A three-dimensional imaging method of ultrasonic phased array batteries based on wave velocity correction

Abstract

The present invention relates to the technical field of non-destructive testing of lithium-ion batteries, and provides a method for three-dimensional imaging of batteries using an ultrasonic phased array based on wave velocity correction. First, the corresponding sound velocities of ultrasonic waves propagating in different directions in the battery are obtained using the full matrix time domain data of the phased array body wave. The corresponding sound velocities are then used in the full focusing method to perform phased array imaging with sound velocity correction, thereby obtaining more accurate imaging results. Finally, the position of the phased array is moved by a robotic arm and a stepper motor, and the collected ultrasonic B-scan image is reconstructed into a three-dimensional image. The results show that this method can obtain more accurate and specific battery defect imaging results compared to traditional ultrasonic phased array imaging due to the realization of sound velocity correction, and the defect location is clearer under the same display threshold. In addition, this method applies wave velocity correction to three-dimensional imaging of batteries, providing a solution to the lack of economical, fast and accurate imaging methods in battery defect detection.

Description

Technical Field

[0001] The present invention relates to the technical field of nondestructive testing of lithium-ion batteries, and more particularly to a three-dimensional imaging method of an ultrasonic phased array battery based on wave velocity correction. Background Art

[0002] Lithium-ion batteries are widely used in consumer electronics, new energy vehicles, and energy storage systems. The integrity of their internal structures is directly related to battery performance and safety. During manufacturing, transportation, and use, battery cells are often subjected to mechanical stress, electrochemical reactions, or thermal stress, which can easily lead to internal structural defects such as wrinkling, delamination, foreign matter, and punctures. Severe defects can cause thermal runaway, fire, or even explosion. Therefore, research on high-precision, non-destructive lithium battery internal defect detection technology has important engineering applications.

[0003] Currently, traditional inspection methods such as X-ray imaging and CT, while offering high resolution, suffer from limitations such as radiation exposure, high cost, and poor portability, making them difficult to meet the practical needs of online battery testing and batch screening. In contrast, ultrasonic phased array technology, with its high penetration, high sensitivity, and excellent imaging capabilities, is particularly well-suited for defect detection in layered structures such as pouch cells.

[0004] However, the complex internal structure and strong anisotropy of batteries, along with the variable ultrasonic propagation paths, present challenges for traditional imaging methods. For example, the total focusing method (TFM) is prone to artifacts and image defocus when assuming a constant sound velocity, resulting in reduced defect detection accuracy. Furthermore, the relatively compact structure of lithium batteries leads to high inter-channel ultrasonic signal noise, making traditional single-shot imaging data susceptible to interference, which can affect imaging stability.

[0005] To improve imaging accuracy, it is urgent to develop an imaging method that can adaptively correct for changes in sound velocity across multi-angle propagation paths. Existing research has limited research on angular sound velocity fitting and fine-tuned imaging techniques based on FMC data. Techniques that maintain detection resolution while improving data utilization efficiency remain underdeveloped.

[0006] In addition, two-dimensional B-scan imaging cannot fully reflect the spatial morphology of complex defects. It is necessary to combine the mechanical scanning system to perform high-density stitching and three-dimensional reconstruction of the B-scan images to achieve spatial visualization of defects within the battery volume, providing a reliable basis for refined defect classification and severity assessment.

[0007] Therefore, developing a comprehensive detection method that combines ultrasonic phased array FMC data processing, adaptive sound velocity correction, TFM enhanced imaging and robotic arm three-dimensional reconstruction is of great significance for improving the accuracy and intelligence level of lithium battery defect detection. Summary of the Invention

[0008] In order to overcome the above-mentioned defects of the prior art, the present invention provides an ultrasonic phased array battery three-dimensional imaging method based on wave velocity correction. To address the image defocus problem caused by sound velocity changes in lithium battery detection and the limitation that two-dimensional imaging cannot reflect the true spatial morphology of defects, the method calculates multi-angle sound velocity values ​​by analyzing the full-matrix time domain data, and constructs an angle-sound velocity distribution function based on this. The sound velocity correction is performed in combination with full-focus imaging, and the two-dimensional image stitching and three-dimensional reconstruction are realized in conjunction with a robotic arm, thereby realizing multi-dimensional detection that integrates sound velocity adaptive modeling and high-density mechanical scanning, and improving the accuracy and spatial resolution capability of battery defect detection.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A three-dimensional imaging method for an ultrasonic phased array battery based on wave velocity correction includes the following steps:

[0011] Step S1, performing ultrasonic phased array scanning on the battery to collect full matrix time domain data;

[0012] Step S2, based on the full matrix time domain data, obtaining the ultrasonic propagation time at different angles through generalized cross-correlation operation; using the relative coordinate positions of the ultrasonic phased array elements and the imaging point, calculating the ultrasonic propagation distance at each angle;

[0013] Step S3, calculating the ultrasonic sound velocity value at the corresponding angle based on the propagation distance and propagation time at each angle, and constructing the angle-sound velocity distribution function through interpolation fitting;

[0014] Step S4, introducing the angle-sound velocity distribution function into the total focusing method to perform sound velocity correction imaging to obtain a two-dimensional B-scan image after sound velocity correction, includes the following steps:

[0015] Step S41, for each imaging point, respectively calculating the excitation propagation distance and excitation angle between the imaging point and the excitation array element, and the receiving propagation distance and receiving angle between the imaging point and the receiving array element;

[0016] Step S42, obtaining the corresponding excitation sound velocity and receiving sound velocity from the angle-sound velocity distribution function according to the excitation angle and the receiving angle, and calculating the propagation time of the imaging point based on the excitation sound velocity, the receiving sound velocity, the excitation propagation distance, and the receiving propagation distance;

[0017] Step S43: extracting the signal value at the corresponding time point from the corresponding full matrix time domain data according to the propagation time, performing Hilbert transform on the signal value, and extracting the envelope value as the grayscale value of the pixel corresponding to the imaging point;

[0018] Step S44, repeating the above steps S41-S43 for all imaging points, and combining the pixel grayscale values ​​of each imaging point into a two-dimensional B-scan image after sound velocity correction;

[0019] Step S5 , by moving the position of the phased array probe, collecting continuous two-dimensional B-scan images and performing three-dimensional reconstruction, to obtain a three-dimensional imaging result of the battery.

[0020] As a further solution of the present invention, in step S2, obtaining the ultrasonic wave propagation time at different angles by generalized cross-correlation operation includes the following steps:

[0021] Step A1: Superimpose and average the ultrasonic signals with the same excitation angle and reception angle in the full matrix time domain data. That is, directly add the ultrasonic signals with the same excitation angle and reception angle in the time domain, and then take the average value to obtain the averaged ultrasonic signal to reduce the influence of noise;

[0022] Step A2: define the ultrasonic signal with an excitation angle of 0° as the reference signal, perform Hilbert transform on it, and extract the time corresponding to the maximum value of the bottom echo signal as the reference ultrasonic propagation time;

[0023] In step A3, a generalized cross-correlation operation is performed on the ultrasonic signals of the remaining excitation angles based on the reference ultrasonic propagation time and the reference signal. The cross-correlation function of each target signal and the reference signal is calculated to determine the time offset value corresponding to its peak value. Finally, the time offset value is added to the reference ultrasonic propagation time to calculate the propagation time of each angle.

[0024] As a further solution of the present invention, in step S3, the angle-sound speed distribution function is constructed by performing cubic interpolation fitting on the sound speed at each angle. The angle-sound speed distribution function is used to estimate the sound speed at any angle, which facilitates the subsequent sound speed correction of the total focusing method.

[0025] As a further solution of the present invention, in step S42, the propagation time is obtained by dividing the excitation propagation distance by the excitation sound speed to obtain the excitation propagation time, and dividing the receiving propagation distance by the receiving sound speed to obtain the receiving propagation time, and then adding the excitation propagation time and the receiving propagation time.

[0026] As a further solution of the present invention, the moving of the phased array probe is achieved by using a robotic arm and a stepping motor to drive the phased array probe to move, and the phased array probe is fixed to the end of the robotic arm.

[0027] As a further solution of the present invention, the ultrasonic signal superposition and averaging processing is performed only on channel pairs with equal excitation angles and receiving angles, so as to suppress non-structural noise and retain structural response characteristics.

[0028] As a further solution of the present invention, the generalized cross-correlation operation includes: performing cross-correlation processing on the ultrasonic signals of the remaining excitation angles and the reference signal, and determining the time delay when the cross-correlation function reaches a maximum value as the time offset value; the generalized cross-correlation operation has the characteristics of strong noise resistance and high time resolution, and is used to improve the accuracy of sound velocity measurement.

[0029] As a further solution of the present invention, the robotic arm and the stepper motor cooperate to control the phased array probe to scan at equal intervals of 0.25 mm in a direction perpendicular to the imaging plane where the two-dimensional B-scan image is located, and spatially align and superimpose the collected continuous two-dimensional B-scan images to reconstruct the three-dimensional spatial distribution of the defect.

[0030] As a further solution of the present invention, the battery is a lithium-ion soft-pack battery, and the three-dimensional imaging results are used to simultaneously detect wrinkling defects, delamination defects, foreign matter defects and puncture defects inside the battery.

[0031] Compared with the prior art, the ultrasonic phased array battery three-dimensional imaging method based on wave velocity correction of the present invention has the following beneficial effects:

[0032] This method uses multi-angle, full-matrix time-domain data to construct a sound velocity angular distribution function and incorporates it into a full-focusing algorithm for adaptive sound velocity correction. Compared to traditional imaging methods that assume a constant sound velocity, this technical solution dynamically adjusts the sound velocity parameters based on the ultrasonic propagation angle, resolving image blur and artifact issues. Furthermore, a generalized cross-correlation method is employed to accurately estimate propagation time. Compared to traditional time-domain analysis methods, this method significantly improves measurement accuracy without increasing hardware complexity, laying the technical foundation for high-precision defect detection.

[0033] This invention uses a precise robotic arm to achieve 3D reconstruction of high-density B-scan images, fully displaying the spatial morphology, depth, and boundary contours of defects. Compared to traditional 2D imaging, which only provides cross-sectional information, this solution overcomes the limitations of 2D images in representing complex defects, providing a reliable, three-dimensional basis for refined defect classification and severity assessment.

[0034] The present invention has outstanding versatility and scalability. It is not only suitable for the structural non-destructive testing of various types of lithium batteries, but also has good industrial application prospects. More importantly, the proposed adaptive sound velocity modeling mechanism provides new ideas for ultrasonic testing of other structurally complex materials (such as composite materials and laminated structures). BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic flow chart of a method for three-dimensional imaging of an ultrasonic phased array battery based on wave velocity correction provided by the present invention.

[0036] Figure 2 Schematic diagram of the experimental device provided by the present invention.

[0037] Figure 3 This is a schematic diagram of the ultrasonic signal after averaging and noise reduction provided by the present invention.

[0038] Figure 4 This is a schematic diagram of the angle-sound velocity distribution function provided by the present invention.

[0039] Figure 5 Schematic diagram of FMC-TFM ultrasonic B-scan imaging with sound velocity correction provided by the present invention.

[0040] Figure 6 Schematic diagram of FMC-TFM ultrasonic three-dimensional imaging with sound velocity correction provided by the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] Example 1

[0043] A three-dimensional imaging method for an ultrasonic phased array battery based on wave velocity correction includes the following steps:

[0044] First, a lithium-ion soft-pack battery was selected as the test object, and a 64-element linear phased array probe was used to perform a single full-matrix time domain data acquisition on the bottom surface of the battery. The acquisition frequency was set to 5 MHz and the sampling frequency was set to 50 MHz to ensure high-quality ultrasonic signal data. Figure 2 This is a schematic diagram of the experimental setup, fully illustrating the configuration of the phased array system, coupling medium, and battery specimen. Full-matrix capture is a data acquisition method that controls every element in the phased array probe to function as both a transmitter and a receiver, thereby obtaining the most complete dataset possible, including information about all acoustic wave propagation paths.

[0045] After the acquisition is completed, in order to reduce the interference of random noise, the original data needs to be preprocessed to improve the data quality. The preprocessing step is specifically: superimpose and average the ultrasonic signals of only the channel pairs with equal excitation angles and receiving angles in the full matrix time domain data. Because the real defect echo signal is stable, and the random noise is chaotic, by taking the average value, the real signal will be enhanced, and the noise will cancel each other out. This processing method can effectively reduce the impact of system thermal noise and random interference on imaging quality, while retaining effective structural response information. The processed signal is as follows Figure 3 As shown in the figure, the noise amplitude is significantly reduced compared with the original signal, which lays a good foundation for subsequent sound speed calculation and imaging processing.

[0046] After preprocessing, the sound velocity calculation phase begins, where the angle-sound velocity distribution function is constructed. First, the ultrasonic signal with an averaged excitation angle of 0° (perpendicular to the bottom surface of the battery) is defined as the reference signal. A Hilbert transform is then performed. The signal envelope is extracted using the Hilbert transform, and the time point corresponding to the maximum echo envelope value is precisely located and recorded as the reference ultrasonic propagation time. The Hilbert transform is a signal processing technique that extracts the instantaneous amplitude or envelope of the signal, enabling more stable determination of the echo peak location.

[0047] Using this reference ultrasonic propagation time and reference signal as a benchmark, a generalized cross-correlation operation is performed on the ultrasonic signals at the remaining excitation angles in the full-matrix time-domain data. Specifically, the MATLAB function xcorr() is used to calculate the cross-correlation function between the target signal (the ultrasonic signal at the remaining excitation angles) and the reference signal (the ultrasonic signal at an excitation angle of 0°). This cross-correlation function is calculated by time-shifting the waveform of the target signal, multiplying it with the waveform of the reference signal point by point at all time points, and then summing the results. This calculation obtains the time offset value corresponding to the peak of the cross-correlation function and defines it as the relative time delay. This time offset value is then added to the reference ultrasonic propagation time to calculate the actual propagation time for each angle.

[0048] After obtaining the propagation time at each angle, combined with the known propagation distance, the corresponding sound speed value at each angle is calculated based on the physical relationship that the speed of sound is equal to the propagation distance divided by the propagation time.

[0049] Through the above steps, the ultrasonic propagation parameter data at different angles were successfully obtained, as shown in Table 1.

[0050] Table 1 Ultrasonic propagation parameter data at different angles

[0051]

[0052] Then, a cubic interpolation fit is performed on the sound velocity values ​​at each angle to construct a complete angle-sound velocity distribution function. The cubic interpolation method can generate a smooth curve between discrete data points, thereby estimating the sound velocity value at any intermediate angle. The angle-sound velocity distribution function can be used to estimate the sound velocity at any angle. Based on the ultrasonic propagation parameter data at different angles, the cubic interpolation fitting coefficient parameters of the angle-sound velocity distribution function are fitted, as shown in Table 2.

[0053] Table 2 Cubic interpolation fitting coefficients of angle-sound velocity distribution function

[0054]

[0055] Among them, a is the coefficient of the cubic term, b is the coefficient of the quadratic term, c is the coefficient of the linear term, and d is the constant term.

[0056] The angle-sound velocity distribution function is realized by a cubic polynomial function, whose structure is: the sound velocity value is equal to the cubic coefficient multiplied by the cube of the angle value, plus the quadratic coefficient multiplied by the square of the angle value, plus the linear coefficient multiplied by the angle value, and finally the constant term. Figure 4 As shown, the interpolation results show good smoothness and conform to physical laws.

[0057] After constructing the function, full-focus imaging with sound velocity correction can be performed. Traditional full-focus algorithms usually assume that the sound velocity is a constant when calculating the ultrasonic propagation time. This simplified assumption can easily lead to focusing errors when facing the complex structure inside the battery, thereby affecting the imaging quality. In order to solve this problem, the process of the embodiment of the present invention introduces the angle-sound velocity distribution function obtained by fitting in the above steps into the full-focus imaging process, performs sound velocity correction on each imaging point in the imaging area, and finally obtains a two-dimensional B-scan image. The specific correction imaging steps include: for each imaging point, first calculate the propagation angle and propagation distance between the point and each excitation array element and receiving array element; then query and obtain the corresponding sound velocity value from the constructed angle-sound velocity distribution function, divide the excitation propagation distance by the corresponding excitation sound velocity, and add the receiving propagation distance by the corresponding receiving sound velocity, and then calculate the accurate propagation time; then for this specific time point, extract the corresponding full-matrix capture signal and perform Hilbert transform, and use the extracted envelope value as the pixel grayscale value of the imaging point; finally, repeat the above processing for all imaging points, and finally obtain a high-quality B-scan image after sound velocity correction, such as Figure 5 shown.

[0058] In order to obtain a three-dimensional image, it is necessary to collect multiple continuous two-dimensional B-scan images by moving the phased array probe. In an embodiment of the present invention, the phased array probe is fixed to the end of the robotic arm, and cooperates with the stepper motor control system to enable the probe to move accurately in a straight line along the Y direction. The interval between each movement is strictly controlled to 0.25 mm to ensure that the scanning density meets the accuracy requirements of three-dimensional reconstruction. At each scanning position, B-scan data acquisition is repeated to obtain a continuous two-dimensional tomographic image sequence. Subsequently, all the collected two-dimensional B-scan images are spatially aligned and orderly stacked. In this way, three-dimensional grayscale data reflecting the internal defect distribution of the battery is constructed, and integrated drawing technology is used for three-dimensional visualization display, ultimately achieving three-dimensional presentation of defects, such as Figure 6 shown.

[0059] To validate the effectiveness of this method, defect imaging experiments were conducted on various battery samples and compared with a traditional full-focusing algorithm. The experiments demonstrated that the use of the acoustic velocity-corrected full-focusing method improved image clarity by approximately 30% and reduced boundary recognition error to below 0.2 mm, significantly improving defect detection accuracy. Furthermore, the 3D reconstructed images intuitively displayed the shape, volume, and distribution depth of internal defects, validating the feasibility and practicality of this method for complex structure inspection.

[0060] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0061] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A three-dimensional imaging method of an ultrasonic phased array battery based on wave velocity correction, characterized in that: The following steps are involved: Step S1, performing ultrasonic phased array scanning on the battery to collect full matrix time domain data; Step S2, based on the full matrix time domain data, obtaining the ultrasonic propagation time at different angles through generalized cross-correlation operation; using the relative coordinate positions of the ultrasonic phased array elements and the imaging point, calculating the ultrasonic propagation distance at each angle; Step S3, calculating the ultrasonic sound velocity value at the corresponding angle based on the propagation distance and propagation time at each angle, and constructing the angle-sound velocity distribution function through interpolation fitting; Step S4, introducing the angle-sound velocity distribution function into the total focusing method to perform sound velocity correction imaging to obtain a two-dimensional B-scan image after sound velocity correction, includes the following steps: Step S41, for each imaging point, respectively calculating the excitation propagation distance and excitation angle between the imaging point and the excitation array element, and the receiving propagation distance and receiving angle between the imaging point and the receiving array element; Step S42, obtaining the corresponding excitation sound velocity and receiving sound velocity from the angle-sound velocity distribution function according to the excitation angle and the receiving angle, and calculating the propagation time of the imaging point based on the excitation sound velocity, the receiving sound velocity, the excitation propagation distance, and the receiving propagation distance; Step S43: extracting the signal value at the corresponding time point from the corresponding full matrix time domain data according to the propagation time, performing Hilbert transform on the signal value, and extracting the envelope value as the grayscale value of the pixel corresponding to the imaging point; Step S44, repeating the above steps S41-S43 for all imaging points, and combining the pixel grayscale values ​​of each imaging point into a two-dimensional B-scan image after sound velocity correction; Step S5 , by moving the position of the phased array probe, collecting continuous two-dimensional B-scan images and performing three-dimensional reconstruction, to obtain a three-dimensional imaging result of the battery.

2. The ultrasonic phased array battery three-dimensional imaging method based on wave velocity correction according to claim 1 is characterized in that: In step S2, obtaining the ultrasonic wave propagation time at different angles by generalized cross-correlation calculation includes the following steps: Step A1, directly adding the ultrasonic signals with the same excitation angle and receiving angle in the full matrix time domain data in the time domain, and then taking the average value to obtain the averaged ultrasonic signal; Step A2: define the ultrasonic signal with an excitation angle of 0° as the reference signal, perform Hilbert transform on it, and extract the time corresponding to the maximum value of the bottom echo signal as the reference ultrasonic propagation time; In step A3, a generalized cross-correlation operation is performed on the ultrasonic signals of the remaining excitation angles based on the reference ultrasonic propagation time and the reference signal. The cross-correlation function of each target signal and the reference signal is calculated to determine the time offset value corresponding to its peak value. Finally, the time offset value is added to the reference ultrasonic propagation time to calculate the propagation time of each angle.

3. The method for three-dimensional imaging of a battery using an ultrasonic phased array based on wave velocity correction according to claim 1, wherein: In step S3, the angle-sound speed distribution function is constructed by performing cubic interpolation fitting on the sound speed at each angle, and the angle-sound speed distribution function is used to estimate the sound speed at any angle.

4. The ultrasonic phased array battery three-dimensional imaging method based on wave velocity correction according to claim 1 is characterized in that: In step S42 , the propagation time is obtained by dividing the excitation propagation distance by the excitation sound velocity to obtain the excitation propagation time, dividing the reception propagation distance by the reception sound velocity to obtain the reception propagation time, and then adding the excitation propagation time and the reception propagation time.

5. The method for three-dimensional imaging of a battery using an ultrasonic phased array based on wave velocity correction according to claim 1, wherein: The phased array probe is moved by using a mechanical arm and a stepping motor to drive the phased array probe to move, and the phased array probe is fixed to the end of the mechanical arm.

6. The ultrasonic phased array battery three-dimensional imaging method based on wave velocity correction according to claim 2, characterized in that: The ultrasonic signal superposition and averaging processing is performed only on the channel pairs with equal excitation angles and receiving angles.

7. The ultrasonic phased array battery three-dimensional imaging method based on wave velocity correction according to claim 2, characterized in that: The generalized cross-correlation operation includes: performing cross-correlation processing on the ultrasonic signals of the remaining excitation angles and the reference signal, and determining the time delay when the cross-correlation function reaches a maximum value as the time offset value.

8. The method for three-dimensional imaging of a battery using an ultrasonic phased array based on wave velocity correction according to claim 5, wherein: The robotic arm and the stepper motor cooperate to control the phased array probe to scan at equal intervals of 0.25 mm in a direction perpendicular to the imaging plane where the two-dimensional B-scan image is located, and spatially align and superimpose the collected continuous two-dimensional B-scan images to reconstruct the three-dimensional spatial distribution of defects.

9. The ultrasonic phased array battery three-dimensional imaging method based on wave velocity correction according to claim 1, characterized in that: The battery is a lithium-ion soft-pack battery, and the three-dimensional imaging results are used to simultaneously detect wrinkling defects, delamination defects, foreign matter defects and puncture defects inside the battery.

Images