A method for orthogonal synthesis and separation of ultrasonic signals

Abstract

The present invention provides an orthogonal synthesis and separation method for ultrasonic signals. First, the analog signal received by the ultrasonic sensor array element is preprocessed, and a set of mutually orthogonal trigonometric functions are used to simulate and modulate the low-frequency components in the ultrasonic echo signal. Then, the orthogonal analog modulated signal is synthesized into one analog signal by an adder, and the synthesized ultrasonic echo digital signal is obtained through a single analog-to-digital conversion channel. With the powerful parallel processing capability of FPGA and a specific digital signal processing algorithm, the parameters of the original signal can be separated, thereby obtaining the independent echo parameters of the two sensor array elements. In the array signal processing link, the dual-channel signal synthesis method can also be used to increase the amplitude strength of the signal and improve the signal-to-noise ratio of the echo signal.

Description

Technical Field

[0001] The invention belongs to the technical field of ultrasonic nondestructive testing, and in particular relates to an ultrasonic signal orthogonal synthesis and separation method. Background Art

[0002] Ultrasonic defect detection is a commonly used nondestructive testing method. In order to obtain more abundant information about the object being tested and to achieve higher quantitative accuracy during defect detection, ultrasonic sensors have also developed from single probes to arrays. With the increase in the number of array elements in the array, the advantage is that the information obtained in a single scan is richer and the detection efficiency is improved. However, it also brings the inconvenience of an increase in the number of analog signal processing channels at the receiving end. For example, in the commonly used ultrasonic phased array probe, each array element requires an independent receiving circuit. When the array element scale increases, the number of processing units increases significantly, which increases the complexity of the circuit and the production cost.

[0003] In order to reduce the number of circuits in ultrasonic sensor signal processing units and improve the transmission efficiency of information, domestic and foreign scholars have also carried out research on ultrasonic signal synthesis and parameter separation methods, mainly time-division multiplexing ultrasonic signal synthesis and parameter separation technology (Yu Jiahui. Research on hardware implementation method of sparse sampling of array ultrasonic signals based on FRI [D]. Jiangsu University, 2019.). This method uses the transmission time characteristics of ultrasonic signals in the test piece. Within the detection sound range, according to the sensor geometry and sound field transmission characteristics, the echo delay range of different channels is a priori determined, thereby realizing the time-division multiplexing synthesis of 8-channel ultrasonic sensor array element signals, and the wave arrival time and peak parameters of each signal can be effectively digitally demodulated in the back-end information processing. However, after the measured object and sensor structure parameters change, this method needs to re-calculate the a priori information, which reduces the convenience of use. It requires scientific and technological personnel with specialized knowledge and skills to implement, which is not conducive to its widespread promotion and use. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides an ultrasonic signal orthogonal synthesis and separation method to solve the problem of ultrasonic signal synthesis and separation in ultrasonic detection.

[0005] The present invention achieves the above technical objectives through the following technical means.

[0006] A method for orthogonal synthesis and separation of ultrasonic signals:

[0007] Signal synthesis:

[0008] Extract the low-frequency components of the two ultrasonic signals to obtain signals x(t) and y(t) respectively; generate two trigonometric functions g(t) and And multiply them with the signals x(t) and y(t) respectively to get x(t)g(t) and For x(t)g(t) and Add them together and get Then z(n) is obtained by analog-to-digital conversion;

[0009] Signal separation:

[0010] Generate two trigonometric function sequences g(n) and And multiply them by z(n) to get z(n)g(n) and Filter out z(n)g(n) and The high frequency components in and

[0011] Furthermore, in the signal synthesis, before extracting the low-frequency component, the ultrasonic signal is first subjected to limiting protection, impedance matching and amplification preprocessing.

[0012] Furthermore, the trigonometric function g(t) and The frequency f0≥2B, B is the main frequency bandwidth with amplitude 0 in the amplitude-frequency of signals x(t) and y(t).

[0013] Furthermore, the frequency spectra of the signals x(t) and y(t) are obtained by Fourier transform, and then the bandwidth B with an amplitude of 0 in the amplitude-frequency spectrum is taken.

[0014] Furthermore, for the trigonometric functions g(t) and Firstly, the ROM and phase-locked loop embedded in the FPGA structure are used to generate two trigonometric function sequences g(n) and g(n) with a phase difference of π / 2. Then the D / A module is used to generate the corresponding trigonometric functions g(t) and

[0015] Furthermore, the A / D module is used to convert z(t) into z(n), where the A / D sampling frequency f s ≥2(B+f0).

[0016] Furthermore, in the signal separation, z(n) is respectively multiplied with g(n) and Multiplying them, we get (n)g(n) and

[0017] Furthermore, in the signal separation, a bandpass filter is used to filter out high-frequency components.

[0018] Furthermore, the bandpass filter is a -6dB Butterworth 12th-order digital bandpass filter.

[0019] Further, from and The peak value and the corresponding delay are calculated respectively.

[0020] The beneficial effects of the present invention are:

[0021] The present invention provides a new method for synthesizing and separating ultrasonic signals. First, the analog signal received by the ultrasonic sensor array element is preprocessed, and a set of mutually orthogonal trigonometric functions are used to simulate and modulate the low-frequency components in the ultrasonic echo signal. Then, the orthogonal analog modulated signal is synthesized into one analog signal by an adder, and the synthesized ultrasonic echo digital signal is obtained through a single analog-to-digital conversion channel. With the powerful parallel processing capability of FPGA and a specific digital signal processing algorithm, the parameters of the original signal can be separated, thereby obtaining the independent echo parameters of the two sensor array elements. In the array signal processing link, the dual-channel signal synthesis method can also be used to increase the amplitude strength of the signal and improve the signal-to-noise ratio of the echo signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of the ultrasonic signal synthesis and separation method of the present invention;

[0023] FIG2(a) is a waveform diagram of the ultrasonic echo signal x0(t) in Test 1;

[0024] FIG2( b ) is a waveform diagram of the ultrasonic echo signal y0(t) in Test 1;

[0025] FIG3(a) is a waveform diagram of the low-frequency component x(t) of the echo signal in test 1;

[0026] FIG3( b ) is a waveform diagram of the low-frequency component y(t) of the echo signal in test 1;

[0027] FIG4(a) is a waveform diagram of x(t)g(t) in test 1;

[0028] Figure 4(b) shows the Waveform diagram of

[0029] Figure 5 The waveform diagram of the synthesized signal sequence z(n) in Test 1;

[0030] FIG6(a) is a waveform diagram of z(n)g(n) in Test 2;

[0031] Figure 6(b) shows the results of test 2. Waveform diagram of

[0032] Figure 7(a) shows the separated signal in test 2 Waveform diagram of

[0033] Figure 7(b) shows the separated signal in test 2 Waveform diagram of

[0034] FIG8( a ) is a waveform diagram of the ultrasonic echo signal x0(t) in Test 2;

[0035] FIG8( b ) is a waveform diagram of the ultrasonic echo signal y0(t) in Test 2;

[0036] FIG9( a ) is a waveform diagram of the low-frequency component x(t) of the echo signal in test 2;

[0037] FIG9( b ) is a waveform diagram of the low-frequency component y(t) of the echo signal in test 2;

[0038] Figure 10(a) is the waveform of x(t)g(t) in test 2;

[0039] Figure 10(b) shows the results of test 2. Waveform diagram of

[0040] Fig.11 This is the waveform diagram of the synthesized signal sequence z(n) in Test 2;

[0041] FIG12( a ) is a waveform diagram of z(n)g(n) in Test 2;

[0042] Figure 12(b) shows the results of test 2. Waveform diagram of

[0043] Figure 13(a) shows the separated signal in test 2 Waveform diagram of

[0044] Figure 13(b) shows the separated signal in test 2 Waveform diagram of . DETAILED DESCRIPTION

[0045] Embodiments of the present invention are described in detail below, examples of the illustrated embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0046] 1. Technical solution

[0047] Two ultrasonic sensors are provided, both of which have transmitting and receiving functions, and the two ultrasonic sensors are independent of each other. The two ultrasonic sensors are attached to the object to be measured, and the two ultrasonic sensors are stimulated by electric pulse signals respectively. The two ultrasonic sensors can be stimulated simultaneously or with a delay (that is, one is stimulated first and the other is stimulated later). After being stimulated, the two ultrasonic sensors generate two ultrasonic beams respectively. The ultrasonic waves are reflected after encountering defects or end faces in the object to be measured, and finally the ultrasonic sensor receives the reflected echo signal.

[0048] like Figure 1As shown, for the echo signals received by the two ultrasonic sensors, the synthesis and separation of the two ultrasonic signals are achieved through the following steps:

[0049] 1. Signal synthesis

[0050] S1, after performing limiting protection, impedance matching and amplification preprocessing on the two ultrasonic (echo) signals, extract the low-frequency components of the two ultrasonic signals, which are signals x(t) and y(t) respectively.

[0051] It should be noted that the limiting protection in the above preprocessing is due to the fact that the ultrasonic array element (ultrasonic sensor) used in this embodiment is both a transmitting unit and a receiving unit, which will cause a signal mutual coupling channel between the excitation circuit and the receiving circuit; therefore, in order to avoid excessive impact of the excitation electrical pulse on the receiving circuit, a limiting protection link is added in the preprocessing.

[0052] S2, obtain the frequency spectrum of the signal x(t) and y(t) through Fourier transform, and then take the main frequency bandwidth with an amplitude of 0 in the amplitude-frequency spectrum, that is, the bandwidth B.

[0053] S3, respectively generates two trigonometric functions g(t) and The trigonometric functions g(t) and The frequency f0≥2B.

[0054] In the specific operation, the ROM and phase-locked loop (PLL) embedded in the FPGA (field programmable gate array) structure can be used to generate two trigonometric function sequences g(n) and g(n) with a phase difference of π / 2. Then the D / A module (digital-to-analog conversion module) is used to generate the corresponding trigonometric functions g(t) and

[0055] S4, the two trigonometric functions g(t) and Use multipliers to multiply x(t) and y(t) respectively to get x(t)g(t) and

[0056] S5, using an adder to add the above multiplied signals, to obtain As a result, two analog signals share one analog channel, and the signals are mixed and superimposed in the time domain to achieve the synthesis of the two analog signals.

[0057] S6, using the A / D module (analog-to-digital conversion module) to digitize the synthesized analog signal z(t). According to the Nyquist minimum sampling frequency requirement, the A / D sampling frequency f s ≥2(B+f0), perform analog-to-digital conversion on the signal to obtain a discrete digital signal sequence z(n).

[0058] 2. Signal separation

[0059] S7, the FPGA still generates the digital sequences g(n) and

[0060] S8, in FPGA, the digital sequence multiplication method is used to obtain z(n)g(n) and

[0061] S9, use a bandpass filter to filter out z(n)g(n) and The high-frequency components in and When designing a specific bandpass filter, we can use z(n)g(n) and Spectral distribution, two -6dB Butterworth 12th-order digital bandpass filters are designed; thus z(n)g(n) and The high frequency components are filtered out by the filter to obtain and and That is, they are the estimated values ​​of the signals x(t) and y(t), respectively; from and By respectively calculating the peak value and the corresponding time delay, the echo parameters originally sensed by the two ultrasonic sensors can be obtained.

[0062] 2. Effect Test

[0063] According to the above technical solution, the ultrasonic signal is synthesized and separated. The following two groups of tests are respectively performed, wherein there is a time delay difference between the two ultrasonic echo signals in test one, while there is no time delay difference in test two. The following two groups of tests are used to demonstrate the synthesis and separation effect of the present invention on echoes with different time delays.

[0064] Test 1

[0065] 1) Synthesis

[0066] In broadband narrow pulse ultrasonic detection, the echo signals of the two ultrasonic sensor array elements are:

[0067]

[0068]

[0069] Among them, A1 and A2 represent the amplitudes of the two echo signals respectively; B1 and B2 represent the bandwidths of the two echo signals respectively; T1 and T2 represent the arrival times of the two echo signals respectively; and f1 is the center frequency of the ultrasonic sensor array element.

[0070] In this test, a group of ultrasonic signals with time delay difference are selected; the specific parameters are: amplitude A1 = 0.7V, A2 = 0.7V, bandwidth B1 = 3MHz, B2 = 3MHz, wave arrival time T1 = 20us, T2 = 22us, ultrasonic sensor center frequency f1 = 2.5MHz; the two ultrasonic signals are respectively as follows Figure 2(a) and 2(b) shown.

[0071] The low-frequency components of the above two ultrasonic signals are extracted respectively, and the corresponding signals x(t) and y(t) are obtained. The specific function expression is:

[0072]

[0073]

[0074] A3 is the gain. In this test, A3=3. The waveforms of x(t) and y(t) are as follows: Figure 3(a) and 3(b) shown.

[0075] The frequency spectrum of the echo signal x(t) and y(t) is obtained by Fourier transform, and the main frequency bandwidth B with an amplitude of 0 is taken. In this embodiment, B = 3KHz. The FPGA is combined with the built-in ROM and phase-locked loop (PLL) to generate two trigonometric function sequences g(n) and g(n) with a phase difference of π / 2. The D / A module then generates g(t) and The trigonometric function frequency f0 ≥ 2B is required. In this embodiment, f0 = 6KHz. Then the two trigonometric functions g(t) and Use multipliers to multiply x(t) and y(t) respectively to get x(t)g(t) and The waveforms are as follows Figure 4(a) , 4(b) shown.

[0076] Use the adder to add x(t)g(t) and Synthesize to obtain At this time, the two analog signals share one analog channel, and the composite signal can be digitized using a single A / D.

[0077] According to the Nyquist minimum sampling frequency requirement, the A / D sampling frequency is taken as f s =18KHz, perform analog-to-digital conversion on the synthesized signal to obtain a discrete sequence z(n). The sequence waveform is as follows Figure 5 shown.

[0078] 2) Separation

[0079] Use FPGA to generate g(t) and The corresponding digital sequence g(n) and Then, we use the digital sequence multiplication method to obtain x1(n)=z(n)g(n) and

[0080] The specific function expressions of x1(n) and y1(n) are as follows:

[0081]

[0082]

[0083] The waveforms of x1(n) and y1(n) are shown in FIG6(a) and FIG6(b) respectively.

[0084] According to the spectrum distribution of x1(n) and y1(n), two -6dB Butterworth 12th-order digital bandpass filters are designed. The passband frequency of the bandpass filter is 7KHz and the stopband frequency is 17KHz. After the synthesized digital sequence passes through the filter, its high-frequency components are filtered out and the separated signals are obtained respectively. and The specific function expression is as follows:

[0085]

[0086]

[0087] and The waveform is as Figure 7(a) and 7(b) shown.

[0088] Finally from and The maximum echo peak is solved respectively, and the corresponding delay of the point is calculated, so as to obtain the amplitude and delay parameters of the two ultrasonic echo signals.

[0089] 3. Test 2

[0090] 1) Synthesis

[0091] The overall test is based on test 1, in which the two ultrasonic echo signals x0(t) and y0(t), the echo signal amplitudes A1=0.7V, A2=1.4V, and the wave arrival time T1=T2=22us; the waveform is as follows Figure 8(a) and 8(b) shown.

[0092] Extract the low-frequency components in the echo signal to obtain x(t) and y(t). The waveform is as follows: Figure 9(a) and 9(b) shown.

[0093] Generate two trigonometric functions g(t) and And multiply by x(t) and y(t) respectively to get x(t)g(t) and Waveform diagram Figure 10(a) and 10(b) shown.

[0094] Use the adder to synthesize the two signals and get Then the synthesized signal is converted into digital form to obtain z(n). The waveform is as follows: Fig.11 shown.

[0095] 2) Separation

[0096] Generate g(n) and And use the digital sequence multiplication method to obtain x1(n)=z(n)g(n) and Waveform diagram Figure 12(a) and 12(b) shown.

[0097] Use a bandpass filter to filter out high-frequency components and get the signal and Waveform diagram Figure 13(a) and 13(b) shown.

[0098] In the description of the present invention, it should be understood that the terms "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0099] The present invention is not limited to the above-mentioned embodiments. Any obvious improvement, substitution or deformation that can be made by those skilled in the art without departing from the essential content of the present invention belongs to the protection scope of the present invention.

Claims

1. A method for orthogonal synthesis and separation of ultrasonic signals, characterized in that: Signal synthesis: Extract the low-frequency components of the two ultrasonic signals to obtain signals x(t) and y(t) respectively; generate two trigonometric functions g(t) and And multiply them with the signals x(t) and y(t) respectively to get x(t)g(t) and For x(t)g(t) and Add them together and get Then z(n) is obtained by analog-to-digital conversion; The trigonometric functions g(t) and The frequency f0≥2B, B is the main frequency bandwidth with amplitude 0 in the amplitude-frequency of the signals x(t) and y(t); the frequency spectrum of the signals x(t) and y(t) is obtained by Fourier transform, and then the bandwidth B with amplitude 0 in the amplitude-frequency is taken; For the trigonometric functions g(t) and Firstly, the ROM and phase-locked loop embedded in the FPGA structure are used to generate two trigonometric function sequences g(n) and g(n) with a phase difference of π / 2. Then the D / A module is used to generate the corresponding trigonometric functions g(t) and Signal separation: Generate two trigonometric function sequences g(n) and And multiply them by z(n) to get z(n)g(n) and Filter out z(n)g(n) and The high frequency components in and 2. The method for orthogonal synthesis and separation of ultrasonic signals according to claim 1, characterized in that: In the signal synthesis, before extracting the low-frequency component, the ultrasonic signal is first subjected to limiting protection, impedance matching and amplification preprocessing.

3. The method for orthogonal synthesis and separation of ultrasonic signals according to claim 1, characterized in that: Use the A / D module to convert z(t) to z(n), where the A / D sampling frequency is f s ≥2(B+f0).

4. The method for orthogonal synthesis and separation of ultrasonic signals according to claim 1, characterized in that: In the signal separation, z(n) is respectively multiplied with g(n) and Multiply them together to get z(n)g(n) and 5. The method for orthogonal synthesis and separation of ultrasonic signals according to claim 1, characterized in that: In the signal separation, a bandpass filter is used to filter out high-frequency components.

6. The method for orthogonal synthesis and separation of ultrasonic signals according to claim 5, characterized in that: The bandpass filter is a -6dB Butterworth 12th-order digital bandpass filter.

7. The method for orthogonal synthesis and separation of ultrasonic signals according to claim 1, characterized in that: from and The peak value and the corresponding delay are calculated respectively.

Images