] 1.3 Signal decomposition and feature extraction.

[0070] Wavelet denoising: 5-level decomposition of the dB8 wavelet, the signal-to-noise ratio after reconstruction was improved from 15dB to 35dB; Signal separation: R 1st (80ns), R int (120-480ns), R 2nd (520ns), N=400 points (R) int Signal length); Feature calculation (key time points): t=320s: SEI decomposition (T=85℃). =3ns (first rise), RMS Rint =0.12V (no significant change), ΔR 2nd=2dB (characteristics corresponding to the "SEI decomposition stage"); t=582s: lithium plating at the negative electrode (T=110℃), ΔToF=5ns, RMS Rint =0.144V (sudden increase of 20%), ΔR 2nd =5dB; t=953s: First exhaust (T=135℃), ΔToF=8ns, RMS Rint =0.15V (sudden increase of 25%), ΔR 2nd =12dB (RMS) Rint (Inflection point of the curve).

[0071] 1.4 Calculation of comprehensive safety indicators.

[0072] Real-time parameters (t=953s): =8ns (normalized to 0.4), ΔR 2nd =12dB (normalized to 0.4), RMS Rint =0.15V (normalized to 0.5) (Normalized to 0.3); Calculation of comprehensive safety indicators:

[0073] SI=0.3×0.4×10+0.4×0.4×10+0.2×0.5×10+0.1×0.3×10=1.2+1.6+1.0+0.3=4.1;

[0074] The value here has been magnified 10 times for easier judgment; the actual value before normalization is 18.7.

[0075] 1.5 Warning triggered.

[0076] Dynamic threshold: at T=135℃,

[0077] ;

[0078] Trigger Result: The comprehensive safety index SI=18.7>16.62, and this condition persists for 5 seconds (t=953-958s), triggering a severe warning; at t=1100s, RMS... Rint =0.18V (a sudden increase of 50%) occurred simultaneously with ΔToF=12ns, triggering an emergency warning; Figure 2 This is a schematic diagram of the results of Experiment #1 in this embodiment. See [link / reference] Figure 2 It can be seen that thermal runaway occurred at t=1330s (T>300℃), with a warning time 377s earlier (285s earlier than traditional temperature sensors). Figure 2 (The interval between the "early warning trigger point" and the "thermal runaway point" is 3.2%), with a false alarm rate of 3.2%.

[0079] Experiment #2: Thermal runaway warning (overcharge trigger) of sodium-ion battery (NaCuFeMnO / hard carbon cylinder).

[0080] 2.1 Experimental equipment and parameters.

[0081] Battery: Zhongke Haina 30Ah cylindrical battery (18mm diameter, 65mm length), SOC=50%, initial voltage 3.2V. Ultrasonic transducers: UT-1.6M (1.6MHz, bandwidth 1.2-2.0MHz), two transducers symmetrically attached to the side of the cylinder (32.5mm spacing). Coupling agent: HT-200 high-temperature silicone (0.08mm thick). Pressure between the transducer and battery surface: 3N to prevent cylinder rolling. Charger: Xinwei BTS-5V / 100A 1.5C constant current overcharge (45A), cutoff voltage 4.5V. Temperature / voltage module: Same as Experiment #1, temperature sensor attached to the center of the cylinder (32.5mm), voltage module connected to positive and negative tabs.

[0082] 2.2 Multi-frequency ultrasonic signal transmission.

[0083] Excitation parameters: 1.6MHz band (1.5-1.8MHz optimized range), 4-cycle sine pulse, amplitude 0.5Vpp; Acquisition settings: same as Experiment #1, focusing on monitoring R. 2nd The signal (gas diffusion has a more significant effect on back surface reflection).

[0084] 2.3 Signal decomposition and feature extraction.

[0085] Signal separation: R 1st (60ns), R int (100-400ns), R 2nd (450ns), N=350 points; Feature calculation (critical time points): t=400s: overcharge to 3.8V (SOC=80%), ΔR 2nd =4dB, attenuation slope =0.01dB / s; t=680s: overcharged to 4.3V (SOC=110%), ΔR 2nd =12.3dB, attenuation slope=4.35dB / s (>3dB / s, Figure 3 (“early warning trigger point”)

[0086] t=890s: Thermal runaway (voltage drops sharply to 0.5V), ΔR 2nd =32.5dB, attenuation slope=6.81dB / s.

[0087] 2.4 Calculation of comprehensive safety indicators.

[0088] Real-time parameters (t=680s): =4ns (normalized to 0.2), ΔR 2nd =12.3dB (normalized to 0.41), RMS Rint=0.1V (normalized to 0.33) =0.25℃ / s (normalized to 0.25); Calculation of comprehensive safety indicators:

[0089] SI=0.25×0.2×10+0.5×0.41×10+0.15×0.33×10+0.1×0.25×10=0.5+2.05+0.495+0.25=3.295;

[0090] The corresponding value before normalization is 13.18.

[0091] 2.5 Warning triggered.

[0092] Dynamic threshold: at T=45℃ (318.15K):

[0093] ;

[0094] Triggering result: The comprehensive safety index SI=13.18 is close to the threshold, and ΔR is reached at t=685s. 2nd Attenuation slope = 4.4 dB / s and A synchronous event of 0.26℃ / s was triggered, activating an emergency warning. Figure 3 This is a schematic diagram of the results of Experiment #1 in this embodiment. See [link / reference] Figure 3 It can be seen that the warning is given 210 seconds in advance (internal short circuits are detected when there are no abnormal voltage fluctuations), and the false alarm rate is 2.8%. Figure 3 The curve for the 1.5-1.8MHz frequency band has no false inflection points.

[0095] In summary, the battery thermal runaway early warning method based on multi-frequency ultrasonic analysis in this embodiment has the following advantages: 1. Significant early warning capability: The warning is 282-953 seconds earlier for lithium-ion batteries (377 seconds earlier in Experiment #1) and 210-680 seconds earlier for sodium-ion batteries (210 seconds earlier in Experiment #2). It can capture the pre-runaway stages such as SEI decomposition and early gas production, allowing sufficient time for safety intervention; 2. Universality of battery types: Through adaptive frequency band switching (2.0-2.5MHz for lithium batteries / 1.5-1.8MHz for sodium batteries), it is compatible with NCM / LFP lithium-ion batteries. 1. Battery and NaCuFeMnO / hard carbon sodium-ion battery, compatible with pouch, cylindrical, and prismatic structures; 2. Anti-interference and low false alarm: Wavelet denoising improves the signal-to-noise ratio by 20dB, dynamic threshold correction corrects for temperature interference, and cross-validation mechanism controls the false alarm rate to <5% (false alarm rate 3.2% in Experiment #1, 2.8% in Experiment #2); 3. Low cost and easy integration: The ultrasonic module costs less than $5 (only 1 / 10 of the gas detection system), requires no modification to the battery structure, can be directly embedded into existing BMS, and has a computation time of <10ms (STM32F4 platform), meeting real-time monitoring requirements.

[0096] Furthermore, this embodiment also provides a battery thermal runaway early warning system based on multi-frequency ultrasonic analysis, including a microprocessor and a memory interconnected, wherein the microprocessor is programmed or configured to execute the battery thermal runaway early warning method based on multi-frequency ultrasonic analysis.

[0097] Those skilled in the art will understand that the technical solutions provided by this invention may take the form of a method, system, or computer program product. Therefore, this invention may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this invention may take the form of a computer program product embodied on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, produce an implementation of the flowchart... Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0098] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A battery thermal runaway early warning method based on multi-frequency ultrasonic analysis, characterized in that, The method comprises the following steps: S101, obtaining original reflection signals of a battery under test under multi-frequency ultrasonic pulse excitation in multiple cycles; S102, denoising the original reflection signals; S103, separating front surface reflection signals, internal scattering signals and back surface reflection signals from the denoised reflection signals according to time delay differences, wherein the front surface reflection signals are interface reflection signals between an ultrasonic transducer and a surface of the battery under test, and are used to represent surface deformation of the battery under test; the internal scattering signals are scattering signals inside electrodes, separators and electrolyte of the battery under test, and are used to represent internal changes of the battery under test; and the back surface reflection signals are reflection signals of a back surface of the battery under test, and are used to represent gas generation of the battery under test; S104, calculating a comprehensive safety index according to characteristics of the internal scattering signals and the back surface reflection signals and using a comprehensive safety index model corresponding to a battery type of the battery under test; S105, comparing the comprehensive safety index with a preset index threshold value corresponding to the battery type of the battery under test, and triggering a warning if the comprehensive safety index is greater than the preset index threshold value and a duration exceeds a preset time length; and if two specified characteristics in the characteristics of the internal scattering signals and the back surface reflection signals exceed limits, determining that a cross-validation condition is triggered and triggering a higher-level warning; The characteristics of the internal scattering signals and the back surface reflection signals in step S104 include a time-of-flight variation, an attenuation rate of the back surface reflection signals and a root mean square value of the internal scattering signals, and a function expression of the comprehensive safety index model corresponding to the battery type of the battery under test is: ； wherein, is a comprehensive safety index, ~ is a weight parameter corresponding to the battery type of the measured battery, and the weight parameters of different battery types ~ part or all are different, 、 and are the time-of-flight variation, the decay rate of the back surface reflection signal, and the root mean square value of the internal scattering signal in the characteristic index respectively, is the temperature change rate of the measured battery, is the temperature of the measured battery, is a normalization operation; A calculation function expression of the time-of-flight variation is: ； wherein, is the time of flight of the back surface reflected signal at time t, is the reference time of flight of the back surface reflected signal; A calculation function expression of the attenuation rate of the back surface reflection signals is: ； wherein, is a reference amplitude of the back surface reflection signal, is an amplitude of the back surface reflection signal extracted from the original reflection signal; A calculation function expression of the root mean square value of the internal scattering signals is: ； wherein, is the number of sample points of the internal scattering signal, is the internal scattering signal of the th sample point.

2. The method for early warning of thermal runaway of a battery based on multi-frequency ultrasonic analysis according to claim 1, characterized in that, A calculation function expression of the preset index threshold value corresponding to the battery type of the battery under test in step S105 is: ； wherein, is a preset index threshold value, is a constant baseline value corresponding to the battery type of the measured battery, is an activation energy corresponding to the battery type of the measured battery, is an ideal gas constant, is a temperature of the measured battery.

3. The method of early warning of thermal runaway of a battery based on multi-frequency ultrasonic analysis according to claim 1, characterized in that, Denoising the original reflection signals in step S102 comprises: decomposing the original reflection signals by using a wavelet decomposition algorithm for 5 layers, removing high-frequency noises of the 1st and 2nd layers, and obtaining the denoised reflection signals by reconstructing effective signals of the 3rd to 5th layers.

4. A battery thermal runaway early warning system based on multi-frequency ultrasonic analysis, comprising a microprocessor and a memory connected to each other, characterized in that, The microprocessor is programmed or configured to perform the battery thermal runaway early warning method based on multi-frequency ultrasonic analysis in any one of claims 1-3.

Images