Austenitic stainless steel sparse array full-focusing imaging method based on TRL probe

Abstract

The invention provides an austenitic stainless steel sparse array full-focus imaging method based on a TRL probe, and the method comprises the steps: employing the amplitude level of a sound beam pointing to a sidelobe as a target function, building an optimization mathematical model through combining with a genetic algorithm, and obtaining a sparse array arrangement mode of the TRL probe. A TRL area array probe is used for collecting full-matrix data of the thick-wall austenitic stainless steel, and sparse array imaging is carried out through optimized TRL array element arrangement. The imaging time of a full array is shortened from 623.91 s to about 50 s, the imaging speed is increased by about 92%, and the defect signal-to-noise ratio is only reduced by 14.69%. According to the invention, the TFM imaging efficiency can be greatly improved under the condition of little loss of imaging quality, and good balance between the detection efficiency and the imaging quality is realized.

Description

Technical Field

[0001] The invention relates to the technical field of full-focus imaging of thick-walled austenitic stainless steel, and in particular to a sparse array full-focus imaging method of austenitic stainless steel based on a TRL probe. Background Art

[0002] Due to the columnar crystal structure and coarse grains in the weld of thick-walled austenitic stainless steel, when ultrasonic waves propagate in the weld, they will be seriously affected by the grain diameter and elastic anisotropy of the columnar crystal structure, causing changes in the sound velocity, increased scattering attenuation, beam deviation, low detection signal-to-noise ratio, etc., which bring difficulties to defect detection and accurate quantification and positioning. Compared with phased array ultrasonic testing technology (PAUT), TFM imaging technology has the advantages of high imaging signal-to-noise ratio (SNR), high defect image resolution, and small defect positioning / quantification error. However, the total focus method (TFM) imaging technology is limited in its industrial application due to the large amount of FMC data involved in imaging, the long time-consuming delay superposition calculation of pixel points, poor real-time processing capability, and high hardware investment cost of the detection system. The number of array elements involved in sparse array imaging is small, and the processing and calculation of data are correspondingly reduced, which can effectively improve the imaging efficiency. Therefore, it is very necessary to design a sparse array total focus imaging method for austenitic stainless steel based on TRL probe. Summary of the invention

[0003] In order to improve the problems of large computational complexity and low imaging speed of TRL probe full focusing imaging, the present invention provides a full focusing imaging method of austenitic stainless steel sparse array based on TRL probe, which greatly improves the imaging speed while maintaining the imaging quality to a certain extent.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] The present invention provides a method for full-focus imaging of an austenitic stainless steel sparse array based on a TRL probe, comprising:

[0006] Based on the TRL array probe, ultrasonic signal acquisition is performed on thick-walled austenitic stainless steel to obtain full matrix capture data;

[0007] The TRL array layout is optimized by using genetic algorithm combined with beam pointing function;

[0008] Sparse array full focusing imaging is performed through optimized TRL array element arrangement.

[0009] Preferably, a genetic algorithm is used in combination with a beam pointing function to optimize the TRL array arrangement, specifically:

[0010] Establish fitness function;

[0011] Genetic algorithm is used to optimize the sparse arrangement of TRL arrays.

[0012] Preferably, a fitness function is established, specifically:

[0013] An optimization mathematical model is established according to the level of the acoustic beam directivity sidelobe amplitude, and the fitness function of the genetic algorithm is determined;

[0014] If the array elements are regarded as isotropic point sources, the beam E(u) of a sparse array with N elements can be expressed as:

[0015]

[0016] Where, d n and I n They represent the coordinates and excitation amplitude of the array element, u = cos(θ), θ is the deflection angle of the beam, 0≤θ≤π, k = 2π / λ, λ is the ultrasonic wavelength in the medium, and the amplitude of each array element in the TRL probe is equal, that is, I n =1, and d 1 =0,d N =L, simplify the formula to:

[0017]

[0018] Taking minimizing the peak sidelobe level of the sparse array as the optimization goal of the algorithm, according to the meaning of the array peak sidelobe, the fitness function of the genetic algorithm is defined as:

[0019]

[0020] In the formula, FF max is the peak value of the main lobe of the beam, and the variable u represents the beam range excluding the main lobe.

[0021] Preferably, sparse array imaging is performed by optimizing the TRL array element arrangement, specifically:

[0022] Calculate the coordinates of the TRL array probe elements;

[0023] Calculate the coordinates of the pixels in the imaging area;

[0024] Discrete multiple sound beam refraction points at the dual-medium interface;

[0025] It takes time to calculate the beam path propagation after array sparse optimization;

[0026] Select the shortest transmission path based on Fermat's principle;

[0027] Traverse all sparse signals and delay superposition the pixel;

[0028] Repeat the above steps to complete the full-focus image of the entire TRL sparse array.

[0029] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0030] The present invention provides a method for full-focusing imaging of a sparse array of austenitic stainless steel based on a TRL probe, the method comprising using the amplitude level of the acoustic beam pointing sidelobe as an objective function, combining a genetic algorithm to establish an optimization mathematical model, and obtaining a sparse array arrangement of the TRL probe. A TRL array probe is used to collect full matrix data of thick-walled austenitic stainless steel, and sparse array imaging is performed through the optimized TRL array element arrangement. Compared with the PAUT detection technology, the present invention has the advantages of high imaging resolution, high signal-to-noise ratio, and small positioning / quantification error. Compared with the TFM imaging detection technology, the TFM imaging efficiency can be greatly improved with only a small loss of imaging quality, and the effect of 25% sparse array imaging is significantly better than PAUT imaging, achieving a good balance between detection efficiency and imaging quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0032] Figure 1 A flow chart of a method provided by an embodiment of the present invention;

[0033] Figure 2 It is a schematic diagram of the genetic algorithm process;

[0034] Figure 3a This is a schematic diagram of the directivity of the sound beam arranged in a full array;

[0035] Figure 3b Schematic diagram of the directivity of the sound beam of a random 25% sparse array arrangement;

[0036] Figure 3c Schematic diagram of the acoustic beam directivity of the 25% sparse array arrangement after GA optimization;

[0037] Figure 4a This is a schematic diagram of the comparison of imaging results of full array arrangement when the sparse rate is 25%;

[0038] Figure 4b A schematic diagram showing the comparison of imaging results of random 25% sparse array arrangement when the sparse rate is 25%;

[0039] Figure 4cA schematic diagram showing the comparison of imaging results of a 25% sparse array arrangement after GA optimization when the sparse rate is 25%;

[0040] Figure 5 Schematic diagram of PAUT imaging results. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments 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 creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] The main content of the present invention is a sparse array imaging method based on a TRL array probe. This method uses a low-frequency TRL array probe to perform TFM imaging on thick-walled austenitic stainless steel welds, and specifically solves the problems of large TFM imaging data volume and slow imaging speed. The present invention writes a sparse array imaging algorithm. Based on the collected Full Matrix Capture (FMC) data, the algorithm optimizes the TRL array arrangement based on a genetic algorithm (GA), and uses the TRL array element arrangement optimized by GA for sparse array imaging. This algorithm can significantly reduce the number of signal receiving channels with a small loss of imaging quality, greatly improving the efficiency of TFM imaging. The following is a detailed introduction to this:

[0044] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a method for full-focus imaging of a sparse array of austenitic stainless steel based on a TRL probe, comprising:

[0045] Step 100: Acquire ultrasonic signals of thick-walled austenitic stainless steel based on a TRL array probe to obtain full matrix capture data;

[0046] Step 200: Optimizing the TRL array arrangement by using a genetic algorithm combined with a beam pointing function;

[0047] Step 300: Perform sparse array imaging through the optimized TRL array element arrangement.

[0048] In step 100, the TRL array probe used is independently developed and designed, and the frequency and structural parameters are shown in Table 1;

[0049] Table 1 TRL array probe parameters

[0050]

[0051]

[0052] In step 200, according to the full matrix capture data, the TRL array arrangement is optimized based on the genetic algorithm, specifically:

[0053] Establish fitness function;

[0054] Genetic algorithm is used to optimize the sparse arrangement of TRL arrays.

[0055] Establish a fitness function, specifically:

[0056] According to the level of the sidelobe amplitude of the acoustic beam directivity, an optimization mathematical model is established and the fitness function of the genetic algorithm is determined. The details are as follows:

[0057] The array optimization goal is to find an optimal array element coordinate vector d = [d 1 , d 2 ,…,d N ], so that the peak side lobe level (PSLL) of the array beam is minimized. The fitness function of the algorithm needs to calculate the peak side lobe level of each array. Taking the array element as an isotropic point source, the beam E(u) of a sparse array with N elements is expressed as:

[0058]

[0059] Where, d n and I n They represent the coordinates and excitation amplitude of the array element, u = cos(θ), θ is the deflection angle of the beam, 0≤θ≤π, k = 2π / λ, λ is the ultrasonic wavelength in the medium, and the amplitude of each array element in the TRL probe is equal, that is, I n =1, and d 1 =0,d N =L, simplify the formula to:

[0060]

[0061] Taking minimizing the peak sidelobe level of the sparse array as the optimization goal of the algorithm, according to the meaning of the array peak sidelobe, the fitness function of the genetic algorithm is defined as:

[0062]

[0063] In the formula, FF maxis the peak value of the main lobe of the beam, and the variable u value area represents the beam range except the main lobe.

[0064] The genetic algorithm is used to optimize the sparse arrangement of TRL arrays, specifically:

[0065] The genetic algorithm process is as follows Figure 2 As shown in the figure, according to the general steps of the genetic algorithm, N random numbers with values ​​between 0 and L can be grouped as the initial individuals of the population. However, it will be difficult to test the validity of these initial individuals (i.e., whether they meet the constraint of the minimum array element spacing). Therefore, a new individual description method is used to initialize the population. The array endpoint position is used as the coordinate origin, so that the coordinate of the first array element is always 0, the coordinate of the Nth array element is always L, and the remaining N-2 array elements are randomly distributed in the array aperture. In order to meet the constraint of the minimum array element spacing, the coordinates of the second array element and the N-1th array element must satisfy d 2 ≥d c and d N-1 ≤Ld c (d c is the minimum array element spacing), and according to the above principle, the sparse optimization arrangement of TRL array elements based on the GA algorithm is realized.

[0066] Sparse array imaging is performed through the optimized TRL array element arrangement, specifically:

[0067] Calculate the coordinates of the TRL array probe elements;

[0068] Calculate the coordinates of the pixels in the imaging area;

[0069] Discrete multiple sound beam refraction points at the dual-medium interface;

[0070] It takes time to calculate the beam path propagation after array sparse optimization;

[0071] Select the shortest transmission path based on Fermat's principle;

[0072] Traverse all sparse signals and delay superposition the pixel;

[0073] Repeat the above steps to complete the full-focus image of the entire TRL sparse array.

[0074] The expanded explanation is: If the coordinates of the TRL probe transmitting array element are known: (x 1 ,y 1 ,z 1 ), pixel coordinates: (x 2 ,y 2 ,z 2 ), receiving array element coordinates: (x 3 ,y 3 ,z 3), the propagation path time of the sound beam can be expressed as:

[0075]

[0076] Where, d 1 ,d 2 ,d 3 It can be expressed as:

[0077]

[0078] Among them, (x, y, z) are the coordinates of the refraction discrete point, c 1 、c 2 The wedge sound velocity and the workpiece sound velocity are used to calculate the value of t of the transmit-receive signal for the pixel after the optimized array element arrangement. The pixel can be delayed and superimposed:

[0079]

[0080] Wherein, h(S(t)) is the signal after Hilbert transformation. According to equations (4), (5) and (6), the amplitude of any pixel in the imaging area can be obtained.

[0081] The present invention provides an embodiment to collect FMC data of a thick-walled austenitic stainless steel weld test block with a thickness of 100 mm. The weld of the test block is distributed with transverse through hole defects with depths of 10 mm, 50 mm, and 90 mm and a diameter of 3.2 mm. The frequency and structural information of the TRL array probe used are shown in Table 1. The wedge block selected has a longitudinal wave refraction angle of 45°, a roof angle focusing depth of 50 mm, and a longitudinal wave sound velocity of 2337 m / s. Based on the FMC data, the calculated sound beam directivities of the full array, the random 25% sparse array without GA optimization, and the 25% sparse array after GA optimization are as follows: Figure 3a-3c As shown in the figure, the sidelobe amplitudes of the three different arrays are -30dB, -12.06dB and -19.57dB respectively. It can be seen that GA optimization can effectively reduce the sidelobe amplitude, improve the directivity of the sound beam, and help improve the sparse array TFM imaging effect.

[0082] The full array, 25% uniform sparse array and 25% sparse array after GA optimization were subjected to TFM imaging respectively. The imaging results are shown in the figure below. Figure 4a-4c shown.

[0083] The imaging speeds of full array, uniform sparse 25% and sparse 25% after GA optimization are 623.91s, 50.12s and 48.44s respectively. It can be seen that the imaging speed of sparse array after GA optimization is more than 12 times higher than that of TFM imaging. Compared with the uniform sparse imaging results, the defect image smear problem of hole 2 after GA optimization is significantly improved, and the noise amplitude in the image is suppressed.

[0084] Using the same test parameters as above, PAUT test was performed on thick-walled austenitic stainless steel test blocks. The imaging results are shown in Figure 2. Figure 5 shown.

[0085] The API, SNR, positioning and quantitative errors of the images corresponding to different imaging methods are compared, and the results are shown in Table 2. Figure 4a-4c and Figure 5 Compared with the sector scanning imaging method, the average signal-to-noise ratio of defects in the full array TFM imaging method is improved by 6.73dB, the positioning error is reduced by 3.29mm, the defect quantification error is reduced by 3.90mm, and the image API is reduced by 9.42, indicating that the full array TFM imaging method has a higher defect characterization capability than the sector scanning imaging method.

[0086] Compared with the TFM imaging method, the GA sparse array imaging method involved in the present invention has a 0.36 reduction in average API and a 2.37 dB decrease in average SNR when the sparsity is 25%, and an imaging speed increased by more than 12 times. In addition, the image API, SNR, and positioning / quantification errors of the GA-optimized 25% sparse array are better than those of PAUT imaging and uniformly sparse 25% array imaging.

[0087] Table 2 Comparative analysis of imaging results of different imaging methods

[0088]

[0089]

[0090] In summary, a method for total focusing imaging of a thick-walled austenitic stainless steel sparse array based on a TRL array probe, according to the embodiments recorded in the present invention, can reduce the number of signal receiving channels by 75% with only a small loss in imaging quality, increase the TFM imaging rate by more than 12 times, and achieve a good balance between detection efficiency and image quality.

[0091] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0092] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for total focusing imaging of austenitic stainless steel sparse array based on TRL (Transmit-Receive Longitudinal) probe, characterized in that: include: Based on the TRL array probe, ultrasonic signal acquisition is performed on thick-walled austenitic stainless steel to obtain full matrix capture data; The TRL array layout is optimized by using genetic algorithm combined with beam pointing function; Sparse array full focusing imaging is performed through optimized TRL array element arrangement.

2. The method according to claim 1, characterized in that The genetic algorithm is combined with the beam pointing function to optimize the TRL array arrangement, specifically: Establish fitness function; Genetic algorithm is used to optimize the sparse arrangement of TRL arrays.

3. The method according to claim 2, characterized in that Establish a fitness function, specifically: An optimization mathematical model is established according to the level of the acoustic beam directivity sidelobe amplitude, and the fitness function of the genetic algorithm is determined; If the array elements are regarded as isotropic point sources, the beam E(u) of a sparse array with N elements can be expressed as: Where, d n and I n They represent the coordinates and excitation amplitude of the array element, u = cos(θ), θ is the deflection angle of the beam, 0≤θ≤π, k = 2π / λ, λ is the ultrasonic wavelength in the medium, and the amplitude of each array element in the TRL probe is equal, that is, I n =1, and d1 = 0, d N =L, simplify the formula to: Taking minimizing the peak sidelobe level of the sparse array as the optimization goal of the algorithm, according to the meaning of the array peak sidelobe, the fitness function of the genetic algorithm is defined as: In the formula, FF max is the peak value of the main lobe of the beam, and the variable u represents the beam range excluding the main lobe.

4. The method according to claim 1, characterized in that: Sparse array imaging is performed through the optimized TRL array element arrangement, specifically: Calculate the coordinates of the TRL array probe elements; Calculate the coordinates of the pixels in the imaging area; Discrete multiple sound beam refraction points at the dual-medium interface; It takes time to calculate the beam path propagation after array sparse optimization; Select the shortest transmission path based on Fermat's principle; Traverse all sparse signals and delay superposition the pixel; Repeat the above steps to complete the full-focus image of the entire TRL sparse array.

Images