Ultrasonic identification method for alpha phase of titanium alloy strip and application

Abstract

The invention belongs to the technical field of nondestructive testing of materials, relates to an ultrasonic identification method for an alpha phase of a titanium alloy strip and application, and solves the problem that the alpha phase of a titanium alloy local strip cannot be identified by traditional ultrasonic detection. The method comprises the following steps: S1, calibrating the detection sensitivity of equipment by using a reference block, and then placing a to-be-detected bar on a detection station; s2, full-wave signal scanning and A scanning full-wave time-domain signal acquisition are sequentially carried out on the bar to be detected, A scanning full-wave time-domain signals at the positions corresponding to the abnormal independent signals are stored, then reconstruction operation is carried out, and reconstruction signals are obtained; and S3, carrying out empirical mode decomposition on the reconstructed signal, extracting a characteristic signal, then calculating the time domain power of the characteristic signal, and carrying out ultrasonic imaging according to the time domain power value and the spatial position corresponding to the abnormal independent signal, thereby obtaining a time domain power imaging graph of the to-be-detected bar, and further determining whether the to-be-detected bar contains the long-strip alpha phase or not.

Description

Technical Field

[0001] This invention belongs to the field of non-destructive testing technology for titanium alloy materials, and relates to an ultrasonic identification method and application for long strip α phase of titanium alloy. Background Technology

[0002] Currently, ultrasonic pulse-echo testing is one of the most commonly used core non-destructive testing methods for titanium alloys, widely applied in industrial production and quality control. When ultrasonic waves propagate in a titanium alloy medium, they undergo reflection, scattering, attenuation, and waveform transformation. Single ultrasonic pulse-echo detection signals have limited information dimensions and therefore low detection resolution.

[0003] Within the complex structure of titanium alloys, only a small number of localized elongated and bulky α phases exist, with these anomalous α phases ranging in size from 100 μm to 400 μm. Their ultrasonic signals are mixed with electrical noise, mechanical vibration noise, and other signals, making them difficult to distinguish. According to ultrasonic scattering models, anomalous α phases in titanium alloys do not cause significant differences in noise and backwave amplitude, thus hindering accurate detection.

[0004] Currently, non-destructive testing (NDT) plays an increasingly important role in product quality control, and it is rapidly moving towards automation, digitalization, and imaging. Among these advancements, ultrasonic testing technology has seen particularly significant development, with the introduction of higher-precision digital imaging NDT systems. In the ultrasonic testing of titanium alloy bars, combining ultrasonic digital signal processing and image characterization techniques can improve the accuracy and reliability of the test results. This allows for the effective assessment of abnormal material characteristics within areas of structural anomalies, thereby enhancing the quality control level of titanium alloy bars.

[0005] As an emerging technology in the field of nondestructive testing, ultrasonic testing technology has shown good application prospects in the detection of defects in metallic materials by using the ultrasonic signal processing results of feature extraction and algorithms to improve the ultrasonic testing capabilities. However, its application in the evaluation of abnormal states of α-phase microstructure in titanium alloys is still in its infancy.

[0006] Therefore, developing a targeted ultrasonic digital signal processing method with high abnormal state recognition has become a key issue that urgently needs to be addressed. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and propose an ultrasonic identification method and application for elongated α phases in titanium alloys. This method uses a visualization characterization and analysis method of "wavelet signal reconstruction + modal feature extraction + time-domain power value imaging" to solve the problem that existing ultrasonic detection technology cannot identify local elongated α phases in titanium alloys.

[0008] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention discloses an ultrasonic identification method for the α phase of a titanium alloy strip, comprising the following steps: S1. First, use a comparison test block to calibrate the detection sensitivity of the equipment, and then place the bar to be tested on the detection station; S2. Perform full-wave signal scanning and A-scan full-wave time-domain signal acquisition sequentially on the bar to be inspected, and obtain the following results: Figure 3 The image shown is then reconstructed to obtain the reconstructed signal; S3. Perform empirical mode decomposition on the reconstructed signal and extract feature signals to obtain, as shown below. Figure 6 The image shown is used to calculate the time-domain power of the feature signal after an improved algorithm. Ultrasonic imaging is then performed based on the improved time-domain power value and the spatial location corresponding to the feature signal to obtain the time-domain power image of the bar under test (e.g., [image description missing]). Figure 7 As shown), thereby determining whether the bar to be tested contains elongated α phase (such as... Figure 2 (As shown).

[0009] Specifically, in S2, the A-scan full-wave time-domain signal is composed of several sampling points arranged in chronological order; More specifically, when the absolute value of the amplitude of a certain sampling point exceeds the set amplitude threshold, the sampling point is determined to be an abnormal sampling point, and the set of all the abnormal sampling points is the feature signal; More specifically, the amplitude threshold is determined as follows: the amplitude of several consecutive sampling points is used as the original data, and the average value is calculated. The average value is used as a reference benchmark, and the amplitude threshold is 20dB lower than the average value. Specifically, in S3, the feature signal is a non-stationary feature signal; Specifically, in S3, the blue area of ​​the time-domain power imaging map represents the area with low time-domain power value, and the yellow to dark red area represents the area with high time-domain power value (more than 1.3 times the average value).

[0010] Furthermore, in S1, the ultrasonic testing instrument uses a 5MHz or 10MHz testing probe.

[0011] Specifically, the 5MHz or 10MHz detection probe is used to adapt to the low attenuation characteristics of titanium alloy materials and to balance detection resolution and penetration. Using the detection probe of this specification can accurately identify the elongated α phase in titanium alloys and meet the penetration requirements of titanium alloy rods.

[0012] Furthermore, in S1, the comparative test block has a flat-bottomed hole with a diameter of 0.4 mm.

[0013] Furthermore, in S2, the steps of the reconstruction operation are as follows: S2.1 Perform wavelet packet decomposition on the A-scan full-wave time-domain signal corresponding to the position of the characteristic signal (e.g., Figure 4 (As shown) to obtain nodes; S2.2. Perform multi-scale threshold denoising on the nodes, filter to obtain wavelet packet node coefficients, and then perform wavelet packet reconstruction, as follows. Figure 5 As shown, the reconstructed signal can be obtained.

[0014] Specifically, in S2.1, the "Symlet" wavelet is selected for packet decomposition; Furthermore, in the multi-scale threshold denoising, the calculation method for each threshold is as follows: taking the statistical standard deviation of the noisy signal as the first parameter and the decibel value of the noise intensity as the second parameter, first calculate the square root of the second parameter, and then multiply the square root of the second parameter by the first parameter to obtain the denoising threshold at that scale.

[0015] Furthermore, in S3, the formula for calculating the time-domain power is: Where n represents the number of sampling points; t represents the time corresponding to a certain sampling point; This represents the time-domain signal amplitude at the i-th sampling point; ωi represents the time-domain power of the i-th sampling point. Specifically, the multi-scale threshold noise reduction involves multi-scale decomposition and separation of noise and the reflection signal of the long strip α phase, and retaining the effective signal characteristics after removing the noise coefficient according to a specific threshold; thereby improving the signal-to-noise ratio and purity of the signal, laying the foundation for subsequent accurate identification, positioning and quantitative analysis of the long strip α phase.

[0016] Furthermore, when the time-domain power is higher than 1.3 times the average value, the rod contains a long strip α-phase region; when the time-domain power is lower than 1.3 times the average value, the rod contains only an equiaxed α-phase region.

[0017] Specifically, the average value of the time-domain power is 1.2.

[0018] Secondly, the present invention discloses an ultrasonic identification method for elongated α phases in titanium alloys, applicable to the detection of abnormal elongated α phases of 100μm to 400μm in titanium alloys.

[0019] Compared with the prior art, the present invention has the following beneficial effects: First, this invention achieves full-volume online ultrasonic testing of titanium alloy bars without damaging their structural integrity or preserving their original service performance. It can accurately identify the elongated α-phase in near-α and α+β titanium alloys, given their complex microstructure. By employing a specific frequency ultrasonic probe and combining it with time-domain power algorithms and image characterization techniques, it overcomes the technical limitation of existing detection methods in identifying the elongated α-phase anomalous microstructure in titanium alloys, achieving rapid localization and reliable identification of this type of anomalous microstructure. Secondly, this invention is applicable to titanium alloy bars produced by different forging processes. It can perform ultrasonic testing and analysis on different α-phase distributions within the bars. Through a visual characterization and analysis method of "wavelet signal reconstruction + modal feature extraction + time-domain power value imaging," the detection process is clear, standardized, and highly operable, and the analysis results can be presented quickly and intuitively. This method effectively reduces the detection difficulty of elongated α-phases in near-α titanium alloys and α+β titanium alloys. It does not rely on special consumables or complex pretreatment processes. With only an optimized non-destructive testing technology path, it can efficiently and fully achieve accurate and rapid online identification of elongated α-phases within the alloy, significantly improving the detection rate and identification reliability of such abnormal structures, and is suitable for the quality control needs of industrial production. Attached Figure Description

[0020] The accompanying drawings are incorporated in and form part of this specification, and together with the description serve to explain the principles of the invention.

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of the ultrasonic identification method for the α phase of titanium alloy strips; Figure 2 It is a metallographic distribution diagram of the elongated α phase in near-α or α+β titanium alloys; Figure 3 It is a full-wave time-domain signal diagram of ultrasound; Figure 4 It is a wavelet packet decomposition diagram; Figure 5 This is a wavelet signal reconstruction diagram; Figure 6 It is signal modal feature extraction; Figure 7 It is a time-domain power imaging map; Figure 8 This is the time-domain power diagram of the core of the bar in Example 1; Figure 9 This is the anatomical result of ultrasound examination of the elongated region in Example 1; Figure 10 This is the time-domain power diagram of the core of the bar in Example 2; Figure 11 This is the ultrasound examination and anatomical result of Example 2; Figure 12 This is the time-domain power diagram of the core of the bar in Example 3; Figure 13 This is the result of ultrasound examination and anatomy in Example 3. Detailed Implementation

[0023] Exemplary embodiments will now be described in detail. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples consistent with some aspects of the invention as detailed in the appended claims.

[0024] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0025] Example 1 According to such Figure 1 The process shown involves the following steps: S1. Place the comparison test block with a flat bottom hole of Φ0.4mm on the testing station of the ultrasonic testing instrument to calibrate the testing sensitivity of the equipment, and then place the first bar on the testing station.

[0026] Specifically, the first bar is made of near-α titanium alloy and has a specification of Φ250mm.

[0027] S2. Perform a full-wave signal scan along the length of the first rod (the scan area is set to the region from 0mm to 1000mm from the left end of the first rod), and perform A-scan full-wave time-domain signal acquisition. Then, extract the A-scan full-wave time-domain signal segment corresponding to the core of the rod with a burial depth of 100mm to 150mm, and perform wavelet packet decomposition on the signal segment using "Symlet" wavelet to obtain the nodes. After performing multi-scale threshold denoising on the nodes and filtering to obtain the wavelet packet node coefficients, wavelet packet reconstruction is then performed to obtain the reconstructed signal.

[0028] S3. Perform empirical mode decomposition on the reconstructed signal and extract non-stationary feature signals. Then, calculate the time-domain power of the feature signals after the improved algorithm, and perform ultrasound imaging based on the improved time-domain power value and the spatial location corresponding to the feature signals. Figure 8As shown, the blue area represents the low time-domain power value area, and the yellow to dark red area represents the high time-domain power value (1.3 times higher than the average value). It can be observed that the cardiac ultrasound image contains areas with time-domain power values ​​higher than 1.3 times the average value. Therefore, the first rod has a long strip α phase in the scanning area.

[0029] Example 2 According to such Figure 1 The process shown involves the following steps: S1. Place the comparison test block with a flat bottom hole of Φ0.4mm on the testing station of the ultrasonic testing instrument to calibrate the testing sensitivity of the equipment, and then place the second bar on the testing station.

[0030] Specifically, the second bar is made of α+β titanium alloy and has a specification of Φ250mm.

[0031] S2. Perform a full-wave signal scan along the length of the second rod (the scan area is set to the region from 1000mm to 2000mm from the left end of the second rod), and perform A-scan full-wave time-domain signal acquisition. Then, extract the A-scan full-wave time-domain signal segment corresponding to the core of the rod with a burial depth of 100mm to 150mm, and perform wavelet packet decomposition on the signal segment using "Symlet" wavelet to obtain the nodes. After performing multi-scale threshold denoising on the nodes and filtering to obtain the wavelet packet node coefficients, wavelet packet reconstruction is then performed to obtain the reconstructed signal.

[0032] S3. Perform empirical mode decomposition on the reconstructed signal and extract non-stationary feature signals. Then, calculate the time-domain power of the feature signals after the improved algorithm, and perform ultrasound imaging based on the improved time-domain power value and the spatial location corresponding to the feature signals. Figure 10 As shown, the blue area represents the low time-domain power value area, and the yellow to dark red area represents the high time-domain power value (more than 1.3 times the average value). It can be observed that the cardiac ultrasound image is entirely blue, that is, there is no area where the time-domain power value is more than 1.3 times the average value. Therefore, the second rod is equiaxed α phase in the scanning area and there is no long strip α phase.

[0033] Example 3 According to such Figure 1 The process shown involves the following steps: S1. Place the comparison test block with a Φ0.4mm flat bottom hole at the testing station of the ultrasonic testing instrument to calibrate the testing sensitivity of the equipment, and then place the third bar at the testing station.

[0034] Specifically, the third bar is made of α+β titanium alloy and has a specification of Φ250mm.

[0035] S2. Scan the entire material along the length of the third bar and perform A-scan full-wave time domain signal acquisition. Then, extract the A-scan full-wave time domain signal segment corresponding to the core of the bar with a burial depth of 100mm~150mm, and perform wavelet packet decomposition on the signal segment using "Symlet" wavelet to obtain the nodes. After performing multi-scale threshold denoising on the nodes and filtering to obtain the wavelet packet node coefficients, wavelet packet reconstruction is then performed to obtain the reconstructed signal.

[0036] S3. Perform empirical mode decomposition on the reconstructed signal and extract non-stationary feature signals. Then, calculate the time-domain power of the feature signals after the improved algorithm, and perform ultrasound imaging based on the improved time-domain power value and the spatial location corresponding to the feature signals. Figure 12 As shown, the blue area represents the low time-domain power value area, and the yellow to dark red area represents the high time-domain power value (1.3 times higher than the average value). It can be observed that in the cardiac ultrasound detection image, there are areas with time-domain power values ​​higher than 1.3 times the average value as well as areas with normal time-domain power values. Therefore, there is a long strip α phase in the corresponding scanning area of ​​the third rod.

[0037] To verify the reliability of the present invention, a section was performed along the radial direction of the first bar in Example 1, in an area with a core embedment depth of 100mm to 150mm. Figure 9 As shown, microscopic tissue analysis revealed that long α-phases could be clearly observed in abnormal regions with time-domain power values ​​1.3 times higher than the average value. This result is completely consistent with the abnormal signal determination conclusion of the ultrasound identification method of this invention. Along the radial direction of the second bar in Example 2, the area with a core embedment depth of 100mm~150mm was sectioned for inspection, such as... Figure 11 As shown, microscopic tissue analysis revealed that the region with a time-domain power value lower than 1.3 times the average value was an equiaxed α phase, and no elongated α phase was observed. This result is completely consistent with the judgment conclusion of the ultrasonic identification of the present invention. Along the radial direction of the third bar described in Example 3, a section was performed on the area with a core embedment depth of 100mm~150mm, as shown below. Figure 13 As shown, microscopic tissue analysis revealed that regions with time-domain power values ​​below 1.3 times the average value were equiaxed α phases, and no elongated α phases were observed. In contrast, abnormal regions with time-domain power values ​​above 1.3 times the average value clearly showed elongated α phases. This result is completely consistent with the judgment conclusion of the ultrasound identification method of this invention. In summary, the ultrasonic identification method of the present invention is reliable.

[0038] To visually present the ultrasound identification parameters and results of Examples 1-3, the present invention also provides the following tables: In addition, this specification also provides an ultrasonic identification method for titanium alloy strips of α phase and the principle of its application, as detailed below: In industry, ultrasonic testing is commonly used to detect defects located below the surface of titanium alloy materials. When ultrasonic waves propagate to the interface between two acoustic impedances, a certain proportion of the ultrasonic waves are reflected back. By scanning the workpiece with an ultrasonic probe, the intensity of the ultrasonic echo signal inside the workpiece is displayed in grayscale or color, thus achieving the purpose of mapping the abnormal cross-section inside the workpiece. This method of defect detection has the advantages of intuitive imaging and high testing accuracy.

[0039] When ultrasound propagates in titanium alloy media, it undergoes reflection, scattering, attenuation, and waveform conversion. Some of these reflections are caused by internal pores, cracks, and inclusions in the material, while others are noise-scattered signals generated by the inhomogeneity of the titanium alloy's microstructure. Current testing techniques often use 5MHz or 10MHz probes to inspect titanium alloy bars. Because the 100μm~400μm elongated α-phase size is smaller than the ultrasonic testing wavelength of titanium alloy materials, the reflection and scattering signals from the elongated α-phase microstructure are weak. Furthermore, the reflection and scattering signals from the elongated α-phase microstructure are easily mixed with other structural noise, electrical noise, and mechanical vibration noise. Therefore, it is necessary to extract useful ultrasonic characteristic signals to characterize the inhomogeneous regions present in the elongated α-phase, thereby achieving ultrasonic evaluation of the elongated α-phase microstructure.

[0040] like Figure 2 As shown, near-α and α+β titanium alloys sometimes contain several relatively long α-phase regions, with macroscopic lengths distributed along the bar in millimeter increments. To fully and rapidly characterize the distribution of long α-phases in the bar online, this invention first establishes the relationship between the α-phase and the ultrasonic noise signal scattering—the larger the aspect ratio of the α-phase, the higher the ultrasonic noise. Second, this invention employs a visualization and characterization analysis method combining "multi-scale signal reconstruction + modal feature decomposition + power and amplitude coupling," and sets a threshold for long α-phase strips in ultrasonic detection images, achieving rapid detection and image characterization of long α-phase strips with different length ranges.

[0041] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention.

[0042] It should be understood that the present invention is not limited to the content already described above, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. An ultrasonic identification method for the α phase of a titanium alloy strip, characterized in that, Includes the following steps: S1. First, use a comparison test block to calibrate the detection sensitivity of the equipment, and then place the bar to be tested on the detection station; S2. Perform full-wave signal scanning and A-scan full-wave time-domain signal acquisition on the bar to be inspected in sequence, and then perform reconstruction operation to obtain the reconstructed signal; S3. Perform empirical mode decomposition on the reconstructed signal and extract the feature signal. Then calculate the time-domain power of the feature signal after the improved algorithm. Perform ultrasonic imaging based on the time-domain power value after the improved algorithm and the spatial position corresponding to the feature signal to obtain the time-domain power imaging map of the bar under test, and then determine whether the bar under test contains a long strip α phase.

2. The ultrasonic identification method for the α phase of a titanium alloy strip according to claim 1, characterized in that, In S1, the ultrasonic testing instrument uses a testing probe with a frequency of 5MHz or 10MHz.

3. The ultrasonic identification method for the α phase of a titanium alloy strip according to claim 1, characterized in that, In S1, the comparative test block has a flat-bottomed hole with a diameter of 0.4 mm.

4. The ultrasonic identification method for the α phase of a titanium alloy strip according to claim 1, characterized in that, In S2, the steps of the reconstruction operation are as follows: S2.1 Perform wavelet packet decomposition on the A-scan full-wave time-domain signal corresponding to the position of the feature signal to obtain the node; S2.2 First, perform multi-scale threshold noise reduction on the nodes, filter to obtain wavelet packet node coefficients, and then perform wavelet packet reconstruction to obtain the reconstructed signal.

5. The ultrasonic identification method for the α phase of a titanium alloy strip according to claim 4, characterized in that, In the multi-scale threshold denoising, the calculation method for each threshold is as follows: taking the statistical standard deviation of the noisy signal as the first parameter and the decibel value of the noise intensity as the second parameter, first calculate the square root of the second parameter, and then multiply the square root of the second parameter by the first parameter to obtain the denoising threshold at that scale.

6. The ultrasonic identification method for the α phase of a titanium alloy strip according to claim 1, characterized in that, In S3, the formula for calculating the time-domain power is: Where n represents the number of sampling points; t represents the time corresponding to a certain sampling point; This represents the time-domain signal amplitude at the i-th sampling point; ωi represents the time-domain power of the i-th sampling point.

7. The ultrasonic identification method for the α phase of a titanium alloy strip according to claim 5, characterized in that, When the time-domain power is higher than 1.3 times its average value, the rod contains a long strip α-phase region; when the time-domain power is lower than 1.3 times its average value, the rod contains only an equiaxed α-phase region.

8. The application of the ultrasonic identification method for the α phase of a titanium alloy strip according to any one of claims 1 to 7, characterized in that, It is suitable for detecting anomalous α phases with lengths of 100μm to 400μm in titanium alloys.

Images