Ultrasonic guided wave quality detection method, system and application thereof

Abstract

The present invention discloses an ultrasonic guided wave quality detection method, system and application thereof, wherein the quality detection method comprises: collecting ultrasonic signals returned during detection; processing the signals by improved adaptive noise complete set empirical mode decomposition, obtaining the intrinsic mode function after signal decomposition, and then determining the quality of the detection target. The detection method of the present invention is efficient, accurate and suitable for wide application.

Description

Technical Field

[0001] The invention relates to the technical field of using ultrasonic guided waves for quality detection. Background Art

[0002] Anchor rods are a type of reinforcement and support tool widely used in the construction of surface projects such as slopes, deep rock and soil foundation pits, and underground projects such as tunnels and mines. The anchor is the part of the anchor rod that extends into the support or foundation for force transmission. Generally speaking, the longer the anchoring section in the anchor rod, the better the reinforcement and support effect.

[0003] In engineering applications, the quality of anchor rods is directly related to the safety of the project. Once the anchor rod fails, it will pose a huge threat to the stability of the geotechnical structure, which may lead to serious on-site accidents and huge losses of the project. There are usually many reasons for the failure of anchor rods, such as: the anchoring section of the anchor rod is not long enough; the anchoring agent fails to effectively bond with the rod body and the surrounding rock; the anchor rod is broken due to changes in geological conditions or the impact caused by blasting, heavy machinery and earthquake forces; the stress state of the anchor rod changes from axial tensile stress to lateral shear stress due to the lateral sliding of the surrounding rock, causing the anchor rod to be sheared; due to the harsh working environment, such as water, humid air, rainwater and other liquids infiltrating into the anchor rod, the anchor rod corrodes and causes stress concentration, which in turn causes the anchor rod to break at the stress concentration point, etc. In addition to the quality defects of the anchor rod itself and the damage caused by strong external forces, most of these reasons are caused by small damage during long-term work, which requires timely and effective non-destructive testing of the anchor rod to identify, prevent and eliminate dangers.

[0004] In the prior art, a variety of non-destructive testing methods for such reinforcement and / or support tools have been developed, such as the acoustic stress wave method, the optical fiber method, the electromagnetic wave method, and the ultrasonic guided wave method. Among them, the acoustic stress wave method mostly uses a small hammer to knock and vibrate to generate stress waves, and its detection results are easily deviated due to different knocking forces; the electromagnetic wave method utilizes the conductive properties of metal steel bars, and achieves the detection purpose through the phenomenon that the conductor will have a sudden change in potential at the impedance change interface, but most electromagnetic wave methods, such as the potential difference method, the electric standing wave method, the self-induction method, etc., can only detect the length of the anchor rod, and the detection ability of the density of the anchor rod is very weak; the optical fiber method is mostly used for the detection of anchoring force, which has high requirements for welding processing, needs to be pre-buried, is not easy to operate, and the sensors and dynamometers used for detection are easily affected by the environment. Compared with other methods, the ultrasonic guided wave method is more sensitive to the anchoring quality and the instrument operation is simple, but its signal processing difficulty is higher than other methods. How to quickly and accurately obtain the quality information of the object being detected in the ultrasonic guided wave signal is a problem that has always been expected to be solved in this field.

[0005] In signal processing, existing technologies mostly rely on Fourier transform or fast Fourier transform, and some technical means use wavelet transform. Among them, Fourier transform is mainly based on linear and stable signals, and it is easy to cause signal distortion for nonlinear and non-stationary reflection signals; although fast Fourier transform and wavelet transform have improved the shortcomings of Fourier transform to a certain extent, they have not been able to fundamentally get rid of the limitations of Fourier analysis methods, and wavelet transform is prone to energy leakage due to the finite length of the wavelet basis, making the quantitative distribution of the energy frequency time of the signal very difficult. Summary of the invention

[0006] The purpose of the present invention is to propose an ultrasonic guided wave quality detection method which is completely independent of Fourier transform and is particularly suitable for nonlinear and non-stationary signal processing, and has a strong extraction capability for weak reflections and multiple reflections generated by the detection object during the detection process.

[0007] The present invention also aims to propose the application of the above detection method.

[0008] The present invention first discloses the following technical solution:

[0009] An ultrasonic guided wave quality detection system comprises: an ultrasonic guided wave transmitter for emitting low-frequency ultrasonic guided waves, an excitation sensor connected to the ultrasonic guided wave transmitter via a conductive medium and fixed to the end of a target to be detected, a receiving sensor fixed to the same end as the excitation sensor and arranged in parallel, an ultrasonic guided wave receiver connected to the conductive medium of the receiving sensor, a data acquisition device connected to the output end of the ultrasonic guided wave receiver, and a data processing device connected to the data acquisition device via a data transmission channel.

[0010] In some preferred embodiments, the low-frequency ultrasonic guided wave is a longitudinal ultrasonic guided wave of 20kHz-100kHz.

[0011] In some preferred embodiments, the ultrasonic guided wave transmitter and the ultrasonic guided wave receiver are selected from pulse generators and receivers having ultrasonic guided wave transmitting and receiving functions.

[0012] In some preferred embodiments, the conductive medium is a coaxial cable.

[0013] In some preferred embodiments, the transmission channel is a USB interface.

[0014] In some preferred embodiments, the data acquisition device is a dual-channel data acquisition card.

[0015] In some preferred embodiments, the data processing device is a computer.

[0016] In some preferred embodiments, the sensor uses a piezoelectric sensor.

[0017] The present invention further discloses an ultrasonic guided wave quality detection method, which comprises:

[0018] S1: Collect low-frequency ultrasonic guided wave signals after being transmitted to the detected target and returned;

[0019] S2: performing an improved adaptive noise complete ensemble empirical mode decomposition (ICEEMDAN) on the returned ultrasonic guided wave signal to obtain an intrinsic mode function (IMF);

[0020] S3: Obtaining the quality of the detected target according to the intrinsic mode function (IMF).

[0021] In some preferred embodiments, the ultrasonic wave emitted into the detected target in step S1 is a longitudinal ultrasonic guided wave of 20-100 kHz.

[0022] In some preferred embodiments, step S2 comprises:

[0023] S21: adding multiple groups of Gaussian white noise to the returned ultrasonic guided wave signal to generate multiple groups of corresponding new signals;

[0024] S22: performing empirical mode decomposition on the signal sequence after adding Gaussian white noise to obtain multiple envelope mean values, and performing summing and averaging on the multiple envelope mean values ​​to obtain a current order residual;

[0025] S23: subtracting the current order residual from the returned ultrasonic guided wave signal to obtain the current order intrinsic mode function, adding multiple new groups of Gaussian white noise to the current order residual, performing empirical mode decomposition again to obtain multiple new envelope mean values, summing and averaging the multiple new envelope mean values ​​to obtain a new residual, and subtracting the new residual from the current order residual to obtain a new intrinsic mode function;

[0026] S24: Repeat step S23 until the extreme values ​​of the current order residual do not exceed two, and the decomposition is completed.

[0027] In some specific embodiments, step S2 may include:

[0028] S21: Add I groups of Gaussian white noise ω after empirical mode decomposition (EMD) and multiply the signal-to-noise ratio to the sequence s of the returned ultrasonic guided wave signal. i (t), get I group of new signal sequences s i ,as follows:

[0029] s i =s+β0E1(ω (i) )

[0030] Among them, ω (i) represents Gaussian white noise with mean and unit variance of 0 and 1 respectively, and i=1, 2, ..., I; where i represents the number of the added white noise, I represents the total number of the added white noise, represents the white noise coefficient, which is used to adjust the required signal-to-noise ratio between the added noise and the residual after adding the noise. ε0 represents the amplitude of the added noise, which is usually set to 0.2. std(·) is the standard deviation operator. E1 represents the operator of the first intrinsic mode function (IMF) obtained by EMD decomposition.

[0031] S22: for the new signal sequence s i Perform I EMD decomposition to obtain the envelope mean that meets the IMF judgment condition, add and average the multiple envelope mean values ​​to obtain the first-order residual r1 of the returned ultrasonic guided wave signal, and subtract the current order residual from the sequence s of the returned ultrasonic guided wave signal to obtain the first-order intrinsic mode function IMF1, as follows:

[0032] r1= <M(s i )>

[0033] IMF1=s-r1

[0034] Where M(·) represents the operator of the local mean of the envelope that meets the IMF judgment condition; <·> represents the average value calculation;

[0035] S23: Continue to add I groups of Gaussian white noise to the first-order residual to construct a new signal to be decomposed r1+β1E2(ω (i) ), and calculate the local mean by EMD decomposition to obtain the second-order residual r2 and the intrinsic mode function IMF2, as follows:

[0036] r2= <M(r1+β1E2(ω (i) ))>

[0037] IMF2 = r1-r2,

[0038] Wherein, β1 represents the white noise coefficient set when white noise is added for the second time, and E2 represents the second intrinsic mode function obtained by performing EMD decomposition on the added white noise.

[0039] S24: Similarly, obtain the k-th order residual r k and intrinsic mode function IMF k , k = 3, ..., K, K represents the total number of IMFs obtained, as follows:

[0040] r k = <M(r k-1 +β k-1 Ek (ω(i)))>

[0041] IMF k =r k-1 -r k

[0042] Among them, β k =ε0std(r k ), represents the white noise coefficient set when adding white noise for the k-1th time, E k (·) is the operator of the kth intrinsic mode function (IMF) obtained by EMD decomposition;

[0043] S26: Repeat step S24, when the residual r k When it is a monotonic function, stop the calculation and record the obtained intrinsic mode function components.

[0044] Wherein, the empirical mode decomposition includes:

[0045] a. Obtain the envelopes of the local maximum and local minimum points of the current signal sequence by cubic interpolation, which are the upper envelope and the lower envelope respectively;

[0046] b. Add the amplitudes of the upper envelope and the lower envelope and average them to obtain the mean of the two envelopes;

[0047] c. Subtracting the mean from the original signal sequence to obtain a new data sequence;

[0048] d. Determine whether the new data sequence is an intrinsic mode function according to the discrimination conditions. If so, the mean is the target mean. If not, repeat steps ac, perform k screenings, and judge again until the discrimination conditions are met.

[0049] Wherein, the discrimination condition is:

[0050] A data sequence h that satisfies the following conditions at the same time 1k (t) is the intrinsic mode function:

[0051] a. In this sequence, the number of extreme points and the number of zero-crossing points are equal or differ by at most 1;

[0052] b. At any point in the sequence, the average value of the two envelope lines formed by the local maximum point and the local minimum point is 0.

[0053] In some preferred embodiments, condition b is obtained by limiting the standard deviation I SD ,

[0054] in, t represents the time point of the signal sequence, and T represents the total length of the signal sequence.

[0055] When I SD When it is between 0.2 and 0.3, it can be considered that condition b is met.

[0056] In some preferred embodiments, step S3 comprises:

[0057] S31: obtaining the propagation velocity of ultrasonic guided waves in the steel bars;

[0058] S32: obtaining the propagation velocity of the ultrasonic guided wave in the anchor compaction section;

[0059] S33: Determine the mass of the detected target according to the propagation wave velocity and the peak point of the intrinsic mode function.

[0060] In some specific implementations, the step S33 includes:

[0061] Find the peak points of the intrinsic mode function and correspond them to different end faces and / or internal abnormalities of the detected target;

[0062] According to the time of different mutation points and the propagation wave speed, the length between different end faces of the detected target and / or the location where the internal anomaly occurs are obtained;

[0063] The present invention further discloses the application of the above detection system and / or detection method on long strip tools.

[0064] In some embodiments, the tool is a rock bolt.

[0065] The anchor rod may contain an anchoring section.

[0066] The present invention has the following beneficial effects:

[0067] The present invention adopts a signal processing process independent of Fourier transform, which is particularly suitable for nonlinear and non-stationary signal processing. It has a strong extraction capability for weak reflections and multiple reflections generated by the detection target during the detection process, and can quickly and accurately obtain the quality information of the target.

[0068] The present invention can quickly and accurately detect the quality of a target containing different components, such as an anchor rod containing an anchoring section.

[0069] The invention can quickly and accurately detect various quality parameters of the target, such as the length of the anchor rod, the length of the anchoring section therein, the density of the anchor rod, the location of internal defects, the abnormal degree of the defects, etc., and belongs to the field of non-destructive testing.

[0070] The present invention detects the target through ultrasonic guided waves, and the signal processing is efficient and accurate. Compared with the existing quality detection methods, the detection effect is better. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a schematic diagram of the assembly of the detection system of the present invention;

[0072] Figure 2 is a reflection waveform diagram of the end of a certain anchor rod in Example 1;

[0073] Figure 3 It is a graph of all IMF components after the anchor rod end reflection signal in Example 1 is decomposed by ICEEMDAN;

[0074] Figure 4 It is the second-order IMF component diagram of the defective anchor rod end reflection signal after ICEEMDAN decomposition in Example 1. DETAILED DESCRIPTION

[0075] The present invention is described in detail below in conjunction with the embodiments and drawings, but it should be understood that the embodiments and drawings are only used to exemplify the present invention and do not constitute any limitation on the protection scope of the present invention. All reasonable changes and combinations within the scope of the inventive concept of the present invention fall within the protection scope of the present invention.

[0076] The anchoring quality is tested through the following process:

[0077] S1: connecting an ultrasonic guided wave transmitter to excite low-frequency ultrasonic guided waves, and collecting return signals of the low-frequency ultrasonic guided waves from the anchor rod to be detected through an ultrasonic guided wave receiver.

[0078] The system used is as shown in the attached Figure 1As shown, it includes an ultrasonic guided wave transmitter and receiver such as a pulse generator and receiver 1 that can transmit and receive low-frequency ultrasonic guided waves to the anchor sample 6, an acquisition device such as a dual-channel data acquisition card 2 that acquires and stores the output data of the pulse generator and receiver 1, a data processing device such as a computer 3 that processes the acquired data, an excitation sensor 4 that is arranged between the transmitting end of the pulse generator and receiver 1 and the transmitting path of the anchor sample 6 and fixed to the end of the anchor sample 6, and a receiving sensor 5 that is arranged between the receiving end of the pulse generator and receiver 1 and the receiving path of the anchor sample 6 and fixed to the end of the anchor sample 6; wherein the excitation sensor and / or the receiving sensor are preferably piezoelectric sensors. The ultrasonic guided wave transmitter and receiver can be connected to the anchor rod through a coaxial cable with good shielding effect. The low-frequency ultrasonic guided waves excited by the transmitter and receiver can be propagated through an excitation sensor fixed to the end of the anchor rod through an adhesive. After the sensor propagates the low-frequency ultrasonic guided waves into the anchoring of the anchor rod, the receiving sensor fixed to the end of the anchor rod receives the transmission wave generated by the ultrasonic guided wave at the defect and the anchoring interface. The reflected wave signal is then transmitted to the ultrasonic guided wave receiver through the coaxial cable of the receiving sensor. The ultrasonic guided wave receiver outputs the reflected wave signal to a signal data acquisition device, such as a dual-channel data acquisition card. The acquisition device stores the signal data and inputs it into a data processing device, such as a computer, through a transmission channel such as a USB interface. The anchor rod related parameters are obtained according to the computer processing results to accurately detect various quality parameters of the target, such as the length of the anchor rod, the length of the anchoring section, the density of the anchor rod, and the position of internal defects such as the debonding defect x shown in the figure, the abnormality of the defect, etc.

[0079] The low-frequency ultrasonic guided wave selects a longitudinal ultrasonic guided wave of 20kHz-100kHz, and the generation process thereof can be performed using the following process:

[0080] A 20kHz 5-cycle sinusoidal function wave is generated by Hanning window modulation, and the wave packet composed of the sinusoidal function wave is excited by the ultrasonic transmitter as an excitation signal. The excitation signal is transmitted to the anchor bolt through a sensor with a center frequency of 30kHz, thereby generating a low-frequency longitudinal ultrasonic guided wave with a frequency of 20kHz-100kHz.

[0081] S2: The return signal is processed by improved adaptive noise complete ensemble empirical mode decomposition (ICEEMDAN).

[0082] It can further select the following process:

[0083] S21: Add I groups of Gaussian white noise ω after empirical mode decomposition (EMD) and multiply the signal-to-noise ratio to the sequence s of the return signal. i ( t ), and obtain a new set of signals s i .

[0084] s i =s+β0E1(ω (i) )

[0085] Among them, ω (i) represents Gaussian white noise with mean and unit variance of 0 and 1 respectively, and i = 1, 2, ..., I; represents the white noise coefficient, which is used to adjust the required signal-to-noise ratio between the added noise and the residual after adding the noise. ε0 is usually set to 0.2, and std(·) is the standard deviation operator. Operator E k (·) is the kth intrinsic mode function (IMF) obtained by EMD decomposition.

[0086] S22: for the said s i Perform I EMD decomposition to obtain the envelope mean that meets the IMF judgment condition, add and average the multiple envelope mean values ​​to obtain the first-order residual r1 of the return signal, subtract the current order residual from the sequence s of the return signal to obtain the first-order intrinsic mode function IMF1, as follows:

[0087] r1=(M(s i )>

[0088] IMF1=s-r1

[0089] Wherein, M(·) represents the operator of the local mean of the envelope that meets the IMF judgment condition; <·> represents the average value calculation.

[0090] S23: Continue to add I groups of Gaussian white noise to the first-order residual to construct a new signal to be decomposed r1+β1E2(ω (i) ), and calculate the local mean by EMD decomposition to obtain the second-order residual and IMF:

[0091] r2= <M(r1+β1E2(ω (i) ))>

[0092] IMF2=r1-r2

[0093] S24: Similarly, for k=3, ..., K, the k-th order residual can be obtained:

[0094] r k = <M(r k-1 +β k-1 E k (ω (i) ))>

[0095] In the formula, β k =ε0std(r k ).

[0096] S25: Calculate the k-th order IMF:

[0097] IMF k =r k-1 -r k

[0098] S26: Repeat steps S24 and S25 when the residual r k When it is a monotonic function, stop the calculation and record the obtained IMF components.

[0099] Wherein, the empirical mode decomposition includes:

[0100] a. Obtain the envelopes of the local maximum and local minimum points of the current signal sequence by cubic interpolation, which are the upper envelope and the lower envelope respectively;

[0101] b. Add the amplitudes of the upper envelope and the lower envelope and average them to obtain the mean of the two envelopes;

[0102] c. Subtracting the mean from the original signal sequence to obtain a new data sequence;

[0103] d. Determine whether the new data sequence is an intrinsic mode function according to the discrimination conditions. If so, the mean is the target mean. If not, repeat steps ac and judge again until the discrimination conditions are met.

[0104] Wherein, the discrimination condition is:

[0105] A data sequence h that satisfies the following conditions at the same time 1k (t) is the IMF:

[0106] c. In this sequence, the number of extreme points and the number of zero-crossing points are equal or differ by at most 1;

[0107] d. At any point in the sequence, the average value of the two envelope lines formed by the local maximum point and the local minimum point is 0.

[0108] Among them, condition b is difficult to fully meet in the actual processing process, and can be limited by limiting the standard deviation I SD ,

[0109] in,

[0110] When I SD When it is between 0.2 and 0.3, it can be considered that condition b is met.

[0111] S3: Based on the processed signal and the wave velocity of the guided wave in the steel bar and the dense anchor section, the quality parameters of the anchor are obtained, such as the anchor length, anchor section length, anchor density and internal defect location.

[0112] The specific process can be as follows:

[0113] S31: Obtain the propagation velocity C of ultrasonic guided waves in the steel bar f .

[0114] Specifically, you can use:

[0115] S310: Perform ultrasonic guided wave detection on a free anchor rod of known length L (i.e., an anchor rod without an anchoring section) according to the process of step S1, and identify the excitation wave signal and the reflected wave packet signal at the bottom of the anchor rod on the time-amplitude spectrum generated by its return signal.

[0116] S311: According to the peak method, the time corresponding to the peak of the excitation wave on the time-amplitude spectrum and the peak of the reflected wave packet at the bottom of the anchor is taken as the excitation time and reflection time respectively, and the difference between the two is the two-way travel time Δt p , from which the one-way propagation time of the guided wave on the free anchor rod and

[0117] S312: The general anchor rod free section wave velocity C is obtained by the ratio of the free anchor rod length L to the one-way propagation time of the guided wave on the free anchor rod. r .

[0118] S32: Obtain the propagation velocity C of the ultrasonic guided wave in the anchor compaction section b .

[0119] Specifically, you can use:

[0120] S320: Perform ultrasonic guided wave detection on the full-length anchored dense anchor rod of known length L according to the process of step S1, and identify the excitation wave signal and the reflected wave packet signal at the bottom of the anchor rod on the time-amplitude spectrum generated by its return signal.

[0121] S321: According to the peak method, the time corresponding to the peak of the excitation wave on the time-amplitude spectrum and the peak of the reflected wave packet at the bottom of the anchor is taken as the excitation time and reflection time respectively, and the difference between the two is the two-way travel time Δt b , from which the one-way propagation time of the waveguide anchor on the dense anchor rod can be obtained

[0122] S322: The general wave velocity C in the anchor bolt dense section is obtained by the ratio of the anchor bolt length L to the one-way propagation time of the guided wave on the anchor bolt densely anchored. b .

[0123] S33: Determine the quality of the anchor bolt according to the peak value of the intrinsic mode function obtained in step S2.

[0124] Specific examples include:

[0125] Find the peak point of the inherent mode function, analyze the correspondence between the peak point and the signal mutation at the bottom of the anchor rod, the signal mutation at the anchor end face and the internal abnormal signal mutation (i.e., the signal mutation caused by internal defects), determine their respective corresponding reflection times t, and multiply the wave velocity of the guided wave in the steel bar and in the anchor dense section by the propagation time of different sections, respectively, to obtain the length of the anchor rod, the length of the anchor section and the position of the internal defect, and further obtain the ratio of the anchor section length to the total length of the anchor rod, i.e., the density of the anchor rod.

[0126] Specific examples include:

[0127] For densely anchored anchor rods, the length of the anchor rod can be calculated as follows:

[0128]

[0129] Where L is the length of the anchor rod, L1 is the length of the free section, L2 is the length of the anchor body, and C f is the wave velocity of ultrasonic guided waves in the steel bar, C b is the wave velocity of ultrasonic guided wave in the anchor compact section, t b is the reflection time of the interface on the anchor body, t r is the reflection time at the bottom of the anchor, t g is the excitation time of the excitation wave.

[0130] For anchor bolts with anchor defects, the defects can be regarded as steel bars, and the length of the anchor bolt is calculated as follows:

[0131]

[0132] Where L3 is the length of the defect, L 21 and L 22 Indicates the length of the densely grouted anchorage section, t dd is the reflection time of the interface under the anchor defect, t ud is the reflection time of the interface on the anchor defect.

[0133] Example 1

[0134] Through the above implementation, a simulation experiment is carried out:

[0135] The detection target is an anchor rod with a total length of 3m and a diameter of 2cm. The length of the anchor body is 2.7m.

[0136] The return signal generated according to the above implementation is as follows: Figure 2 As shown in the figure, two signal features can be observed. The first is the excitation signal, and the second is the reflection signal at the bottom of the anchor rod, with the corresponding peak time of 2.6ms. There is no other obvious reflection signal between the two signals, indicating that there are no defects in the anchor body and the anchoring quality is good.

[0137] After ICEEMDAN decomposition, the 10th-order IMF is obtained, such as Figure 3 shown.

[0138] The first-order IMF represents the waveform of the excitation signal. The characteristic reflection signal of the interface on the anchor body can be observed from the second-order IMF. Point A represents the peak value of the reflection signal, and the corresponding time is 0.29 ms.

[0139] In addition, the wave velocities of ultrasonic guided waves in the steel bar and anchor dense section were 5172m / s and 2285m / s, respectively.

[0140] According to the peak time and guided wave velocity, the anchor length can be calculated to be 3.03m, of which the anchoring section length is 2.66m and the anchor density is 87.8%.

[0141] The relative error of the anchor length test is 1%, the relative error of the anchor section test is 1.5%, and the relative error of the anchor density test is 2.5%.

[0142] The above embodiments are only preferred implementations of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An ultrasonic quality detection method, characterized in that: include: S1: Collect low-frequency ultrasonic guided wave signals after being transmitted to the detected target and returned; S2: performing improved adaptive noise complete set empirical mode decomposition on the returned ultrasonic signal to obtain an intrinsic mode function; S3: Obtaining the quality of the detected target according to the intrinsic mode function; The ultrasonic wave emitted to the detected target is a longitudinal ultrasonic guided wave of 20-100kHz; Wherein, step S2 comprises: S21: adding multiple groups of Gaussian white noise to the returned ultrasonic guided wave signal to generate multiple groups of corresponding new signals; S22: performing empirical mode decomposition on the multiple groups of new signals to obtain multiple envelope mean values, and performing summing and averaging on the multiple envelope mean values ​​to obtain a current order residual; S23: subtracting the current order residual from the returned ultrasonic guided wave signal to obtain the current order intrinsic mode function, adding multiple new groups of Gaussian white noise to the current order residual, performing empirical mode decomposition again to obtain multiple new envelope mean values, summing and averaging the multiple new envelope mean values ​​to obtain a new residual, and subtracting the new residual from the current order residual to obtain a new intrinsic mode function; S24: repeat step S23 until the extreme values ​​of the current order residual do not exceed two, and the decomposition is completed; The new signal in S21 is obtained by the following calculation model: s i =s+P0E1(ω (i) ),i=1,2,...,I Among them, s i represents the new signal, s represents the returned ultrasonic guided wave signal, ω (i) represents Gaussian white noise with mean and unit variance of 0 and 1 respectively, i represents the number of white noise added, I represents the total number of white noise added, β0 represents the white noise coefficient, which is used to adjust the required signal-to-noise ratio between the added noise and the residual after adding noise, ε0 represents the amplitude of the added noise, which is set to 0.2, and std(·) is the standard deviation operator; E1 represents the operator of the first intrinsic mode function obtained by EMD decomposition; And / or, the k-th order residual r obtained in S22-S23 k and the kth intrinsic mode function IMF k The following calculation model is used to obtain: r k = <M(r k-1 +b k-1 E k (oh (i) ))>,k=3,...,K, IMF k =r k-1 -r k b k =e k std( · r k ), Where M(·) represents the envelope local mean operator that satisfies the intrinsic mode function judgment condition, <·> represents the average value, and E k (·) is the operator of the kth intrinsic mode function obtained by EMD decomposition, β k =ε0std(r k ), which indicates the white noise coefficient set when adding white noise for the k-1th time.

2. The quality inspection method according to claim 1, characterized in that: The empirical mode decomposition includes: The envelopes of the local maximum and minimum points of the current signal sequence are obtained by cubic interpolation, which are the upper envelope and the lower envelope respectively; The amplitudes of the upper envelope and the lower envelope are added and averaged to obtain the mean of the two envelopes; Subtracting the mean value from the original signal sequence to obtain a new data sequence; According to the discrimination conditions, determine whether the new data sequence is an intrinsic mode function. If so, the mean is the target mean. If not, repeat steps ac and make judgments again until the discrimination conditions are met. The judgment condition is: the data sequence h that satisfies the following conditions at the same time 1k (t) is the intrinsic mode function: a. In this sequence, the number of extreme points and the number of zero-crossing points are equal or differ by at most 1; b. At any point in the sequence, the average value of the two envelope lines formed by the local maximum point and the local minimum point is 0.

3. The quality inspection method according to claim 2, characterized in that: The condition b is determined as follows: if the limit standard deviation I SD is 0.2~0.3, which can be considered to meet condition b. in: t represents the time point of the signal sequence, and T represents the total length of the signal sequence.

4. The quality inspection method according to claim 3, characterized in that: The step S3 comprises: S31: obtaining the propagation velocity of ultrasonic guided waves in the steel bars; S32: Obtain the propagation velocity of ultrasonic guided waves in the anchor compaction section S33: determining the mass of the detected target according to the propagation wave velocity and the peak point of the intrinsic mode function; The step S33 comprises: Find the peak points of the intrinsic mode function and correspond them to different end faces and / or internal abnormalities of the detected target; According to the time of different mutation points and the propagation wave speed, the length between different end faces of the detected target and / or the location where the internal anomaly occurs are obtained.

5. An ultrasonic guided wave quality detection system using the ultrasonic quality detection method according to any one of claims 1 to 4, comprising: An ultrasonic guided wave transmitter for transmitting low-frequency ultrasonic guided waves, an excitation sensor connected to the ultrasonic guided wave transmitter via a conductive medium and fixed to the end of a target to be detected, a receiving sensor fixed to the same end as the excitation sensor and arranged in parallel, an ultrasonic guided wave receiver connected to the receiving sensor via a conductive medium, a data acquisition device connected to the output end of the ultrasonic guided wave receiver, and a data processing device connected to the data acquisition device via a data transmission channel.

6. The quality inspection system according to claim 5, characterized in that: The conducting medium is a coaxial cable, and / or the transmission channel is a USB interface, and / or the data acquisition device is a dual-channel data acquisition card, and / or the data processing device is a computer, and / or the ultrasonic guided wave transmitter and the ultrasonic guided wave receiver are selected from pulse generating and receiving instruments with ultrasonic guided wave transmitting and receiving functions, and / or the sensor is a piezoelectric sensor.

7. The quality inspection system according to claim 5, characterized in that: The low-frequency ultrasonic guided waves are longitudinal ultrasonic guided waves of 20kHz-100kHz.

8. Application of the detection method according to any one of claims 1 to 4 or the detection system according to any one of claims 5 to 7 in anchor rod detection.

Images