Method for rapidly detecting defects of large shaft of horizontal tail of airplane based on flexible phased array

Abstract

The invention provides an aircraft horizontal tail large shaft defect rapid detection method based on a flexible phased array. The method comprises the following steps: step 1, determining a fitting position of the flexible phased array on a to-be-detected workpiece and determining an imaging area; step 2, carrying out grid division on the imaging area; 3, transmitting an ultrasonic signal and collecting echo data; and step 4, based on the echo data, calculating to obtain a defect imaging picture of the imaging area, and completing rapid detection of the airplane horizontal tail large shaft defect. According to deep analysis of the imaging principle of each frame, it is found that a time delay matrix has invariance, imaging acceleration is carried out by adopting a GPU-based parallel acceleration method, and a rapid full-focus algorithm is finally provided. Verification shows that the imaging rate is improved from average 4.6 s / frame to average 0.3 s / frame by adopting the fast full-focus algorithm provided by the invention, and the imaging rate is improved by 14.3 times.

Description

Technical Field

[0001] The invention relates to a method for quickly detecting defects of an aircraft horizontal tail shaft based on a flexible phased array, and belongs to the technical field of ultrasonic detection. Background Art

[0002] The aircraft horizontal tail shaft is the transmission shaft of the aircraft's horizontal tail stabilizer. The fixed end of the shaft is connected to the aircraft fuselage, and the aircraft's horizontal tail is suspended on two force-bearing fulcrums: the fixed end and the suspension end of the horizontal tail shaft. Therefore, the aircraft horizontal tail shaft is an important load-bearing component connecting the aircraft's horizontal tail and the aircraft fuselage. The forces it bears are mainly bending stress and torsional stress transmitted through the aircraft's horizontal tail. If the horizontal tail shaft has defects such as holes or seams due to process errors during the casting process or due to stress fatigue during flight, it may cause the shaft to break, causing the aircraft's horizontal tail to lose balance out of control, seriously affecting the aircraft's flight safety. Therefore, factory inspection, regular inspection, maintenance and repair of the horizontal tail shaft are of great significance to ensure aircraft safety.

[0003] The ultrasonic method generates ultrasonic waves by exciting the sound field inside the workpiece. When the sound waves encounter the internal defects of the test piece, they will be reflected back to generate defect echoes. By processing the defect echoes, information such as the size and location of the internal defects of the test piece can be obtained. Limited by the curved shape of the horizontal tail shaft, the traditional single-element transmitting and single-element receiving ultrasonic method has low detection accuracy and cannot accurately detect the size and location of defects inside the shaft. Therefore, based on the single-transmitting and single-receiving ultrasonic method, the phased array ultrasonic detection method came into being.

[0004] Ultrasonic phased arrays are divided into two types: rigid phased arrays and flexible phased arrays. Rigid phased arrays cannot be bent and are difficult to perfectly fit the curved surface of the shaft, while flexible phased arrays have good bendability and high fit. Therefore, based on the comparison of the advantages and disadvantages of these two phased arrays and the actual detection environment, the flexible phased array ultrasonic defect detection solution for the horizontal tail shaft is selected. Since the actual industry has high requirements for the automation level, detection efficiency and noise resistance of the horizontal tail shaft defect detection, it is necessary to optimize the design of the traditional defect imaging algorithm to achieve an increase in imaging speed. Summary of the invention

[0005] In order to solve the above problems, the present invention further proposes a method for rapid detection of defects in an aircraft horizontal tail shaft based on a flexible phased array.

[0006] The technical solution adopted by the present invention to solve the above problems is: the present invention proposes a method for rapid detection of defects of aircraft horizontal tail shaft based on flexible phased array, comprising:

[0007] Step 1: Determine the fitting position of the flexible phased array on the workpiece to be measured and determine the imaging area;

[0008] Step 2: Grid the imaging area;

[0009] Step 3: Transmit ultrasonic signals and collect echo data;

[0010] Step 4: Based on the echo data, the defect imaging map of the imaging area is calculated to complete the rapid detection of defects in the aircraft horizontal tail shaft.

[0011] Optionally, the grid division of the imaging area in step 2 specifically includes:

[0012] The imaging area is divided into two sub-areas with the center of the aircraft horizontal tail axis as the pole O and the ray perpendicular to the pole as the polar axis. A polar coordinate system is established, and the polar axis and polar angle directions of the imaging area are evenly divided into grids.

[0013] Optionally, the step of transmitting an ultrasonic signal and collecting echo data in step 3 includes:

[0014] Step 3.1: Take the grid intersection point P in the divided grid as the imaging focus point;

[0015] Step 3.2: Select sparse transmit array elements and sparse transmit array elements;

[0016] Step 3.3: Calculate the acoustic wave propagation distances between the imaging focus point and the sparse transmitting array element and the sparse receiving array element respectively;

[0017] Step 3.4: Calculate the delay time of the ultrasonic signal from the imaging focus point from reception to transmission based on the sound wave propagation distance;

[0018] Step 3.5: If there is a defect at the imaging point, the transmitted ultrasonic signal is reflected at the corresponding imaging point to form echo data.

[0019] Optionally, the sparse transmit array element N in step 3.2 e The number is 32, sparse receiving array element N r The number is 32.

[0020] Optionally, the step of calculating the defect imaging image of the imaging area in step 4 includes:

[0021] Step 4.1: Integrate the GPU parallel computing acceleration framework into the sparse transceiver full-focus imaging algorithm, perform time delay compensation at the corresponding imaging point on all echo data received by the receiving array element, and then superimpose them to obtain the beamforming amplitude at the corresponding imaging point;

[0022] Step 4.2: Traverse all imaging points in the imaging area, repeatedly calculate the synthetic beam amplitude of all imaging points in the imaging area, obtain a synthetic beam amplitude intensity distribution map, and output the synthetic beam amplitude intensity distribution map as a defect imaging map.

[0023] The beneficial effects of the present invention are:

[0024] 1. The present invention proposes a sparse transmission full focusing algorithm by sparsely selecting the transmission array elements, and on this basis, selects sparse receiving array elements, and proposes a sparse transmission and reception full focusing algorithm, which further improves the imaging rate and solves the problem of long time consumption of the traditional full matrix full focusing algorithm.

[0025] 2. Based on the in-depth analysis of the imaging principle of each frame, the present invention finds that the delay matrix is ​​invariant. The present invention uses a GPU-based parallel acceleration method to accelerate imaging, and finally proposes a fast full-focusing algorithm. Through verification, it is found that the fast full-focusing algorithm proposed by the present invention increases the imaging rate from an average of 4.6s / frame to an average of 0.3s / frame while ensuring the imaging quality, and the imaging rate is increased by 14.3 times. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flowchart of a method for rapid defect detection of a large shaft of an aircraft horizontal tail based on a flexible phased array provided by the present invention and a schematic diagram of defect imaging within a curved surface structure;

[0027] Figure 1 In the figure, (a) is a flow chart of a method for rapid defect detection of aircraft horizontal tail shaft based on flexible phased array, and (b) is a schematic diagram of defect imaging in a curved surface structure;

[0028] Figure 2 A schematic diagram of the transmission and reception mode of the sparse array element transmission full focusing algorithm provided by the present invention;

[0029] Figure 3 The sparse emission full-focus imaging diagram of different excitation modes provided by the present invention;

[0030] Figure 3 Among them, (a) is the imaging diagram when the sparse emission fully focused sparse emission array element is 64, (b) is the imaging diagram when the sparse emission fully focused sparse emission array element is 32, (c) is the imaging diagram when the sparse emission fully focused sparse emission array element is 16, (d) is the imaging diagram when the sparse emission fully focused sparse emission array element is 8, (e) is the imaging diagram when the sparse emission fully focused sparse emission array element is 4, (f) is the imaging diagram when the sparse emission fully focused sparse emission array element is 2;

[0031] Figure 4 A schematic diagram of the transmission and reception mode of the sparse transmission and reception full-focusing algorithm provided by the present invention;

[0032] Figure 5 A comparison diagram of the sparse transmission full focus and sparse transmission and reception full focus imaging provided by the present invention;

[0033] Figure 6A flow chart of the fast all-focusing algorithm provided by the present invention;

[0034] Figure 7 This is a comparison chart of the imaging results of the fast full focusing algorithm for defects at different depths and angles provided by the present invention. DETAILED DESCRIPTION

[0035] Combination Figure 1-7 To illustrate this embodiment, Figure 1 As shown, the method for rapid defect detection of aircraft horizontal tail shaft based on flexible phased array described in this embodiment includes the following steps:

[0036] S1: Determine the flexible phased array fitting position and imaging area;

[0037] S2: Divide the imaging area into grids, each grid including a number of imaging points;

[0038] The detection surface of the traditional linear array hard probe is a rectangular plane, so the rectangular coordinate system is generally used for modeling and calculation of imaging algorithms. When the flexible phased array is fitted to the horizontal tail axis, the detection surface of the element is a circular arc surface. If the rectangular coordinate system is continued to be used for calculation and imaging, there will be many inconveniences. Therefore, this embodiment establishes a polar coordinate system for the imaging area with the center of the axis as the pole O, and the ray passing through the pole is used as the polar axis to vertically divide the imaging area into two, and the imaging points are evenly divided in the polar axis and polar angle directions.

[0039] S3: transmit ultrasonic signals and collect echo data;

[0040] S301: Figure 1 As shown, in this embodiment, the polar axis and polar angle are evenly divided into grids by the imaging area inside the horizontal tail shaft ring, the grid intersection point P is calculated, and the grid intersection point P is used as the imaging focus point P;

[0041] S302: Selecting sparse transmitting array elements and sparse receiving array elements;

[0042] S30201: Selection of sparse transmitting array elements: The sparse array element transmits the fully focused array element transmit and receive mode such as Figure 2 As shown, by sparsely exciting the phased array elements, all the elements are selected to receive echoes after each excitation, so as to collect the echo data matrix RF of the sparse transmitting elements. N×Ne , where N e The sparse selection method greatly reduces the number of transmitting elements required for sparse excitation. Compared with the full-matrix full-focusing algorithm, the number of channels of the sparse echo data matrix collected is less, thereby reducing the number of delayed superposition calculations for each imaging point and greatly improving the imaging speed.

[0043] As the number of sparsely selected excitation array elements decreases, the API parameter value of imaging becomes larger and larger, that is, the imaging quality becomes worse and worse. e When N is ≤16, obvious artifacts will appear in the imaging area, thus reducing the imaging contrast at the defect point. e =4, the artifact amplitude near the defect point in the near field area at the top of the imaging area is equivalent to the imaging amplitude at the defect point, which leads to the visual appearance of two defect points at this location, interfering with the judgment of the actual number of defect points; when the number of sparsely selected excitation array elements N e =32, there is no obvious artifact in the imaging area, and the imaging quality and imaging API parameters are similar to those of the full-matrix full-focusing algorithm, but the imaging time is greatly reduced. Therefore, considering the balance between imaging quality and imaging rate, the number of transmitting array elements N for sparse excitation full-focusing imaging is selected. e =32.

[0044] S30202: Selection of sparse receiving array elements: Although the sparse transmission full focusing imaging algorithm reduces echo data redundancy by sparsely selecting transmitting array elements without affecting the imaging quality, thereby greatly improving the imaging speed. According to the acoustic reciprocity theorem, there are still S in the echo data matrix of sparse excitation full focusing acquisition. ij With S ji The two types of acoustic beams have equal acoustic path redundant echo data with reciprocal transmitting and receiving array elements, and because the echo data matrix collected by sparse excitation full focusing is not a real symmetric matrix, it is impossible to divide the collected full matrix real symmetric echo data set into two parts from the diagonal line as in the previous optimization of the full matrix full focusing algorithm, and only perform superposition calculation on half of the data. Therefore, in order to solve the echo data redundancy problem of the sparse transmission full focusing algorithm and further improve the imaging speed, a sparse transmission and reception full focusing imaging algorithm is proposed on this basis.

[0045] The sparse transmit and receive full-focus imaging algorithm is to maintain the original N e = 32, under the premise that the excitation mode of sparse transmission full focusing remains unchanged, the full focusing mode of receiving echoes by all array elements is not adopted, but the remaining 32 receiving array elements N adjacent to the 32 excitation array elements are selected. r To perform sparse reception of echo data, the array element transmission and reception control is as follows: Figure 4 shown.

[0046] S303: Calculating the acoustic wave propagation distances between the imaging focus point and the sparse transmitting array elements and the sparse receiving array elements respectively;

[0047] S304: Calculating the delay time of the ultrasonic signal from the imaging focus point from reception to transmission based on the sound wave propagation distance;

[0048] S305: If there is a defect at the imaging point, the transmitted ultrasonic signal is reflected at the corresponding imaging point to form echo data.

[0049] S4: Ultrasound imaging algorithm focused imaging;

[0050] S401: Incorporating a GPU parallel computing acceleration framework into the sparse transceiver full-focus imaging algorithm, performing time delay compensation at corresponding imaging points on all echo data received by the receiving array element and then superimposing them to obtain a beamforming amplitude at the corresponding imaging point;

[0051] Assume that the number of array elements of the phased array sparse excitation is N e , the number of sparse array elements receiving echo signals is N r , the time taken to calculate the propagation time of an ultrasonic beam is t1, the time taken for each time from the excitation transmitting array element to the sparse array element receiving the ultrasonic echo signal is t2, and the total number of imaging points divided into the imaging area is O, then the total time T for a single frame imaging calculation of ultrasonic detection is:

[0052] T=t2×N e +2t1×N e ×N r ×M (1)

[0053] From the calculation principle of sparse transceiver full-focus imaging, it is found that the imaging time can be further optimized. Since the distance between the center of the phased array element and the already divided imaging grid points in the imaging area is unchanged each time the flexible phased array performs single-frame imaging, it is only necessary to calculate N e +N r By combining and adding the propagation time of the sub-ultrasonic sound beam, the delay result of all echo data in the echo data matrix can be obtained, which can greatly reduce the imaging calculation amount, thereby reducing the calculation time T of a single frame imaging to:

[0054] T=t2×N e +2t1×(N e +N r )×M (2)

[0055] Similarly, when the flexible phased array is fitted to the surface of the large shaft for continuous scanning and imaging, due to the symmetric shape of the large shaft, in different areas of the fitting scan, the imaging grid points divided inside the area are fixed relative to the center position of the phased array element, the material of the shaft to be measured remains unchanged, the speed of sound propagation of the ultrasonic beam and the sound field parameters in the shaft are also fixed, so the delay compensation of the echo data at the fixed imaging point is also unchanged. The delay parameters at each calculation point during the first frame imaging can be stored in the form of a matrix, and the delay matrix can be directly called in the subsequent frame calculation imaging process to further reduce the actual scanning detection imaging time. In summary, the calculation time T of sparse transceiver full-focus single-frame imaging is finally optimized and reduced to:

[0056] T=t2×N e (3)

[0057] Since both the delay time and the echo data are stored in matrices, and GPU is best at matrix parallel computing, the GPU parallel computing acceleration framework is integrated into the optimized sparse transmit-receive full-focus imaging algorithm, and a fast full-focus algorithm is proposed. The process framework of the algorithm is as follows: Figure 6 shown.

[0058] S402: traverse all imaging points in the imaging area, repeatedly calculate the synthetic beam amplitude of all imaging points in the imaging area, obtain a synthetic beam amplitude intensity distribution diagram, and output the synthetic beam amplitude intensity distribution diagram as a defect imaging diagram. During actual defect detection, the flexible phased array is fitted to the outer surface of the main shaft and slides circumferentially, so that the internal defects of the shaft body can be imaged frame by frame, and the internal health status of the horizontal tail main shaft can be judged conveniently and intuitively from the image.

[0059] The present invention proposes a sparse transmission full focusing algorithm by sparsely selecting transmitting array elements, and selects sparse receiving array elements on this basis, and proposes a sparse transmission and reception full focusing algorithm, which further improves the imaging rate and solves the problem of long time consumption of the traditional full-matrix full focusing algorithm.

[0060] like Figure 3 As shown, the number of channels of the echo data set collected by sparse transmission and reception with full focus is N e ×N r , is N e =32, N r = 64, which is half of the number of channels in the sparsely transmitted fully focused echo data set, N e =64, N r = 64, the number of channels of the full-matrix fully focused echo data set is one-fourth of that of the full-matrix fully focused echo data set. Although the echo data matrix obtained is a real symmetric matrix, since the transmitting and receiving array elements are staggered in odd and even distribution when they are selected, there is no overlap, so there is no such thing as S in the matrix. ij With S jiSuch symmetrical equal-path redundant data of the transmitting and receiving array elements are used. This method utilizes all the phased array elements while removing the redundant data, thus greatly improving the imaging speed while ensuring the imaging quality.

[0061] from Figure 5 It can be seen that the sparse transmit-receive full-focus imaging algorithm has no obvious artifacts, high imaging contrast, and the six defect points in the area can be clearly distinguished. Compared with the sparse transmit full-focus imaging algorithm, the sparse transmit-receive full-focus imaging algorithm reduces the number of echo data matrix channels by half by further sparsely selecting receiving array elements without almost affecting the imaging quality, thereby shortening the single-frame imaging time by about half. However, the sparse transmit-receive full-focusing algorithm needs to be further optimized to meet the imaging speed index requirements of real-time detection.

[0062] The imaging results of defects at different depths and angles by the fast full focusing algorithm optimized by the underlying computing principle and GPU parallel accelerated computing and the traditional full focusing algorithm are shown in the following figure. Figure 7 The imaging quality and API parameter values ​​of the two imaging algorithms are basically the same. While ensuring the imaging quality, the single-frame imaging time is shortened from 26.4s to about 0.3s, and the imaging speed is increased by 87 times, which can meet the needs of real-time detection. As the number of phased array elements used in the future increases, the fast full focusing algorithm will improve the efficiency of defect scanning.

[0063] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement made to the above embodiments without departing from the content of the technical solution of the present invention, based on the technical essence of the present invention, within the spirit and principles of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for rapid defect detection of aircraft horizontal tail shaft based on flexible phased array, characterized in that: The method for rapid detection of defects of an aircraft horizontal tail shaft based on a flexible phased array comprises the following steps: Step 1: Determine and fix the fitting position of the flexible phased array on the workpiece to be tested; Step 2: Grid the imaging area; Step 3: Transmit ultrasonic signals and collect echo data; Step 4: Based on the echo data, the defect imaging map of the imaging area is calculated to complete the rapid detection of defects in the aircraft horizontal tail shaft.

2. According to the method of claim 1, the method is characterized in that: The grid division of the imaging area in step 2 specifically includes: The imaging area is divided into two sub-areas with the center of the major axis of the aircraft horizontal tail as the pole O and the ray perpendicular to the pole as the polar axis, a polar coordinate system is established, and the polar axis and polar angle directions of the imaging area are evenly divided into grids.

3. The method for rapid defect detection of aircraft tail shaft based on flexible phased array according to claim 1 is characterized in that: The steps of transmitting ultrasonic signals and collecting echo data in step 3 include: Step 3.1: Take the grid intersection point P in the divided grid as the imaging focus point; Step 3.2: Select sparse transmit array elements and sparse transmit array elements; Step 3.3: Calculate the acoustic wave propagation distances between the imaging focus point and the sparse transmitting array element and the sparse receiving array element respectively; Step 3.4: Calculate the delay time of the ultrasonic signal from the imaging focus point from reception to transmission based on the sound wave propagation distance; Step 3.5: If there is a defect at the imaging point, the transmitted ultrasonic signal is reflected at the corresponding imaging point to form echo data.

4. The method for rapid defect detection of aircraft horizontal tail shaft based on flexible phased array according to claim 3 is characterized in that: The sparse transmit array element N described in step 3.2 e The number is 32, sparse receiving array element N r The number of is 32.

5. The method for rapid defect detection of aircraft horizontal tail shaft based on flexible phased array according to claim 1 is characterized in that: The step of calculating the defect imaging image of the imaging area in step 4 includes: Step 4.1: Integrate the GPU parallel computing acceleration framework into the sparse transceiver full-focus imaging algorithm, perform time delay compensation at the corresponding imaging point on all echo data received by the receiving array element, and then superimpose them to obtain the beamforming amplitude at the corresponding imaging point; Step 4.2: Traverse all imaging points in the imaging area, repeatedly calculate the synthetic beam amplitude of all imaging points in the imaging area, obtain a synthetic beam amplitude intensity distribution map, and output the synthetic beam amplitude intensity distribution map as a defect imaging map.

Images