A steel rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology

Abstract

The application discloses a kind of steel rail fracture monitoring methods based on multivariate coding ultrasonic guided wave technology, first multivariate coding sequence is based on traditional binary coding Barker code and binary orthogonal complementary Golay code, then through series mathematical operation, form a new multivariate excitation coding sequence, through the multivariate coding sequence modulation ultrasonic guided wave single pulse excitation signal, form new coding excitation waveform, monitor steel rail fracture damage by new coding excitation waveform.The multivariate coding excitation ultrasonic guided wave signal proposed in the application combines the advantages of two basic coding methods, increases the flexibility of coding length, improves the signal-to-noise ratio gain and peak sidelobe level of multivariate coding excitation ultrasonic guided wave signal, improves the working efficiency of steel rail fracture monitoring.

Description

Technical Field

[0001] This invention belongs to the field of nondestructive testing technology, specifically relating to a rail fracture monitoring method based on multi-element encoded ultrasonic guided wave technology. Background Technology

[0002] Non-destructive testing (NDT) utilizes the changes in thermal, acoustic, electrical, optical, and magnetic responses caused by the presence of anomalies and defects in the internal structure of materials to evaluate structural anomalies and defects. In other words, it is an emerging discipline that detects the presence of cracks, inclusions, or other defects in the internal structure, physical properties, or state of the workpiece or material being tested without damaging it. Ultrasonic guided waves are generated when ultrasonic waves are confined within the boundary of a waveguide medium such as a rod or tube, resulting in repeated reflections at the boundary. Compared to traditional ultrasonic waves, ultrasonic guided wave monitoring offers a relatively lower monitoring frequency and a longer transmission and monitoring distance. Ultrasonic guided wave technology is currently widely used in the health monitoring of rail and pipeline structures.

[0003] In structural health monitoring technology based on ultrasonic guided wave technology, enhancing the excitation energy of the signal, improving monitoring distance and resolution, and enhancing the qualitative and quantitative assessment of damage are among the key technologies for non-destructive structural health monitoring. A common approach is to increase the transmission voltage, but this complicates the hardware circuitry and places higher demands on the withstand voltage of the ultrasonic transducer. Encoding compression technology, under the premise of existing circuitry and the withstand voltage of the ultrasonic transducer, can indirectly increase the transmission energy by transmitting multiple continuous waves. The receiving end continuously receives the signal and uses pulse compression technology to concentrate the received signal energy.

[0004] Many scholars have conducted in-depth research on coding compression methods. However, some researchers have studied the combination of the two, which improved the signal-to-noise ratio and peak sidelobe level, but still required two encoding and decoding operations to achieve a single compression. Some scholars have focused on how to reduce the two encoding and decoding operations of GolayA and GolayB codes to a single encoding and decoding operation. However, compared to the previous two encoding and decoding operations of GolayA and GolayB codes, this only improved the efficiency of encoding and decoding, without improving the signal-to-noise ratio and peak sidelobe level. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a rail fracture monitoring method based on multi-element encoded ultrasonic guided wave technology, which improves the energy of the entire rail fracture monitoring system and helps to improve the signal-to-noise ratio of the echo signal.

[0006] Technical Solution: This invention provides a rail fracture monitoring method based on multi-element encoded ultrasonic guided wave technology, specifically including the following steps:

[0007] (1) Perform convolution operation on the A code sequence and the Barker code sequence in the Golay code to form a new multi-element coding sequence, denoted as GA(m)_BK(n); perform convolution operation on the B code sequence and the Barker code sequence in the Golay code to form a new multi-element coding sequence, denoted as GB(m)_BK(n).

[0008] (2) Perform convolution operation on the multi-encoding sequence GA(m)_BK(n) and the multi-encoding sequence GB(m)_BK(n) to form a new multi-encoding sequence, denoted as GAB(m)_BK(n);

[0009] (3) The waveform S(t) of the sine wave modulated by the Hanning window is modulated by the binary coding sequences GA(m), GB(m) and BK(n). The waveforms after modulation by the binary coding sequences are denoted as GA_S(t), GB_S(t) and BK_S(t), respectively.

[0010] (4) The waveform of the sinusoidal wave modulated by the Hanning window is modulated by the multi-element coding sequence GA(m)_BK(n) and GB(m)_BK(n). The waveforms after modulation by the binary coding sequence are denoted as GA_BK_S(t) and GB_BK_S(t), respectively.

[0011] (5) The waveform S(t) of the sinusoidal wave modulated by the Hanning window is modulated by the multi-electrode coding sequence GAB(m)_BK(n). The waveform after modulation by the multi-electrode coding sequence is denoted as GAB_BK_S(t), which is the multi-electrode coding excitation ultrasonic guided wave signal used to monitor rail fracture.

[0012] (6) After the excitation signal is boosted and amplified by the power amplifier, the multi-element coded ultrasonic guided wave signal is excited by the ultrasonic transducer. After passing through the rail medium, when the rail is broken, the multi-element coded ultrasonic guided wave signal will be reflected when it encounters the damaged part. Then, the receiver at the excitation end of the multi-element coded ultrasonic guided wave receives an echo signal, denoted as Echo_signal(t). The echo signal is deconvolved with the GB(m)_BK(n) signal in step (1) to obtain the echo signal Echo_GA_BK_Signal(t). The echo signal is deconvolved with the GA(m)_BK(n) signal in step (1) to obtain the echo signal Echo_GB_BK_Signal(t).

[0013] (7) When the multi-encoded excitation ultrasonic guided wave signal encounters rail fracture damage, the echo signals Echo_GA_BK_Signal(t) and Echo_GB_BK_Signal(t) are pulse compressed using multi-encoded GA_BK code and GB_BK code respectively to obtain Echo_GA_BK_Compress(t) and Echo_GB_BK_Compress(t) respectively. Then, vector summation is performed to obtain the echo signal Echo_Compress_Signal(t). This signal is the pulse compression echo signal returned when the multi-encoded excitation ultrasonic guided wave encounters rail fracture damage during rail monitoring.

[0014] Furthermore, the calculation formulas for the multivariate sequences GA(m)_BK(n) and GB(m)_BK(n) mentioned in step (1) are as follows:

[0015] GA(m)_BK(n)=GA(m)*BK(n)

[0016] GB(m)_BK(n)=GB(m)*BK(n)

[0017] In the formula, * represents the convolution operator.

[0018] Furthermore, the formula for calculating the multivariate sequence GAB(m)_BK(n) mentioned in step (2) is as follows:

[0019] GAB(m)_BK(n)=GA(m)_BK(n)*GB(m)_BK(n)

[0020] In the formula, * represents the convolution operator.

[0021] Furthermore, the calculation formulas for the waveforms GA_S(t), GB_S(t), and BK_S(t) modulated by the binary coded sequence in step (3) are as follows:

[0022] GA_S(t)=GA(m)·S(t)

[0023] GB_S(t)=GB(m)·S(t)

[0024] BK_S(t) = BK(n)·S(t)

[0025] The formula for calculating the Hanning window modulated sine wave, S(t), is:

[0026]

[0027] In the formula, · represents the product operation symbol, f is the center frequency of the Hanning window modulated sine wave, and N Tt is the number of periods of the Hanning window modulated sine wave, t is the duration of the Hanning window modulated sine wave waveform, and A is the amplitude of the Hanning window modulated sine wave.

[0028] Furthermore, the calculation formulas for the waveforms GA_BK_S(t) and GB_BK_S(t) after modulation of the multi-code sequence in step (4) are as follows:

[0029] GA_BK_S(t)=GA(m)_BK(n)·S(t)

[0030] GB_BK_S(t)=GB(m)_BK(n)·S(t)

[0031] In the formula, · represents the product operation symbol, and S(t) represents the waveform of the Hanning window modulated sine wave.

[0032] Furthermore, the calculation formula for the multi-electrode coded excitation ultrasonic guided wave monitoring rail signal GAB_BK_S(t) in step (5) is as follows:

[0033] GAB_BK_S(t)=GAB(m)_BK(n)·S(t)

[0034] In the formula, · represents the product symbol, and S(t) represents the waveform of the Hanning window modulated sine wave.

[0035] Step (6) is achieved through the following formula:

[0036] Echo_signa((t)=h())·GAB_BK_S(t).

[0037] Echo_GA_BK_Signal(t)=h())·GAB_BK_S(t)* -1 GB(m)_BK(n)

[0038] Echo_GB_BK_Signal(t)=h(x)·GAB_BK_S(t)* -1 GA(m)_BK(n)

[0039] In the formula, · represents the product operator, and * represents the multiplication operator. -1 Let h(x) be the deconvolution operator, and h(x) be the medium transfer function.

[0040] Furthermore, the calculation formulas for Echo_GA_BK_Compress(t), Echo_GB_BK_Compress(t), and Echo_Compress_Signal(t) mentioned in step (7) are as follows:

[0041] Echo_GA_BK_CmopRess(t)=Echo_GA_BK_signal(t)*GA_BK_S(-t)

[0042] Echo_GB_BK_CmopRess(t)=Echo_GB_BK_signal(t)*GB_BK_S(-t)

[0043] Echo_Compress_Signal(t)

[0044] =Echo_gA_BK_Cmopress(t)+Echo_GB_BK_Cmopress(t)

[0045] In the formula, * represents the convolution operator, and GA_BK_S(-t) and GB_BK_S(-t) are respectively...

[0046] The mirrored signal waveforms of GA_BK_S(t) and GB_BK_S(t) about the y-axis.

[0047] Beneficial effects: Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention is based on Barker code and positive interactive Golay code, and through a series of mathematical operations, it can improve the excitation transmission energy of ultrasonic guided waves, increase the flexibility of the coding length, improve the signal-to-noise ratio and peak sidelobe level, and avoid the disadvantage of requiring two excitation transmissions and two receptions to complete one coding; The present invention improves the energy of the entire rail fracture monitoring system, which helps to improve the signal-to-noise ratio of the echo signal. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the present invention;

[0049] Figure 2 This is a sequence diagram of the binary encoding Barker code and Golay code, and the newly generated multi-element encoding sequence in this invention; wherein, (a) is a sequence diagram of the binary sequence GolayA code; (b) is a sequence diagram of the binary sequence GolayB code; (c) is a sequence diagram of the binary sequence Barker code; (d) is a sequence diagram of the multi-element sequence GA(4)_BK(3) code; (e) is a sequence diagram of the multi-element sequence GB(4)_BK(3) code; and (f) is a sequence diagram of the multi-element sequence GAB(4)_BK(3) code.

[0050] Figure 3 This is a sine wave waveform diagram of five cycles of Hanning window modulation in this invention;

[0051] Figure 4These are the single-pulse waveforms of the binary coded Barker code and Golay code, and the newly generated multi-coded sequence in this invention; wherein, (a) is the single-pulse waveform of the binary sequence GolayA code sequence; (b) is the single-pulse waveform of the binary sequence GolayB code sequence; (c) is the single-pulse waveform of the binary sequence Barker code sequence; (d) is the single-pulse waveform of the multi-coded GA(4)_BK(3) code sequence; (e) is the single-pulse waveform of the multi-coded GB(4)_BK(3) code sequence; and (f) is the single-pulse waveform of the multi-coded GAB(4)_BK(3) code sequence.

[0052] Figure 5 This is the echo signal waveform of the multi-encoded excitation ultrasonic guided wave after passing through the rail medium in this invention;

[0053] Figure 6 These are the deconvolution decoding waveforms of the rail fracture damage echo signal in this invention; where (a) is the echo deconvolution decoding waveform diagram; and (b) is the echo deconvolution decoding waveform diagram.

[0054] Figure 7 These are the waveforms of the rail fracture damage echo signal pulse compression and decoding in this invention; wherein, (a) is the echo pulse compression and decoding waveform diagram; (b) is the echo pulse compression and decoding waveform diagram; and (c) is the final decoded waveform diagram of the echo pulse compression.

[0055] Figure 8 These are pulse compression echo signal diagrams when ultrasonic guided waves are excited to monitor rails using conventional binary-coded Barker codes and Golay codes in this invention. (a) is a pulse compression waveform diagram of the binary-coded Barker code; (b) is a pulse compression waveform diagram of the binary-coded Golay code.

[0056] Figure 9 The gain values ​​of amplitude and energy changes of signals from different coded excitation types monitoring rails at different rail fracture depths are shown. Among them, (a) represents the gain values ​​of amplitude difference changes of traditional single-pulse ultrasonic guided waves, two types of binary coded excitation ultrasonic guided waves, and multi-coded excitation ultrasonic guided waves at different rail fracture depths.

[0057] (b) is the gain of the energy difference of traditional single-pulse ultrasonic guided wave, two types of binary coded excited ultrasonic guided waves and multi-coded excited ultrasonic guided waves at different rail fracture depths. Detailed Implementation

[0058] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] like Figure 1 As shown, the present invention provides a rail fracture monitoring method based on multi-element encoded ultrasonic guided wave technology, the implementation process of which is as follows:

[0060] Convolve the A-code sequence and the Barker code sequence in the Golay code to form a new multi-element coding sequence, denoted as GA(m)_BK(n); convolve the B-code sequence and the Barker code sequence in the Golay code to form a new multi-element coding sequence, denoted as GB(m)_BK(n).

[0061] GA(m)_BK(n)=GA(m)*BK(n)

[0062] GB(m)_BK(n)=GB(m)*BK(n)

[0063] In the formula, * represents the convolution operator.

[0064] The multivariate coding sequences GA(m)_BK(n) and GB(m)_BK(n) are convolved to form a new multivariate coding sequence, denoted as GAB(m)_BK(n):

[0065] GAB(m)_BK(n)=GA(m)_BK(n)*GB(m)_BK(n).

[0066] The waveform S(t) of a sine wave modulated by a Hanning window is modulated using binary coded sequences GA(m), GB(m), and BK(n). The waveforms after modulation by the binary coded sequences are denoted as GA_S(t), GB_S(t), and BK_S(t), respectively.

[0067] GA_S(t)=GA(m)·S(t)

[0068] GB_S(t)=GB(m)·S(t)

[0069] BK_S(t) = BK(n)·S(t)

[0070] The formula for calculating the Hanning window modulated sine wave, S(t), is:

[0071]

[0072] In the formula, · represents the product operation symbol, f is the center frequency of the Hanning window modulated sine wave, and N T t is the number of periods of the Hanning window modulated sine wave, t is the duration of the Hanning window modulated sine wave waveform, and A is the amplitude of the Hanning window modulated sine wave.

[0073] The waveform of the sine wave modulated by the Hanning window is modulated using the multi-element coded sequences GA(m)_BK(n) and GB(m)_BK(n). The waveforms after modulation by the binary coded sequences are denoted as GA_BK_S(t) and GB_BK_S(t), respectively.

[0074] GA_BK_S(t)=GA(m)_BK(n)·S(t)

[0075] GB_BK_S(t)=GB(m)_BK(n)·S(t)

[0076] In the formula, · represents the product operation symbol, and S(t) represents the waveform of the Hanning window modulated sine wave.

[0077] The waveform S(t) of the sinusoidal wave modulated by the Hanning window is modulated using the multi-electrode coding sequence GAB(m)_BK(n). The waveform after multi-electrode coding sequence modulation is denoted as GAB_BK_S(t), which is the multi-electrode coded excitation ultrasonic guided wave signal used to monitor rail fracture.

[0078] GAB_BK_S(t)=GAB(m)_BK(n)·S(t)

[0079] In the formula, · represents the product symbol, and S(t) represents the waveform of the Hanning window modulated sine wave.

[0080] After the excitation signal is amplified by a power amplifier, it is used to excite the multi-element coded ultrasonic guided wave signal through an ultrasonic transducer. After passing through the rail medium, when the rail is damaged, the multi-element coded ultrasonic guided wave signal will be reflected when it encounters the damaged part. The receiver at the excitation end of the multi-element coded ultrasonic guided wave receives an echo signal, denoted as Echo_signal(t). The echo signal is deconvolved with the step GB(m)_BK(n) signal to obtain the echo signal Echo_GA_BK_Signal(t). The echo signal is deconvolved with the GA(m)_BK(n) signal to obtain the echo signal Echo_GB_BK_Signal(t).

[0081] The medium is transmitted through the rail, and its transfer function is denoted as h(x). When the echo signal Echo_signal(t) is received, the formula for calculating the echo signal when the coded excitation ultrasonic guided wave encounters damage is:

[0082] Echo_signal(t)=h(x)·GAB_BK_S(t)

[0083] The formulas for calculating the echo signals Echo_GA_BK_Signal(t) and Echo_GB_BK_Signal(t) are as follows:

[0084] Echo_GA_BK_Signal(t)=h(x)·GAB_BK_S(t)* -1 GB(m)_BK(n)

[0085] Echo_GB_BK_Signal(t)=h(x)·GAB_BK_S(t)* -1 GA(m)_BK(n)

[0086] In the formula, · represents the product operator, and * represents the multiplication operator. -1 This is the deconvolution operator.

[0087] When the multi-encoded excited ultrasonic guided wave signal encounters rail fracture damage, the echo signals Echo_GA_BK_Signal(t) and Echo_GB_BK_Signal(t) are pulse-compressed using multi-encoded GA_BK and GB_BK codes, respectively, to obtain Echo_GA_BK_Compress(t) and Echo_GB_BK_Compress(t). Then, vector summation is performed to obtain the echo signal Echo_Compress_Signal(t). This signal is the pulse-compressed echo signal returned when the multi-encoded excited ultrasonic guided wave monitors the rail and encounters rail fracture damage.

[0088] Echo_GA_BK_CmopRess(t)=Echo_GA_BK_signal(t)*GA_BK_S(-t)

[0089] Echo_GB_BK_CmopRess(t)=Echo_GB_BK_signal(t)*GB_BK_S(-t)

[0090] Echo_Compress_Signal(t)

[0091] =Echo_GA_BK_Cmo0Ress(t)+Echo_GB_BK_CmopRess(t)

[0092] In the formula, * represents the convolution operator, and GA_BK_S(-t) and GB_BK_S(-t) are respectively...

[0093] Mirror waveforms of GA_BK_S(t) and GB_BK_S(t) about the y-axis

[0094] In this embodiment, a 4-bit GolayA code (encoded as 111-1), a GolayB code (encoded as 11-11), and a 3-bit Barker code (encoded as 11-1) are used as examples.

[0095] S1: Perform convolution operation on the A code (GA(4)) sequence and the Barker code (BK(3)) sequence in the Golay code to form a new multi-element coding sequence, denoted as GA(4)_BK(3); perform convolution operation on the B code (GB(4)) sequence and the Barker code (BK(3)) sequence in the Golay code to form a new multi-element coding sequence, denoted as GB(4)_BK(3).

[0096] S2: Perform convolution operation on the multivariate coding sequence GA(4)_BK(3) and the multivariate coding sequence GB(4)_BK(3) to form a new multivariate coding sequence, denoted as GAB(4)_BK(3).

[0097] The above encoded sequence is as follows Figure 2 As shown, where:

[0098] The binary sequence GolayA code: GA(4): = [111-1]; the sequence is as follows Figure 2 As shown in (a).

[0099] The binary sequence GolayB code: GB(4): = [11-11]; the sequence is as follows Figure 2 As shown in (b).

[0100] The binary sequence Barker code is: BK(3): = [11-1]; the sequence is as follows: Figure 2 As shown in (c).

[0101] Multivariate sequence: GA(4)_BK(3)=[121-1-21]; sequence as follows Figure 2 As shown in (d).

[0102] Multivariate sequence: GB(4)_BK(3)=[12-1-12-1]; sequence as follows Figure 2 As shown in (e).

[0103] Multivariate sequence: GAB(4)_BK(3)=[144-2-505-2-44-1]; sequence as follows Figure 2 As shown in (f).

[0104] S3: The waveform of the sine wave modulated by the Hanning window is obtained by modulating the above coding sequence. The sine wave waveform S(t) is used in rail fracture monitoring. The period N of the signal sine wave is set. T Five cycles were used, with amplitude A set to 1V, center frequency set to 30kHz, and t representing the duration of the five cycles, approximately 1678s.

[0105]

[0106] Hanning window modulation 5-cycle waveform, such as Figure 3 As shown.

[0107] S4: The waveforms S(t) of the sinusoidal waves modulated by the Hanning window are respectively denoted as GA_S(t), GB_S(t), BK_S(t), GA_BK_S(t), GB_BK_S(t), and GAB_BK_S(t), as shown below. Figure 4 As shown, the calculation formula is:

[0108] GA_S(t)=GA(4)·S(t)

[0109] GB_S(t)=GB(4)·S(t)

[0110] BK_S(t)=BK(3)·S(t)

[0111] GA_BK_S(t)=GA(4)_BK(3)·S(t)

[0112] GB_BK_S(t)=GB(4)_BK(3)·S(t)

[0113] GAB_BK_S(t)=GAB(4)_BK(3)·S(t).

[0114] S5: The multi-code modulated waveform GAB_BK_S(t) is transmitted through the rail. The transfer function of the rail's transmission medium is denoted as h(x). Here, the transfer function h(x) is set to 1. At this time, the echo signal Echo_signal(t) is received, as shown below. Figure 5 As shown, the formula for calculating the rail fracture damage echo signal is:

[0115] Echo_signal(t)=1·GAB_BK_S(t).

[0116] S6: Perform a series of pulse compression decoding on the rail fracture echo signal Echo_signal(t). First, perform deconvolution operations with GB(m)_BK(n) and GA(m)_BK(n) respectively to obtain the echo deconvolution decoded signals Echo_GA_BK_Signal(t) and Echo_GB_BK_Signal(t). The calculation formula is:

[0117] Echo_GA_BK_Signal(t)=1·GAB_BK_S(t)* -1 GB(m)_BK(n)

[0118] Echo_GB_BK_Signal(t)=1·GAB_BK_S(t)* -1GA(m)_BK(n)

[0119] In the formula, · represents the product operator, and * represents the multiplication operator. -1 This is the deconvolution operator. The signal waveforms of the rail fracture echo signal deconvolution decoded signals Echo_GA_BK_Signal(t) and Echo_GB_BK_Signal(t) are as follows: Figure 6 As shown, (a) is the waveform diagram of echo deconvolution decoding; (b) is the waveform diagram of echo deconvolution decoding.

[0120] S7: The deconvolution decoded signals Echo_GA_BK_Signal(t) and Echo_GB_BK_Signal(t) of the rail fracture echo are subjected to multi-element encoding GA_BK code and GB_BK code sequence pulse compression, respectively, to obtain Echo_GA_BK_Compress(t) and Echo_GB_BK_Compress(t), respectively. Then, vector summation is performed to obtain the echo signal Echo_Compress_Signal(t). The calculation formulas for Echo_GA_BK_Compress(t), Echo_GB_BK_Compress(t), and Echo_Compress_Signal(t) are as follows:

[0121] Echo_GA_BK_Cmopress(t)=Echo_GA_BK_signal(t)*GA_S(-t)

[0122] Echo_GB_BK_Cmopress(t)=Echo_GB_BK_signal(t)*GB_S(-t)

[0123] Echo_Compress_Signal(t)

[0124] =Echo_GA_BK_Cmop1ess(t)+Echo_GB_BK_Cmopress(t)

[0125] In the formula, * represents the convolution operator, and GA_S(-t) and GB_S(-t) are the mirror waveforms of GA_S(t) and GB_S(t) about the y-axis, respectively. The signal waveforms of Echo_GA_BK_Compress(t), Echo_GB_BK_Compress(t), and Echo_Compress_Signal(t) are shown below. Figure 7 As shown, (a) is the echo pulse compression decoding waveform; (b) is the echo pulse compression decoding waveform; and (c) is the final echo pulse compression decoding waveform.

[0126] S8: To further illustrate the performance of this multi-element coded ultrasonic guided wave excitation signal for rail fracture monitoring, a comparison was made with the performance of ultrasonic guided wave excitation encoded using separate binary coded Barker and Golay codes. The 3-bit Barker code and 4-bit Golay code are used. The excitation waveforms of the binary coded excitation signals (3-bit Barker code and 4-bit Golay A and Golay B codes, used for rail fracture monitoring) are shown in step S4 as BK_S(t), GA_S(t), and GB_S(t). The corresponding pulse compressions are denoted as Barker_Compress(t), GA_Compress(t), and GB_Compress(t); summing the vectors of GA_Compress(t) and GB_Compress(t) yields the Golay code compressed signal Golay_Compress(t). The calculation formulas for Barker_Compress(t), GA_Compress(t), GB_Compress(t), and Golay_Compress(t) are as follows:

[0127] Barker_Cmopress(t)=BK_S(t)*BK_S(-t)

[0128] GA_Cmopress(t)=GA_S(t)*GA_S(-t)

[0129] GB_Cmopress(t)=GB_S(t)*GB_S(-t)

[0130] Golay_Cmopress(t)=GA_Cmopress(t)+GB_Cmopress(t)

[0131] The decoded waveforms of Barker and Golay codes, Barker_Compress(t) and Golay_Compress(t), are as follows: Figure 8 As shown, (a) is a pulse compression waveform of the binary-coded Barker code; (b) is a pulse compression waveform of the binary-coded Golay code.

[0132] S9: When evaluating coding compression performance, the relationship between the main lobe peak power and average power, and the signal-to-noise ratio gain are two key parameters, denoted as PSL and SNR, respectively. gain ,P peak P represents the peak power of the main lobe of the signal. mean The average power of the signal is A. bm Let A be the signal strength of the main lobe of the encoded signal, and let A be the signal strength of a simple conventional pulse. cgCalculations show that the PSL value for multi-level coding is 16.2039 dB, the PSL value for the binary code Barker (3 bits) is 10.0126 dB, and the PSL value for the binary code Golay complementary code (4 bits) is 14.0685 dB. The gain SNR obtained by multi-level coding... gain The gain SNR obtained by binary-coded Barker codes is 71.48 dB. gain The gain SNR obtained by binary-coded Golay (A, B) codes is 53.41 dB. gain It is 61.94dB.

[0133]

[0134]

[0135] The above-mentioned coded excitation signals were applied to rail fracture monitoring. The coded excitation signals BK_S(t), GA_S(t), GB_S(t), and GAB_BK_S(t) were modulated by a single-pulse ultrasonic guided wave signal S(t), a binary coded 3-bit Barker code (BK(3)), a 4-bit positive complementary Golay code (GA(4) and GB(4)), and a multi-element coded sequence (GAB(4)_BK(3)). The excitation waveforms are shown below. Figure 3 and Figure 4 As shown, Figure 4 In the diagram, (a) is a single pulse of the binary sequence GolayA code modulation; (b) is a single pulse of the binary sequence GolayB code modulation; (c) is a single pulse of the binary sequence Barker code modulation; (d) is a single pulse of the multi-element sequence GA(4)_BK(3) code modulation; (e) is a single pulse of the multi-element sequence GB(4)_BK(3) code modulation; and (f) is a single pulse of the multi-element sequence GAB(4)_BK(3) code modulation. The final pulse compression decoding signals of the Barker code, Golay code, and multi-element coded sequences are Barker_Cmopress(t), Golay_Cmopress(t), and Echo_Compress_Signal(t), respectively.

[0136] The aforementioned conventional single-pulse excitation, binary-coded excitation, and multi-coded excitation ultrasonic guided waves were boosted and used for rail monitoring. In the experiment, the monitored rail was 12.5 meters long, and the simulated rail fracture location was 7 meters. The artificial rail fracture depth was cut downwards in increments of 10% of the rail height, and the echo signal intensity increased with the increase of the cutting depth.

[0137] To more intuitively illustrate the monitoring of rail fracture performance using different excitation methods, the gain of the signal strength difference between lossy and lossless signals was calculated from both amplitude and energy perspectives. The calculation formulas are as follows:

[0138] Amplitude Angle:

[0139] ΔV gain =20*log 10 (V lossy - V lossless)

[0140] ΔE gain =20*log 10 (E is lossy - E is lossless)

[0141] In the formula, Vlossy and Elossy refer to the amplitude and energy of the signal received by the ultrasonic transducer when the rail has fractured (lossy echo signal); Vlossless and Elossless refer to the amplitude and energy of the signal received by the ultrasonic transducer when the rail has not fractured (lossless signal). ΔV gain and ΔE gain These are the lossy and lossless amplitude difference gain and energy difference gain, respectively.

[0142] The signal calculation results extracted from the monitored rail are as follows: Figure 9 As shown, (a) represents the gain of the amplitude difference variation of traditional single-pulse ultrasonic guided waves, two types of binary-coded excited ultrasonic guided waves, and multi-coded excited ultrasonic guided waves at different rail fracture depths (compared to the amplitude and energy of a healthy rail); (b) represents the gain of the energy difference variation of traditional single-pulse ultrasonic guided waves, two types of binary-coded excited ultrasonic guided waves, and multi-coded excited ultrasonic guided waves at different rail fracture depths (compared to the amplitude and energy of a healthy rail). Figure 9 It can be seen that the performance of ultrasonic guided wave monitoring of rails using multi-element coding sequence excitation is superior to that of traditional single-pulse ultrasonic guided wave excitation, traditional binary coding Barker code and Golay code excitation for rail fracture monitoring. The coding method is flexible and theoretically infinitely long. In addition, it has high monitoring efficiency and high damage echo gain.

Claims

1. A method for monitoring rail fracture based on multi-element encoded ultrasonic guided wave technology, characterized in that, Includes the following steps: (1) Perform convolution operation on the A code sequence and the Barker code sequence in the Golay code to form a new multi-element coding sequence, denoted as GA(m)_BK(n); perform convolution operation on the B code sequence and the Barker code sequence in the Golay code to form a new multi-element coding sequence, denoted as GB(m)_BK(n). (2) Perform convolution operation on the multi-encoding sequence GA(m)_BK(n) and the multi-encoding sequence GB(m)_BK(n) to form a new multi-encoding sequence, denoted as GAB(m)_BK(n); (3) The waveform S(t) of the sine wave modulated by the Hanning window is modulated by the binary coding sequences GA(m), GB(m) and BK(n). The waveforms after modulation by the binary coding sequences are denoted as GA_S(t), GB_S(t) and BK_S(t), respectively. (4) The waveform of the sinusoidal wave modulated by the Hanning window is modulated by the multi-element coding sequence GA(m)_BK(n) and GB(m)_BK(n). The waveforms after modulation by the binary coding sequence are denoted as GA_BK_S(t) and GB_BK_S(t), respectively. (5) The waveform S(t) of the sinusoidal wave modulated by the Hanning window is modulated by the multi-electrode coding sequence GAB(m)_BK(n). The waveform after modulation by the multi-electrode coding sequence is denoted as GAB_BK_S(t), which is the multi-electrode coding excitation ultrasonic guided wave signal used to monitor rail fracture. (6) After the excitation signal is boosted and amplified by the power amplifier, the multi-element coded ultrasonic guided wave signal is excited by the ultrasonic transducer. After passing through the rail medium, when the rail is broken, the multi-element coded ultrasonic guided wave signal will be reflected when it encounters the damaged part. Then, the receiver at the excitation end of the multi-element coded ultrasonic guided wave receives an echo signal, denoted as Echo_signal(t). The echo signal is deconvolved with the GB(m)_BK(n) signal in step (1) to obtain the echo signal Echo_GA_BK_Signal(t). The echo signal is deconvolved with the GA(m)_BK(n) signal in step (1) to obtain the echo signal Echo_GB_BK_Signal(t). (7) When the multi-encoded excitation ultrasonic guided wave signal encounters rail fracture damage, the echo signals Echo_GA_BK_Signal(t) and Echo_GB_BK_Signal(t) are pulse compressed using multi-encoded GA_BK code and GB_BK code respectively to obtain Echo_GA_BK_Compress(t) and Echo_GB_BK_Compress(t) respectively. Then, vector summation is performed to obtain the echo signal Echo_Compress_Signal(t). This signal is the pulse compression echo signal returned when the multi-encoded excitation ultrasonic guided wave encounters rail fracture damage during rail monitoring.

2. The rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology according to claim 1, characterized in that, The calculation formulas for the multivariate sequences GA(m)_BK(n) and GB(m)_BK(n) mentioned in step (1) are as follows: GA(m)_BK(n)=GA(m)*BK(n) GB(m)_BK(n)=GB(m)*BK(n) In the formula, * represents the convolution operator.

3. The rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology according to claim 1, characterized in that, The formula for calculating the multivariate sequence GAB(m)_BK(n) mentioned in step (2) is as follows: GAB(m)_BK(n)=GA(m)_BK(n)*GB(m)_BK(n) In the formula, * represents the convolution operator.

4. The rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology according to claim 1, characterized in that, The calculation formulas for the waveforms GA_S(t), GB_S(t), and BK_S(t) after modulation of the binary coded sequence in step (3) are as follows: GA_S(t)=GA(m)·S(t) GB_S(t)=GB(m)·S(t) BK_S(t) = BK(n)·S(t) The formula for calculating the Hanning window modulated sine wave, S(t), is: In the formula, · represents the product operation symbol, f is the center frequency of the Hanning window modulated sine wave, and N T t is the number of periods of the Hanning window modulated sine wave, t is the duration of the Hanning window modulated sine wave waveform, and A is the amplitude of the Hanning window modulated sine wave.

5. The rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology according to claim 1, characterized in that, The calculation formulas for the waveforms GA_BK_S(t) and GB_BK_S(t) after modulation of the multi-code sequence in step (4) are as follows: GA_BK_S(t)=GA(m)_BK(n)·S(t) GB_BK_S(t)=GB(m)_BK(n)·S(t) In the formula, · represents the product operation symbol, and S(t) represents the waveform of the Hanning window modulated sine wave.

6. The rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology according to claim 1, characterized in that, The calculation formula for the multi-electrode coded excitation ultrasonic guided wave monitoring rail signal GAB_BK_S(t) in step (5) is as follows: GAB_BK_S(t)=GAB(m)_BK(n)·S(t) In the formula, · represents the product symbol, and S(t) represents the waveform of the Hanning window modulated sine wave.

7. The rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology according to claim 1, characterized in that, Step (6) is achieved through the following formula: Echo_signa((t)=h(x)·GAB_BK_S(t) Echo_GA_BK_Signal(t)=h(x)·GAB_BK_S(t)* -1 GB(m)_BK(n) Echo_GB_BK_Signal(t)=h(x)·GAB_BK_S(t)* -1 GA(m)_BK(n) In the formula, · represents the product operator, and * represents the multiplication operator. -1 Let h(x) be the deconvolution operator, and h(x) be the medium transfer function.

8. The rail fracture monitoring method based on multi-element coded ultrasonic guided wave technology according to claim 1, characterized in that, The calculation formulas for Echo_GA_BK_Compress(t), Echo_GB_BK_Compress(t), and Echo_Compress_Signal(t) mentioned in step (7) are as follows: Echo_GA_BK_Cmopress(t)=Echo_GA_BK_signal(t)*GA_BK_S(-t) Echo_GB_BK_CmopRess(t)=Echo_GB_BK_signal(t)*GB_BK_S(-t) Echo_Compress_Signal(t)=Echo_GA_BK_Cmo0Ress(t)+Echo_GB_BK_CmopRess(t) In the formula, * represents the convolution operator, and GA_BK_S(-t) and GB_BK_S(-t) are the mirror waveforms of GA_BK_S(t) and GB_BK_S(t) about the y-axis, respectively.

Images