WO2022143102A1 - Object defect detection method and system based on active electric field - Google Patents

Abstract

An object defect detection method and system based on an active electric field. The method comprises the following steps: S1, placing, in a detection electric field, an object to be subjected to detection in liquid; S2, moving electrical detection signals on a surface of said object according to a preset route, and synchronously collecting an electric field signal; S3, acquiring position information of a defect of said object by means of a joint time-frequency spectrogram; S4, calculating a corner frequency of said object according to the joint time-frequency spectrogram and by using a corner frequency recognition algorithm; and S5, among pre-constructed correlations between crack widths of said object and corner frequencies of said object, finding the crack width of said object according to the corner frequency of said object. Crack information of said object is acquired by means of the joint time-frequency spectrogram and the corner frequency, and whether said object has defects, such as corrosion, coating peeling, incompletion and debris coverage, can be further determined.

Description

A kind of object defect detection method and system based on active electric field technical field

The invention relates to the technical field of metal defect detection, in particular to an object defect detection method and system based on an active electric field.

Background technique

Defect detection of underwater metal objects has a wide range of applications and plays an important role in ship surface coating detection, submersible surface structure detection, underwater pipeline detection, salvage and archaeology.

With the deepening of marine operations and marine development and exploration, the application environment of underwater detection is becoming more and more complex. The current underwater non-destructive testing technology, according to the detection principle, can be divided into: electromagnetic sensing detection technology, detection technology based on ultrasonic sensing and detection technology based on optical sensing.

The detection technology based on optical sensing is mainly underwater visual inspection. Visual inspection can find some small surface cracks and discontinuities, can evaluate the overall underwater structure, and can use video or photography to record relevant information damage.

Magnetic field-based detection technologies include magnetic particle inspection technology, alternating current magnetic field measurement (ACFM) technology, and eddy current detection technology.

The principle of magnetic particle inspection technology is that the magnetized component will form a leakage magnetic field to attract ferromagnetic substances. If the component has defects, the magnetic field formed by the defective part will attract the magnetic powder, which can directly display the size, position, shape and other information of the defect.

AC magnetic field measurement technology is an electromagnetic field non-destructive testing technology that can accurately measure cracks on metal surfaces, and is widely used in the detection and evaluation of structural defects. Its principle is based on the principle of electromagnetic induction, which loads the excitation coil with a low-frequency excitation signal. When the probe is close to the surface of the object to be measured, a uniform electric field will be induced on the surface of the object to be measured. When the induced electric field encounters a defect, the electric field will Bypassing from both ends and bottom of the defect, resulting in the distortion of the magnetic field around the defect. Finally, the defect is quantitatively analyzed by extracting the magnetic field signal in the space, and the signal is input into the computer for analysis, and the changes of the magnetic field Bx and Bz components are obtained.

The principle of eddy current detection technology is that when the coil carrying alternating current is close to the metal material, eddy current will be generated inside the metal material. If the material is defective, the eddy current path in the material will be changed due to the defect, and the impedance inside the coil will be Changes will also occur, and the existence of defects can be judged by detecting changes in the internal impedance of the coil.

The detection technology based on ultrasonic sensing is mainly ultrasonic flaw detection technology, which is a method of using some physical properties of ultrasonic waves in the structure to find defects inside the structure. Ultrasonic testing is not only suitable for metals, but also for non-metallic materials. The principle of ultrasonic flaw detection technology is that ultrasonic waves can penetrate into the material through the ultrasonic transmitter, and then analyze by receiving the signals reflected and refracted by the components themselves and defects. Thereby obtaining characteristic information about defects or materials.

In the existing underwater inspection technology, there are various defects. For example, underwater visual inspection is the most common and basic method, which is easy to operate and has a wide range of applications. However, it can only be used to detect surface crack defects. At the same time, visual inspection also requires underwater inspection personnel to master various underwater inspection techniques, diving techniques and adaptability to combat various accidents, time and labor costs generated underwater. high.

The principle of underwater magnetic particle inspection technology is the same as that on land. The key difficulty is to clean the metal surface or weld to be inspected underwater. Magnetic particle inspection technology is suitable for parts with surface defects that are difficult to see with the naked eye, but the material to be inspected must be ferromagnetic material, suitable for martensitic stainless steel material, not for austenitic stainless steel material, and the tested part needs to be magnetized before flaw detection , Demagnetization is required after flaw detection.

Eddy current method belongs to surface flaw detection method. This method can realize quantitative detection of crack-type defect depth, but cannot give intuitive defect information, and is not suitable for detecting objects with complex shapes.

Underwater ultrasonic has many and complex influencing factors. Ultrasonic testing requires the probe to be closely combined with the surface of the structure. But because the surface of underwater structures can be uneven, the reflected and transmitted ultrasonic waves are more complicated. Coupled with the complex acoustic signals in the ocean, the ultrasonic signal is susceptible to interference, the location of the defect is determined by the time it takes to receive the transmitted ultrasonic waves, and the size of the defect is based on the height of the ultrasonic echo or the defect echo Therefore, the more complex the ultrasonic signal, the lower the reliability and the more difficult the detection.

As the closest prior art to the present invention, the patent "A Method and Device for Underwater Metal Shape Detection Based on Active Electric Field Principle" (Publication No. CN 109188534) discloses an underwater metal shape detection method and device based on the active electric field principle. device. The invention provides a metal shape detection method and a device based on the active electric field detection technology. The underwater active electric field is used to detect metal objects in all directions from different angles, and the collected electric field information is processed to obtain the turning frequencies of the objects in different directions. , and then process to determine the shape of the metal object. The method can overcome the influence of many complex factors such as darkness and turbidity in the underwater environment, is simple in operation, has a wide range of applications, and has a good effect on the shape detection of metal objects in a liquid environment.

SUMMARY OF THE INVENTION

The purpose of the present invention is to propose a new technical means of underwater metal defect detection based on the above-mentioned shape detection of metal objects in a liquid environment based on the turning frequency, which can overcome the existing problems of the existing detection means, and proposes An object defect detection method and system based on an active electric field.

In order to achieve the above-mentioned purpose of the invention, the present invention provides the following technical solutions:

An object defect detection method based on an active electric field, comprising the following steps:

S1, place the measured object in the liquid in the detection electric field, and the detection electric field is composed of the detection electric signal;

S2, moving the detected electrical signal on the surface of the object to be measured according to a preset route, and synchronously collecting the electrical field signal;

S3, obtain the position information of the defect of the tested object through the joint time spectrogram, and the joint time spectrogram is obtained by short-time Fourier transform of the electric field signal;

S4, according to the joint time-spectrogram, the turning frequency identification algorithm is used to calculate the turning frequency of the measured object;

S5 , in the pre-built corresponding relationship between the crack width of the tested object and the turning frequency, find out the crack width of the tested object according to the turning frequency of the tested object.

As a preferred solution of the present invention, step S3 specifically includes the following steps:

S31, perform short-time Fourier transform on the electric field signal to obtain the joint time-spectrogram of the measured object;

S32, find the end points of the convex or concave energy amplitude in the frequency spectrum diagram of the joint time, and obtain the time corresponding to the end points according to the end points,

S33, according to the starting point position, the end position, the moving speed and the time corresponding to the end point of the detected electrical signal moving on the surface of the tested object, determine the position information of the defect of the tested object.

As a preferred solution of the present invention, step S4 specifically includes the following steps:

S41, converting the joint time spectrogram into an energy spectral density matrix;

S42, eliminating noise in the energy spectral density matrix to obtain a filtered energy spectral density matrix;

S43, calculate the distortion value of each frequency component in the filtered energy spectral density matrix;

S44 , performing polynomial fitting on the frequency-distortion value to obtain a turning frequency fitting curve, and the frequency corresponding to the point where the distortion value is 0 in the turning frequency fitting curve is the turning frequency of the measured object.

As a preferred solution of the present invention, in step S41, the step of converting the joint time spectrogram into an energy spectral density matrix is: converting the three-dimensional data of the joint time spectrogram into an energy spectral density matrix, and the energy spectral density matrix is a two-dimensional matrix, In the two-dimensional matrix, the abscissa represents the time gradient, the ordinate represents the frequency, and the coordinate perpendicular to the vertical axis of the plane where the abscissa and the ordinate represent the energy amplitude of the current frequency at the current time.

As a preferred solution of the present invention, step S42 specifically includes the following steps:

S421, calculate the total average value avg of the energy amplitude values in the energy spectral density matrix and the average value avg _i of the energy amplitude values in each row, where i is the row number;

S422, if avg _i <avg, the energy amplitudes in the i-th row are all set to 0.

As a preferred solution of the present invention, the calculation formula of the distortion value in step S43 is:

h _i =(max _i -avg _i )-(avg _i -min _i )

where hi is the distortion value in row _{i, max i is} the maximum value of energy amplitude in row i, min _i is the minimum value of energy amplitude in row _i , and avgi is the amplitude value of energy in row i Value mean, i is the row number.

Based on the same concept, the present invention also proposes an object defect detection system based on an active electric field, including a transmitter electrode, a receiver electrode, a detection pan-tilt and a signal processor,

The transmitting electrode emits a detection electrical signal to form a detection electric field, and the transmitting electrode and the receiving electrode are fixed on the detection platform.

The detection pan/tilt is used to drive the transmitting electrode and the receiving electrode to move on the surface of the measured object according to a preset route, and the receiving electrode synchronously collects the electric field signal during the moving process;

The signal processor receives the electric field signal, uses any of the above methods to detect the underwater measured object, and obtains the defect information of the measured object.

As a preferred solution of the present invention, it is applied to one of the following scenarios: for oil pollution information detection on underwater metals; for crack information detection on underwater metals; for rust information detection on underwater metals; Underwater metal for crack information detection.

As a preferred solution of the present invention, the material of the metal includes iron, aluminum, copper or stainless steel.

Compared with the prior art, the beneficial effects of the present invention:

1. The detection device of the present invention is simple, and does not require professional diving personnel to perform underwater operations, and only needs to operate the underwater detection device to launch into the water. The present invention is applicable to both metallic and non-metallic materials, and is not limited to ferromagnetic materials. If the object to be tested is defective, the method and system of the present invention can visually observe the change, unlike the eddy current method, which cannot give intuitive defect information. There are many and complex influencing factors of underwater ultrasonic waves. Ultrasonic detection requires the probe to be closely combined with the surface of the structure. The present invention does not require the probe to contact the surface of the structure, and unlike ultrasonic waves, it is less affected by marine factors.

2. This method is a completely new method based on the detection of the corner frequency of the active electric field. Based on the active electric field detection technology, the position information of the defect of the tested object can be obtained through the combined time-spectrogram. When the tested object is metal, the turning frequency of the tested object under the active electric field can also be obtained, and the measured object can be measured under the active electric field. Compare and analyze the turning frequency of the lower part and the turning frequency of the standard part, and judge whether the tested object has defects such as rust, coating peeling, object defect, sundries covering and so on.

3. The present invention has the advantages of convenient operation, low energy consumption, strong portability, wide application range, and can be applied to wide temperature and wide pressure occasions. It has important value in underwater exploration, especially in the field of deep-sea exploration. The present invention only needs to perform a single active electric field scan on the tested metal to obtain the turning frequency of the tested object, and analyzes the defect information of the tested object relative to the standard part accordingly, and has the advantages of defect detection for underwater metal objects. good effect.

4. The method has various forms, and the detection signal can be a square wave, a sine wave, or other signals. The materials to be detected are also varied, which can be any common and uncommon metals such as iron, aluminum, copper, stainless steel, etc. The control device of the method is one of a controller, a processor, a single-chip computer or a PC with signal and data processing capabilities; the control device is connected with the detection device in a wired or wireless manner. The detection information of this method includes, but is not limited to, cracks, oil stains, and rust, etc., and is widely used.

Description of drawings:

1 is a flowchart of an object defect detection method based on an active electric field in Embodiment 1 of the present invention;

2 is a joint time-spectrogram obtained by processing an electric field signal through short-time Fourier transform in Embodiment 1 of the present invention;

Fig. 3 is the front panel schematic diagram of adopting Labview to manipulate the movement of the experimental device in the embodiment of the present invention 1;

4 is a fitting curve obtained by fitting the frequency-distortion value in Embodiment 1 of the present invention;

Fig. 5 is the time-frequency joint spectrogram of brass column in the embodiment of the present invention 1;

Fig. 6 is the fitting curve diagram of the turning frequency of the brass column in the embodiment of the present invention 1;

7 is a time-frequency joint spectrogram of a 25mm brass column in Example 1 of the present invention;

Fig. 8 is the time-frequency joint spectrogram of 35mm brass column in the embodiment of the present invention 1;

9 is a specific test environment diagram of a metal defect based on an active electric field corner frequency in Embodiment 2 of the present invention;

10 is a schematic diagram of detecting the movement direction of the pan/tilt in Embodiment 2 of the present invention;

FIG. 11 is a graph showing the oil coverage defect rate-turnover frequency curve in Example 3 of the present invention;

12 is a graph of crack width-turnover frequency curve ratio in Example 3 of the present invention;

Fig. 13 is the combined time spectrum diagram obtained by the crack experiment of the PVC plastic pipe in Example 4 of the present invention

FIG. 14 is a combined time spectrum diagram obtained by the oil pollution coverage experiment of PVC plastic pipes in Example 4 of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with test examples and specific embodiments. However, it should not be construed that the scope of the above-mentioned subject matter of the present invention is limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

Example 1

An object defect detection method based on active electric field, the flow chart is shown in Figure 1, including the following steps:

S1 , constructing a detection electric field, and placing the underwater measured object in the detected electric field, where the detection electric field is composed of a detection electric signal.

The main application scenario of the detection method of the present invention is in liquids, and the liquids referred to in the present invention are conductive electrolyte liquids, such as liquids with ions such as fresh water and seawater. This embodiment mainly takes a liquid such as water as an example to specifically describe the method of the present invention.

The detection electric signal is emitted by a pair of emitting electrode dipoles, one of the emitting electrode dipoles is used for emitting the detection electric signal, and the other is grounded. Therefore, a detection electric field is established between the pair of emitting electrode dipoles. The metal to be tested is placed in water and placed in the probe electric field. The detection signal can be sine, square wave and other signals. The spectrum of the square wave signal has frequency components on odd multiples of the fundamental wave, and the turning frequency can be determined. The sine wave can be used to determine the turning frequency more accurately. A pair of receiving electrodes are arranged adjacent to the dipoles of the transmitting electrodes for synchronously collecting electric field signals and acquiring detection information.

As a preferred solution, the detection signal is a multi-frequency signal, and the detection signal of multiple frequencies is emitted at the same time, then through one detection, multiple sets of data for calculating the turning frequency can be obtained. During the calculation process, multiple sets of data can be directly obtained. The time-frequency distribution spectrogram corresponding to each frequency is used to calculate the corner frequency, instead of using a single frequency signal, multiple tests, and collecting data sets to calculate the corner frequency.

S2, move the detection signal along the surface of the metal to be tested, and collect the electric field signal synchronously. When the metal object is in the underwater environment, due to the existence of the induced polarization effect, it will show high impedance in the low frequency region and low impedance in the high frequency region, and when it is near the corner frequency, it will show Impedance characteristics similar to the surrounding environment. Specifically, when there is an electric field around the object, if the object exhibits low impedance, the electric field lines will appear more densely near the conductor; if the object exhibits high impedance, the electric field lines will appear sparse. When the detection signal is near the turning frequency, the electric field distortion information is the smallest because the measured object exhibits the same characteristic as the resistivity of the surrounding liquid environment.

S3: Process the electric field signal by using the short-time Fourier transform to obtain a joint time-spectrogram of the measured object, and obtain position information of the defect of the measured object through the joint time-spectrogram.

When using the active electric field to detect the object to be measured, the electric field signal is processed through the short-time Fourier transform, and the obtained joint time-spectrogram is shown in Figure 2. As can be seen from the figure, each frequency signal reflects the sag in the amplitude caused by the defect of the measured object on the spectrogram.

The acquisition of the combined time spectrogram is obtained by three softwares, namely Labview software, Labview signal express software, and NI DIAdem software.

The main function of Labview is to manipulate the motion problem of the experimental device, the front panel is shown in Figure 3, and the data is entered.

The Labview signal express software is mainly used to obtain the collected data during the operation of the experimental detection device, and save the data as a file in tdms format.

The main function of NI DIAdem software is to analyze the saved data files in tdms format, and display the files in the form of combined time spectrograms through the program. Select the SCRIPT option in the software, press the run button, select the edited program, select the file to be analyzed, and then select the configuration information, and then the combined time spectrogram can be displayed.

Through the observation of the combined time-spectrogram, the concave and convex changes of the collected electric field signals can be observed more intuitively and accurately, so as to obtain the position information of the defects of the tested object.

Obtaining the position information of the defect of the object under test includes the following steps:

S31, using short-time Fourier transform to process the electric field signal to obtain a joint time-spectrogram of the measured object;

S32, find the end point of the energy amplitude protrusion or depression in the spectrogram during the joint time, and find the corresponding moment of the end point according to the end point;

S33: Calculate the position information of the defect of the tested object according to the starting point position, the ending point position, the moving speed and the time corresponding to the end point of the detected electrical signal moving on the surface of the tested object.

For example, if the measured object is a cylinder, the detected electrical signal moves linearly along the axis direction on the side surface of the cylindrical object, and correspondingly, the electric field signal is collected. Therefore, the starting position, end position and moving speed of the linear motion can be known. After finding the time corresponding to the end point, multiply the time by the moving speed to obtain the distance between the end point of the energy amplitude protrusion or depression corresponding to the starting point position (or end position), and then the position of the protrusion or depression on the measured object can be found. In addition, after knowing the positions of the two end points of the protrusion or depression, the length of the protrusion or depression in the axial direction can be known, and the size information of the crack in the axial direction can be obtained.

Before generating the joint time spectrogram, an appropriate window function needs to be selected. Through pre-experimental research, the processing results of different window functions are compared, and this method finally selects the Hanning window as the window function for signal processing. Within the scope of application of this method, the Hanning window can take into account both the frequency resolution and the time resolution. The Hanning window decomposes the original signal (non-stationary) into a set of approximately stationary short-term signals, and then uses Fourier transform to analyze and process each segment of the short-term signal separately. The time-frequency joint spectrogram can be obtained, from which the change of the original signal spectrum over time can be observed.

S4 , according to the joint time-spectrogram, a turning frequency identification algorithm is used to calculate the turning frequency of the measured object.

Since the detection signal is a multi-frequency signal, the amplitude-frequency characteristic curve can be fitted according to the degree of distortion of the different spectral components caused by the excitation polarization effect of the measured object, so as to avoid errors caused by visual observation and obtain more accurate corner frequency. Specifically include the following steps:

S41 , converting the joint time spectrogram into an energy spectral density matrix.

S42, remove noise in the energy spectral density matrix to obtain a filtered energy spectral density matrix.

S43: Calculate the distortion value of each frequency component in the filtered energy spectral density matrix.

S44 , performing polynomial fitting on the frequency-distortion value to obtain a turning frequency fitting curve, where the frequency value corresponding to the point with a distortion value of 0 in the turning frequency fitting curve is the turning frequency of the measured object.

In step S41, the step of converting the joint time spectrogram into an energy spectral density matrix is: converting the three-dimensional data of the joint time spectrogram into an energy spectral density matrix, where the x-axis of the three-dimensional data is the frequency, the unit is Hz, and the y-axis is is time, the unit is s, the z-axis is energy, the unit is dB; the energy spectral density matrix is a two-dimensional matrix, the horizontal direction in the two-dimensional matrix represents the time gradient, the vertical direction represents the frequency, and the numbers in the matrix represent the current frequency The energy magnitude (in dB) at the current time.

Step S42 specifically includes the following steps:

S421, calculate the total average value avg of the energy spectral density matrix and the average value avg _i of each row, where i is the row number;

S422, if avg _i <avg, the amplitude values in the i-th row are all set to 0.

The calculation formula of the distortion value in step S43 is:

h _i =(max _i -avg _i )-(avg _i -min _i )

where hi is the distortion value in row _{i, max i is} the maximum amplitude value in row i, min _i is the minimum amplitude value in row _i , and avgi is the average value of amplitude in row i , i is the row number.

Steps S42 and S43 are illustrated with the following specific data as examples.

If the obtained data of the energy spectral density matrix p are shown in Table 1. The numbers in the matrix represent the energy amplitude (unit is dB) of the current frequency at the current time.

Table 1 Energy spectral density matrix p

57.5	57.5	59.2	57.4	57.5	57.4
56	56	57.1	56.1	55.8	55.9
52.3	52.3	53.2	52	52.4	52.2
48.1	48.2	49	47.5	48.2	48.3
44.1	44.1	44.8	43.2	44.2	44.1
42.2	42.2	42.2	41.5	42.2	42.3
41.1	41.2	41.8	39.5	41.2	41.1
40.2	40.1	40.7	38.1	40.2	40.2
39.2	39.2	39.5	36.9	39.2	39.2
38.6	38.6	39.1	36	38.6	38.6
38.3	38.3	38.6	35.3	38.3	38.3

38	38	38.3	34.5	38	38
37.6	37.5	38	34.2	37.4	37.4
37	37.1	37.8	33.3	37.1	37
36.5	36.5	37	32.8	36.4	36.4
34.3	34.3	35	30.2	34.2	34.3
30.7	30.7	31	26.4	30.5	30.5
28.2	28.2	29.2	23.2	28.3	28
27.6	27.6	28	22.8	27.8	27.8
26.8	26.8	27.6	twenty one	26.5	26.5

It is calculated that the overall average value of the matrix in Table 1 is around 39.6, and the average value of each row can also be calculated. Then all those less than 39.6 are set to 0, so Table 2 is obtained, that is, the energy spectral density matrix after denoising . Then calculate the distortion value, calculate the mean value avgi of each row in the energy spectral density matrix after denoising, and then calculate the minimum value min _i and the maximum value max _i of each row in the denoised energy spectral density matrix, and the distortion value h _i =(max _i -avg _i )-(avg _i -min _i ), thus the distortion value of each row is obtained.

Table 2 Energy spectral density matrix after denoising

57.5	57.5	59.2	57.4	57.5	57.4
56	56	57.1	56.1	55.8	55.9
52.3	52.3	53.2	52	52.4	52.2
48.1	48.2	49	47.5	48.2	48.3
44.1	44.1	44.8	43.2	44.2	44.1
42.2	42.2	42.2	41.5	42.2	42.3
41.1	41.2	41.8	39.5	41.2	41.1
40.2	40.1	40.7	38.1	40.2	40.2
0	0	0	0	0	0
0	0	0	0	0	0
0	0	0	0	0	0
0	0	0	0	0	0

In step S44, in the energy spectral density matrix after denoising, the vertical direction represents the frequency, and at the same time the distortion value of each row is obtained in the previous step, then the frequency value and the distortion value can be fitted. The polynomial fitting of the frequency-distortion value is implemented in MATLAB. Enter cftool in matlab and call up the cftool toolbox in matlab. Save the frequency value and distortion value of the frequency-distortion value curve as arrays and put them in matlab, and then select the two sets of data of frequency value and distortion value in the cftool toolbox as the horizontal axis and vertical axis respectively. Using Exponential interpolation approximation, the fitting curve can be obtained. The fitting curve is shown in Figure 4, and the turning frequency in the figure is around 200Hz.

S5 , in the pre-built corresponding relationship between the crack width of the tested object and the turning frequency, find out the crack width of the tested object according to the turning frequency of the tested object.

For example, in the previous experiment, the experimental conditions were water temperature 25 degrees Celsius, conductivity 280 μS/cm, a brass column with a length of 50 mm and a diameter of 50 mm was used for the experiment, and the signal generator emitted a square wave with a duty cycle of 50%. , the frequency is 20Hz, and the amplitude is 1V.

Use the test program to test the brass columns with crack widths of 0mm, 5mm, and 15mm, respectively, and obtain their corresponding time-frequency joint spectrum and data. ) is the time-frequency joint spectrogram of 0mm, (b) is the time-frequency joint spectrogram of 5mm, and (c) is the time-frequency joint spectrogram of 15mm. Taking the time-frequency joint spectrogram of 0 mm as an example, the energy spectral density matrix obtained by reading the picture is shown in Table 3.

Table 3 The energy spectral density matrix corresponding to the time-frequency joint spectrogram of 0mm

57.5	57.5	59.2	57.4	57.5	57.4
56	56	57.1	56.1	55.8	55.9
52.3	52.3	53.2	52	52.4	52.2
48.1	48.2	49	47.5	48.2	48.3
44.1	44.1	44.8	43.2	44.2	44.1
42.2	42.2	42.2	41.5	42.2	42.3
41.1	41.2	41.8	39.5	41.2	41.1
40.2	40.1	40.7	38.1	40.2	40.2

Using matlab, the fitting curves were obtained: brass columns with crack widths of 0mm, 5mm, and 15mm, and turning frequencies of 210Hz, 420Hz, and 500Hz, respectively.

After obtaining the turning frequency, use the fitting toolbox cftool of matlab to obtain the turning frequency fitting curve. The turning frequency fitting curve is shown in Figure 6. The corresponding relationship between the crack width of the tested object and the turning frequency can be obtained from the turning frequency fitting curve.

Further, the cracks of 25mm and 35mm are carved in the brass column, and the time-frequency joint spectrum of the 25mm brass column obtained is shown in Figure 7, and the time-frequency joint spectrum of the 35mm brass column obtained is shown in Figure 8. Correspondingly, the measured corner frequencies are 580 Hz and 660 Hz, which are near the fitted curve. Therefore, the crack-turn frequency fitting curve of the brass column is credible, which directly reflects the corresponding relationship between the crack width and the turning frequency. In the actual test, after obtaining the turning frequency, the corresponding crack width can be obtained by looking up the corresponding relationship between the crack width and the turning frequency in the fitting curve.

Example 2

An object defect detection system based on an active electric field includes a transmitter electrode, a receiver electrode, a detection pan-tilt and a signal processor.

The transmitting electrode transmits the detection electrical signal, the receiving electrode obtains the electric field signal, and the transmitting electrode and the receiving electrode are fixed on the detection platform; the detection platform is used to drive the transmitting electrode and the receiving electrode to move along the surface of the measured object, and the receiving electrode is moving During the process, the electric field signal is collected synchronously;

The signal processor receives the electric field signal, and uses the method in Embodiment 1 to detect the underwater metal.

As a specific embodiment, as shown in FIG. 9, a specific test environment is given. The test head includes a detection device and an analysis device. The detection device is equipped with two pairs of detection electrode dipoles, which are arranged in a right-angled trapezoid. Signal to obtain detection information, graphite detection electrodes are used in this experiment. The pan/tilt can be moved. During the movement, the receiving electrodes are used to synchronously collect electric field signals and obtain detection information. The analysis device adopts a computer with signal and data processing capabilities.

The purpose of arranging the transmitting electrode and the receiving electrode in a right-angled trapezoid is to obtain more abundant information. This time, the transmitter electrode and the receiver electrode are placed on the bottom and top bottom of the right-angled trapezoid, and the measured object is on the vertical line of the transmitter electrode. Move the detection pan/tilt in the test area, keep the object under test on the vertical line of the transmitting electrode to ensure the strength of the detection signal, start the stepper motor, and drive the electrode pan/tilt to move in the direction shown in Figure 10. After passing the target object, continue to move forward the same distance and then stop. During the movement of the pan-tilt head, the receiving electrode dipole is used to collect the electric field signal synchronously. Subsequent processing of the electric field signal is the same as that in Embodiment 1, and details are not repeated here.

Example 3

The object defect detection system based on the active electric field of the present invention can not only detect metal defects, but also detect oil pollution coverage and crack defects of metal bodies. The test pan/tilt includes a detection device and an analysis device. In this experimental example, the used detection device includes two pairs of graphite detection electrodes, which are respectively used as transmitting electrode dipoles and two receiving electrode dipoles. The transmitting electrodes are used to transmit excitation signals to establish a detection electric field; the receiving electrodes are used to receive electric field signals. The arrangement of the transmitting electrode and the receiving electrode should be asymmetrical. The analysis device is a computer with signal and data processing capabilities, and the detection device and the analysis device can be connected in a wired or wireless manner.

In this experimental example, a square wave with a peak-to-peak value of 2V and a frequency of 20Hz is used as the detection signal emitted by the transmitting electrode dipole. The material to be tested is a copper solid cylinder with a diameter of 5cm, an iron solid cylinder, an aluminum solid cylinder, a steel solid cylinder and a steel pipe with a length of 10cm and a diameter of 5cm for the experiment. The defect design is divided into two types: 1. Create cracks and defects to conduct underwater object cracks and defects experiments; 2. Use insulating spray paint to spray the metal surface to simulate oil coverage. The experiment was carried out in an underwater environment with a water temperature of 25°C, and the conductivity of the water body was measured to be 320 μS/cm. On the constructed detection pan-tilt, the signal was transmitted through the transmitting electrode dipole, and the detection pan-tilt was moved and received by the detection pan-tilt. The electrode obtains the received signal, processes the received signal through short-time Fourier transform, obtains the joint time-spectrogram, and then obtains the corner frequency through the corner frequency identification algorithm, and compares the standard parts to check the offset. to fit the object defect rate-turnover frequency offset rate curve for subsequent use. The oil coverage defect rate-turnover frequency curve is shown in Figure 11. After calculating the turning frequency of the brass block, brass column, iron column, and steel column in the future, the oil coverage defect rate can be found correspondingly through this curve. The crack width-turning frequency curve rate graph is shown in Figure 11. After calculating the turning frequency of aluminum cylinder, brass column, steel pipe and iron cylinder in the future, the corresponding crack width can be found through the curve. It can also be seen from Fig. 12 that with the increase of cracks or the increase of defect rate, the turning frequency gradually becomes larger.

In practical applications, active frequency-based underwater metal defect detection can be used to detect defects or blockages in underwater pipelines. It can detect the defects of underwater sluices and underwater metal structures in hydropower stations and reservoirs; underwater oil pipelines in petroleum engineering; underwater beam and column defects in bridge engineering, etc.

Example 4

Further, defect detection can also be performed on non-metallic objects. A method for detecting defects of underwater non-metallic objects, comprising the following steps:

A1, construct the detection electric field, and place the object under test in the detection electric field, and the detection electric field is composed of the detection signal;

A2, move the detection signal along the surface of the non-metal to be tested, and collect the electric field signal at the same time;

A3, use the short-time Fourier transform to process the electric field signal to obtain the joint time-spectrogram of the measured object;

A4: Acquire defect information of the non-metal to be tested according to the convex and concave change of the amplitude in the frequency spectrum graph when combined.

A specific example is: a PVC plastic pipe with a length of 10cm, an outer diameter of 50mm, and a thickness of 3.5mm is used, and two cases of defect coverage and cracks are also designed. The rest of the experimental conditions were the same as in Example 3. Figure 13 shows the combined time spectrum of the crack test of PVC plastic pipes, and Figure 14 shows the combined time spectrum of the oil pollution coverage test of PVC plastic pipes. Non-metallic objects have no turning frequency and cannot be judged by the turning frequency. Its defect information, but it can be judged by the combined time spectrogram. It can be seen that when the plastic tube is scanned, the spectrogram appears obviously convex, and when there are defects or cracks, the spectrogram is compared with the convexity. Obvious depression. Find the end point of the depression, and find the time corresponding to the end point according to the end point; according to the starting position, end position, moving speed and the time corresponding to the end point of the electric field signal moving on the surface of the tested object, the defect of the tested object can be calculated. location information.

In practical applications, active frequency-based underwater non-metallic defect detection can be used to detect defects or blockages of underwater non-metallic pipelines. It can detect the defects of some concrete pipes underwater, and can also perform flaw detection on some underwater non-metallic structures.

Claims (7)

1. An object defect detection method based on an active electric field, characterized in that it comprises the following steps:

   S1, placing the object to be measured in the liquid in a detection electric field, and the detection electric field is composed of a detection electrical signal;

   S2, moving the detection electrical signal on the surface of the measured object according to a preset route, and synchronously collecting the electric field signal;

   S3, obtaining the position information of the defect of the tested object through a joint time-spectrogram obtained by performing short-time Fourier transform on the electric field signal;

   S4, according to the joint time spectrogram, adopt the turning frequency identification algorithm to calculate the turning frequency of the measured object;

   S5, in the pre-built corresponding relationship between the crack width of the measured object and the turning frequency, find out the crack width of the measured object according to the turning frequency of the measured object;

   Step S3 specifically includes the following steps:

   S31, performing short-time Fourier transform on the electric field signal to obtain a joint time-spectrogram of the measured object;

   S32, find the end point where the energy amplitude is convex or concave in the spectrogram during the joint time, and obtain the time corresponding to the end point according to the end point,

   S33, determine the position information of the defect of the tested object according to the starting point position, the end position, the speed of movement and the time corresponding to the end point of the detected electrical signal moving on the surface of the tested object;

   Step S4 specifically includes the following steps:

   S41, converting the joint time spectrogram into an energy spectral density matrix;

   S42, eliminate the noise in the energy spectral density matrix, obtain the energy spectral density matrix after filtering;

   S43, calculating the distortion value of each frequency component in the filtered energy spectral density matrix;

   S44 , performing polynomial fitting on the frequency-distortion value to obtain a turning frequency fitting curve, where the frequency value corresponding to the point with a distortion value of 0 in the turning frequency fitting curve is the turning frequency of the measured object.
2. The object defect detection method based on the active electric field according to claim 1, wherein in step S41, the step of converting the joint time spectrogram into an energy spectral density matrix is: converting the joint time spectrogram The three-dimensional data is converted into an energy spectral density matrix, and the energy spectral density matrix is a two-dimensional matrix. In the two-dimensional matrix, the abscissa represents the time gradient, and the ordinate represents the frequency. The coordinates of the vertical axis represent the energy amplitude of the current frequency at the current time.
3. The method for detecting object defects based on an active electric field according to claim 1, wherein step S42 specifically includes the following steps:

   S421, calculate the total average value avg of the energy amplitude in the energy spectral density matrix and the average value avg _i of the energy amplitude in each row, where i is the row number;

   S422, if avg _i <avg, the energy amplitudes in the i-th row are all set to 0.
4. The method for detecting object defects based on an active electric field according to claim 1, wherein the calculation formula of the distortion value in step S43 is:

   h _i =(max _i -avg _i )-(avg _i -min _i )

   where hi is the distortion value in row _{i, max i is} the maximum value of energy amplitude in row i, min _i is the minimum value of energy amplitude in row _i , and avgi is the amplitude value of energy in row i Value mean, i is the row number.
5. An object defect detection system based on an active electric field, characterized in that it includes a transmitter electrode, a receiver electrode, a detection pan-tilt and a signal processor,

   The transmitting electrode emits a detection electrical signal to form a detection electric field, and the transmitting electrode and the receiving electrode are fixed on the detection platform,

   The detection pan/tilt is used to drive the transmitting electrode and the receiving electrode to move on the surface of the measured object according to a preset route, and the receiving electrode synchronously collects electric field signals during the movement;

   The signal processor receives the electric field signal, detects the underwater measured object by using the method according to any one of claims 1-4, and obtains defect information of the measured object.
6. The application of an object defect detection system based on an active electric field according to claim 5, characterized in that it is applied to one of the following scenarios: for detecting oil pollution information on underwater metals; for performing cracks on underwater metals Information detection; used for corrosion information detection of underwater metal; used for crack information detection of underwater metal.
7. The application of an object defect detection system based on an active electric field according to claim 6, wherein the material of the metal comprises iron, aluminum, copper or stainless steel.

Images