Ultrasonic sparse full focusing method based on signal variance and minimum redundancy

Abstract

The invention belongs to the field of nondestructive testing, and discloses an ultrasonic sparse full focusing method based on signal variance and minimum redundancy. The method comprises the following steps of: after acquiring array signals by using full matrix capture, calculating a two-dimensional array signal variance which is transmitted by the same array element and received by all array elements, and reordering and screening the array elements according to the two-dimensional array signal variance; on the basis, combining with the minimum redundancy constraint to expand the array aperture, and constructing a sparse array according to the effective array element position; sound propagation time is obtained through the spatial position relation between the effective array elements and the grid points in the detection area, corresponding signals are searched for in full-matrix data to implement amplitude superposition, and finally ultrasonic imaging in the detection area is achieved. According to the method, the sparse array can be adaptively selected, and full-focusing imaging is performed by using a small amount of data, so that the defect detection capability and the imaging efficiency are further improved.

Description

Technical Field

[0001] The invention belongs to the field of nondestructive testing and relates to an ultrasonic sparse full focusing method based on signal variance and minimum redundancy. Background Art

[0002] In recent years, ultrasonic imaging technologies such as phased array, synthetic aperture and full focusing have developed rapidly. Among them, the full focusing method has the most ideal imaging resolution and sensitivity. This method discretizes the detection area into grid points and performs multiple time-delay superposition and virtual focusing on each grid point to achieve good defect imaging effect and recognition ability. However, the time-delay superposition algorithm based on large-scale full matrix data and high time complexity makes it difficult for the full focusing method to image quickly and in real time, which limits the application of this method in the actual industrial field (Zhou Zhenggan, et al. Research, Application and Development of Phased Array Ultrasonic Post-Processing Imaging Technology [J]. Journal of Mechanical Engineering).

[0003] Sparse arrays can reduce the amount of data and running time required for imaging by selecting a subset of the full matrix data for imaging. This is an effective way to improve the imaging efficiency of the full focusing method. Particle swarm optimization and genetic algorithms are currently commonly used array sparse methods. They mainly design sparse arrays by controlling the directivity of array beams to achieve higher imaging efficiency (Hu HW, et al, Ultrasonic sparse-TFM imaging for a two-layer medium using genetic algorithm optimization and effective aperture correction [J]. NDT&E International). However, these sparse methods only consider the physical properties of sparse arrays, such as the sound field intensity distribution and sidelobe height, and do not fully consider the correlation between the structure to be tested and the received signal. The defect characterization capability needs to be improved.

[0004] Therefore, it is necessary to develop an adaptive array sparse method that takes into account the structural characteristics of the measured structure and apply it to full-focus imaging to further improve the imaging efficiency and resolution. Summary of the invention

[0005] The present invention provides an ultrasonic sparse total focusing method based on signal variance and minimum redundancy, and its purpose is to design a sparse array construction method for adaptively selecting array elements and use it for ultrasonic total focusing imaging, and further improve the defect recognition effect and imaging efficiency of the total focusing method by combining signal characteristics with the sparse array construction.

[0006] The technical solution of the present invention:

[0007] An ultrasonic sparse full focusing method based on signal variance and minimum redundancy uses full matrix capture to obtain array signals, calculates the variance of two-dimensional array signals transmitted by the same array element and received by all array elements, and reorders and screens the array elements accordingly; on this basis, the array aperture is expanded in combination with the minimum redundancy constraint, and a sparse array is constructed according to the effective array element position; the sound propagation time is obtained by using the spatial position relationship between the effective array element and the grid point of the detection area, and the corresponding signal is found in the full matrix data to implement amplitude superposition, and finally ultrasonic imaging in the detection area is realized; the specific steps are as follows:

[0008] (a) Acoustic parameter measurement and detection parameter determination

[0009] Measure the ultrasonic longitudinal wave velocity c of the workpiece being tested; select the appropriate phased array probe type, number of array elements n, and number of sampling points N according to the material of the workpiece being tested and the range of the area to be tested t ;

[0010] (b) Obtaining the full matrix signal

[0011] Place the phased array probe above the area to be detected to collect signals and obtain the dimension N t ×n×n full matrix signal M, which consists of n 2 The time domain A scan signal is composed of the u-th point in the signal transmitted by the array element i and received by the array element j as M ij (u), where 1≤u≤N t , 1≤i≤n, 1≤j≤n;

[0012] (c) Full-matrix signal sparse processing

[0013] According to the characteristics of signals from different transmitting array elements, M is divided into n two-dimensional sub-matrices, where the sub-matrix signal transmitted by array element i and received by all array elements is defined as M i , the data dimension is N t ×n; calculate the signal variance Var(M i ), as shown in formula (1)

[0014]

[0015] Press Var(M i ) values, reorder the transmitting array elements in descending order, and take the first kn array elements to form the array P Var , k is a coefficient related to the probe center frequency, sampling frequency, number of array elements and spacing, 0<k≤1; the array redundancy R is calculated by formula (2), and the minimum redundancy constraint requires that the redundancy is always not less than 1 under any array element spacing, as shown in formula (3)

[0016]

[0017] Where p represents the array element spacing, P(q) and P(q+p) are the function values ​​at the position q and q+p of the phased array. When there is a valid array element at this position, P = 1, otherwise P = 0;

[0018] Combining formula (3) in P Var Add the minimum number of valid array elements to satisfy the minimum redundancy constraint as the final sparse array P Var-MRLA , the total number of array elements is N, and the array satisfies the constraints in formula (4);

[0019]

[0020] (d) Ultrasound Total Focusing Imaging

[0021] Extract from M the sparse array P Var-MRLA The signal is used as the data for ultrasound full focusing imaging. This data is a subset of M and is defined as M Var-MRLA ; Establish a coordinate system with the center position of the first array element in the phased array probe as the origin, the array element arrangement direction as the positive direction of the x-axis, and the positive direction of the z-axis perpendicular to the array element arrangement direction; discretize the area to be detected into several grid points according to a fixed accuracy, and the relative position relationship between the grid points and the array elements can be calculated by formula (5)

[0022]

[0023] In the formula, (x,z), (x i ,z i ), (x j ,z j ) represent the coordinates of discrete grid points, transmitting array element i and receiving array element j respectively; according to the ultrasonic propagation time t ij (x,z), from the signal M Var-MRLA The corresponding signal of the grid point (x, z) in the index is processed by delay superposition as shown in equation (6) to obtain the sparse fully focused image I(x, z) of the grid point;

[0024]

[0025] The strongest peak position of the image is read as the defect point, and the defect is quantified in the -6dB range.

[0026] Beneficial effects of the present invention: The present invention is based on an ultrasonic sparse full focusing method based on signal variance and minimum redundancy. The full matrix data is split into two-dimensional arrays according to different sources and its signal variance is calculated. The array elements are reordered and selected accordingly, and the minimum number of effective array elements is added on this basis to meet the minimum redundancy constraint. The sparse array composed of these effective array elements simultaneously considers the influence of the structural characteristics to be measured and the array aperture on the full focusing method, and realizes ultrasonic imaging in the detection area quickly and with high quality. The method can adaptively select array elements according to the relative position relationship between the phased array and the defect, further improving the defect recognition effect and imaging efficiency of the full focusing method. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a schematic diagram of the ultrasonic testing system used.

[0028] Figure 2 It is a schematic diagram of an aluminum alloy test block with adjacent transverse through holes processed.

[0029] Figure 3 The adjacent STH imaging result using the original full matrix signal.

[0030] Figure 4 The adjacent STH imaging results using sparse full matrix signals. DETAILED DESCRIPTION

[0031] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and technical solutions.

[0032] Ultrasonic sparse full focusing method based on signal variance and minimum redundancy, ultrasonic signal acquisition and detection system such as Figure 1 As shown, the specific processing steps are as follows:

[0033] (a) The test block to be tested is a 40.0 mm thick aluminum alloy sample with a longitudinal wave velocity of c = 6300 m / s. Two transverse holes with a spacing of 4.0 mm are machined inside the test block, with a center depth and radius of 15.0 mm and 1.0 mm respectively. Figure 2 shown.

[0034] (b) The parameters of the phased array probe used for full matrix signal acquisition are: center frequency 2.25MHz, number of array elements n = 32. The phased array probe is placed directly above the defect to perform signal acquisition. According to the thickness of the test block, the defect location and the probe parameters, the number of sampling points N for signal acquisition is t is 2000, and the final size of the full matrix signal M is 2000×32×32.

[0035] (c) When any array element i transmits ultrasound, it will be received by 32 receiving array elements to obtain a two-dimensional array signal M i . Calculate the variance Var(M) of these received signalsi ), and take the first 8 array elements in the order of variance from large to small to form array P Var .P Var The array elements in the array are always taken as effective array elements. When the sparse array satisfies that the redundancy is always greater than or equal to 1 under any array element spacing and the number of effective array elements in the array is the minimum, it is taken as the sparse array P for the full focusing method. Var-MRLA .for Figure 2 The test block shown, P Var-MRLA There are 13 effective array elements in total.

[0036] (d) Establish a coordinate system with the center position of the first array element as the origin, and discretize the detection area into a grid of 0.1 mm × 0.1 mm. Var-MRLA The effective array element corresponding data M Var-MRLA , establish the acoustic-temporal relationship between the effective array element and each grid point and perform amplitude superposition to obtain a sparse fully focused image.

[0037] (e) Figure 3 and Figure 4 The full matrix signal M and the sparse full matrix signal M are given respectively. Var-MRLA The full-focus imaging result when the two defects are centered is shown in Figure 2. The amplitude difference between the two defects and the array performance index (API) are used to characterize the imaging performance. The smaller the amplitude difference and the lower the API value, the higher the imaging performance. The imaging results show that the use of M and M Var-MRLA Defect quantification can be performed by performing full-focus imaging, and the defect depth measurement error does not exceed 0.4mm, and the spacing measurement error does not exceed 0.2mm. Among them, the defect image API value obtained by using M is 1.04, the difference between the two defect amplitudes is 1.46dB, and the imaging time is 512.2s; using M Var-MRLA During imaging, the API value of the defect image was 0.97, the amplitude difference between the two defects was 0.10dB, and the imaging time was 84.5s. By comparison, when using sparse full matrix data based on signal variance and minimum redundancy for imaging, the API value was reduced by 6.7%, the amplitude difference between defects decreased from 1.46dB to 0.1dB, and the imaging efficiency was increased by 6.1 times. The imaging quality is better than the original full matrix data imaging, and the imaging time is greatly shortened.

[0038] The description presented in the above exemplary embodiments is only used to illustrate the technical solution of the present invention, and is not intended to be exhaustive, nor is it intended to limit the present invention to the precise form described. Obviously, it is possible for a person of ordinary skill in the art to make many changes and variations based on the above teachings. The exemplary embodiments are selected and described to explain the specific principles of the present invention and its practical application, so that other technicians in the field can easily understand, implement and use the various exemplary embodiments of the present invention and its various selected forms and modified forms. The scope of protection of the present invention is intended to be defined by the attached claims and their equivalent forms.

Claims

1. An ultrasonic sparse full focusing method based on signal variance and minimum redundancy, characterized in that: Here are the steps: (a) Acoustic parameter measurement and detection parameter determination Measure the ultrasonic longitudinal wave velocity c of the workpiece being tested; select the phased array probe type, the number of array elements n, and the number of sampling points N according to the material of the workpiece being tested and the range of the area to be tested t ; (b) Obtaining the full matrix signal Place the phased array probe above the area to be detected to collect signals and obtain the dimension N t ×n×n full matrix signal M, the full matrix signal M is composed of n 2 The time domain A scan signal is composed of the u-th point in the signal transmitted by the array element i and received by the array element j as M ij (u), where 1≤u≤N t , 1≤i≤n, 1≤j≤n; (c) Full-matrix signal sparse processing According to the characteristics of signals from different transmitting array elements, M is divided into n two-dimensional sub-matrices, where the sub-matrix signal transmitted by array element i and received by all array elements is defined as M i , the data dimension is N t ×n; calculate the signal variance Var(M i ), as shown in formula (1): Press Var(M i ) values, reorder the transmitting array elements in descending order, and take the first kn array elements to form the array P Var , k is a coefficient related to the probe center frequency, sampling frequency, and number of array elements, 0<k≤1; the array redundancy R is calculated by formula (2), and the minimum redundancy constraint requires that the redundancy is always not less than 1 under any array element spacing, as shown in formula (3): Where p represents the array element spacing, P(q) and P(q+p) are the function values ​​at the position q and q+p of the phased array. When there is a valid array element at this position, P = 1, otherwise P = 0; Combining formula (3) in P Var Add the minimum number of valid array elements to satisfy the minimum redundancy constraint as the final sparse array P Var-MRLA , the total number of array elements is N, and the array satisfies the constraints in formula (4); (d) Ultrasound Total Focusing Imaging Extract from M the sparse array P Var-MRLA The signal is used as the data for ultrasound full focusing imaging. This data is a subset of M and is defined as M Var-MRLA ; Establish a coordinate system with the center position of the first array element in the phased array probe as the origin, the array element arrangement direction as the positive direction of the x-axis, and the direction perpendicular to the array element arrangement direction downward as the positive direction of the z-axis; discretize the area to be detected into a number of grid points according to a fixed accuracy, and the relative position relationship between the grid points and the array elements is calculated by formula (5); In the formula, (x,z), (x i ,z i ), (x j ,z j ) represent the coordinates of discrete grid points, transmitting array element i and receiving array element j respectively; according to the ultrasonic propagation time t ij (x,z), from the signal M Var-MRLA The corresponding signal of the grid point (x, z) in the index is processed by delay superposition as shown in equation (6) to obtain the sparse fully focused image I(x, z) of the grid point; The strongest peak position of the image is read as the defect point, and the defect is quantified in the -6dB range.

Images