A method and system for identifying damage cracks of large equipment in a port

Abstract

The application provides a port large equipment damage crack identification method and system, and relates to the technical field of crack identification. An insulating isolation layer and a conductive primer layer containing ferromagnetic powder are sequentially formed on a metal substrate surface to obtain a pretreated surface; an alternating magnetic field is applied to the conductive primer layer to obtain a magnetized area; a differential array coil probe is used to collect an induced current signal from the magnetized area as a first signal and collect a geomagnetic field disturbance signal as a second signal; the first signal and the second signal are subjected to differential operation to generate a differential signal; an abnormal characteristic value is extracted from the differential signal; and the abnormal characteristic value is compared with a preset threshold value to generate a crack identification result. The application can non-contact identify the metal substrate crack without removing the paint coating, and avoids the misjudgment of the surface paint crack.

Description

Technical Field

[0001] This invention relates to the field of crack identification technology, and in particular to a method and system for identifying damage cracks in large port equipment. Background Technology

[0002] Large port equipment operates under high-load, high-salt-spray, and high-humidity marine environments, making its metal substrate structure prone to fatigue cracks. To prevent corrosion, the equipment surface is typically covered with a paint coating. Existing crack identification methods suffer from significant signal attenuation due to the paint coating lift-off effect; magnetic particle detection requires removing the paint coating, which is complex and can damage the equipment; and visual inspection methods can only identify surface paint cracks, failing to distinguish between paint cracks and metal substrate cracks, leading to frequent misjudgments.

[0003] Therefore, how to identify cracks in the metal substrate of large port equipment without removing the paint coating and avoiding misjudging surface paint cracks as cracks in the metal substrate is an urgent technical problem to be solved. Summary of the Invention

[0004] To address the aforementioned problems in existing technologies, this invention proposes a method for identifying damage cracks in large port equipment, comprising: Step 1: An insulating isolation layer and a conductive primer layer containing ferromagnetic powder are sequentially formed on the surface of a metal substrate to obtain a pretreated surface; Step 2: Based on the pretreated surface, apply an alternating magnetic field to the conductive primer layer to obtain a magnetized region; Step 3: Based on the magnetized region, a differential array coil probe is used to acquire the first and second signals from the magnetized region; Step 3 includes: Step 3.1: Based on the magnetized region, arrange the differential array coil probe above the conductive primer layer; Step 3.2: Based on the excitation coil and receiving coil in the differential array coil probe, the induced current signal is collected from the magnetized region as the first signal; Step 3.3: Based on the magnetic sensor in the differential array coil probe, collect the geomagnetic field disturbance signal from the magnetized region as the second signal. Step 4: Perform a differential operation on the first signal and the second signal to generate a differential signal; Step 5: Extract abnormal feature values ​​from the differential signal; Step 6: Compare the abnormal feature values ​​with the preset threshold to generate crack identification results.

[0005] Compared with the prior art, the beneficial effects of the present invention are as follows: An insulating layer and a conductive primer layer containing ferromagnetic powder are sequentially formed on the surface of a metal substrate to obtain a pretreated surface. The insulating layer provides electrical insulation between the conductive primer layer and the metal substrate, avoiding direct electrical contact interference. The conductive primer layer containing ferromagnetic powder possesses both conductivity and ferromagnetic response capabilities. An alternating magnetic field is applied to the conductive primer layer based on the pretreated surface to obtain a magnetized region. The alternating magnetic field simultaneously magnetizes both the conductive primer layer and the metal substrate, generating different electromagnetic responses in each. A differential array coil probe is used to collect a first signal and a second signal based on the magnetized region. The induced current signal is collected by the excitation coil and the receiving coil as the first signal, which reflects the eddy current distribution induced by the alternating magnetic field in the conductive primer layer and the metal substrate. The geomagnetic field disturbance signal is collected by a magnetic sensor as the second signal. This signal is generated only by the geomagnetic field disturbance caused by the deformation of the conductive primer layer and does not contain information about cracks in the metal substrate. Differential operations are performed on the first signal and the second signal to generate a differential signal. The differential operation eliminates the interference component from the conductive primer layer in the first signal, while retaining the abnormal component caused by cracks in the metal substrate. Abnormal feature values ​​are extracted from the differential signal, and the electromagnetic characteristics of cracks in the metal matrix are quantified into comparable values. The abnormal feature values ​​are compared with a preset threshold to generate crack identification results, thus achieving non-contact determination of cracks in the metal matrix.

[0006] This method utilizes a conductive primer layer containing ferromagnetic powder as the response medium for an alternating magnetic field, without removing the conductive primer layer or damaging the surface protective layer. By simultaneously acquiring signals from two different physical sources and performing differential operations through a differential array coil probe, the method effectively separates the metal substrate crack signal from the surface layer deformation signal, thereby avoiding misjudging surface paint cracks as metal substrate cracks and achieving non-contact identification of cracks in the metal substrate of large port equipment. Attached Figure Description

[0007] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0008] Figure 1 The diagram shown is a flowchart illustrating a method for identifying damage cracks in large port equipment according to an embodiment of the present invention. Figure 2 The diagram shown is a schematic diagram of a damage and crack identification system for large port equipment provided in an embodiment of the present invention. Detailed Implementation

[0009] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0010] The specific embodiments of the present invention will be described below.

[0011] Example 1 like Figure 1 As shown, in a first aspect, the present invention proposes a method for identifying damage cracks in large port equipment, comprising: Step 1: An insulating isolation layer and a conductive primer layer containing ferromagnetic powder are sequentially formed on the surface of a metal substrate to obtain a pretreated surface; Step 2: Based on the pretreated surface, apply an alternating magnetic field to the conductive primer layer to obtain a magnetized region; Step 3: Based on the magnetized region, a differential array coil probe is used to acquire the first and second signals from the magnetized region; Step 3 includes: Step 3.1: Based on the magnetized region, arrange the differential array coil probe above the conductive primer layer; Step 3.2: Based on the excitation coil and receiving coil in the differential array coil probe, the induced current signal is collected from the magnetized region as the first signal; Step 3.3: Based on the magnetic sensor in the differential array coil probe, collect the geomagnetic field disturbance signal from the magnetized region as the second signal. Step 4: Perform a differential operation on the first signal and the second signal to generate a differential signal; Step 5: Extract abnormal feature values ​​from the differential signal; Step 6: Compare the abnormal feature values ​​with the preset threshold to generate crack identification results.

[0012] In step 1, the surface of the metal substrate of the large port equipment is treated. The metal substrate surface may have an original paint coating or rust layer, which needs to be removed to expose the metal substrate itself. Removal methods can include mechanical grinding or chemical cleaning, without damaging the metal substrate. On the cleaned metal substrate surface, an insulating layer and a conductive primer layer containing ferromagnetic powder are formed sequentially. The insulating layer prevents electrical contact between the subsequent conductive primer layer and the metal substrate, avoiding electrochemical corrosion or signal short circuits. The conductive primer layer containing ferromagnetic powder has conductive properties and can respond to externally applied magnetic fields. The formation order cannot be reversed: the insulating layer is formed first, followed by the conductive primer layer. The insulating layer can be made of epoxy resin, polyurethane resin, or other organic coatings with good insulating properties. The coating method includes brushing, rolling, or spraying. After coating, it needs to be dried or cured to form a continuous and dense insulating film. A conductive primer layer is formed on the surface of the insulating layer. The conductive primer layer is composed of a conductive paint substrate and ferromagnetic powder. The conductive paint substrate can be epoxy conductive paint, acrylic conductive paint, etc., and the ferromagnetic powder can be iron oxide powder, nickel powder, or iron powder. The ferromagnetic powder is dispersed in the conductive paint at a certain ratio, stirred evenly, and then sprayed onto the surface of the insulating layer. After drying at room temperature or curing by heating, a pre-treated surface is obtained. The pre-treated surface has the following characteristics: the insulating layer electrically isolates the conductive primer layer from the metal substrate; the conductive primer layer is conductive and can generate induced current under the action of an alternating magnetic field; the ferromagnetic powder in the conductive primer layer gives it magnetic permeability, allowing it to be magnetized and influence the distribution of the Earth's magnetic field.

[0013] In step 2, based on the pretreated surface, an alternating magnetic field is applied to the conductive primer layer to obtain a magnetized region. The alternating magnetic field can be applied using an excitation coil. The excitation coil is arranged parallel above the conductive primer layer, maintaining a certain lift-off distance. A sinusoidal alternating current is passed through the excitation coil, generating an alternating magnetic field around it. The alternating magnetic field penetrates the conductive primer layer and the insulating layer, acting on the metal substrate. Under the excitation of the alternating magnetic field, both the conductive primer layer and the metal substrate are magnetized, forming a magnetized region. The extent of the magnetized region is determined by the size of the excitation coil, the current frequency, and the magnetic field strength. Within the magnetized region, induced eddy currents are generated in both the conductive primer layer and the metal substrate. Simultaneously, the magnetic domains of the ferromagnetic powder periodically align with the direction of the alternating magnetic field, producing a magnetization response.

[0014] In step 3, based on the magnetized region, a differential array coil probe is used to acquire a first signal and a second signal from the magnetized region. The differential array coil probe is a composite probe that integrates an excitation coil, a receiving coil, and a magnetic sensor. Step 3.1: Based on the magnetized region, the differential array coil probe is positioned above the conductive primer layer. The probe maintains a fixed lift-off distance from the surface of the conductive primer layer, which is controlled by the support structure of the probe housing. Step 3.2: Based on the excitation coil and receiving coil in the differential array coil probe, the induced current signal is acquired from the magnetized region as the first signal. The differential array coil probe itself contains an excitation coil. After the probe is positioned, an alternating current is passed through the excitation coil to generate a local alternating magnetic field. This local alternating magnetic field is superimposed on the alternating magnetic field applied in step 2, further inducing eddy currents in the conductive primer layer and the metal substrate. The receiving coil is close to the surface under test to detect the secondary magnetic field generated by the eddy currents and convert it into a voltage signal, i.e., the induced current signal. The amplitude and phase of the signal are affected by cracks in the conductive primer layer and the metal substrate: cracks in the metal substrate disturb the eddy current distribution, thereby changing the signal output by the receiving coil. Step 3.3: Based on the magnetic sensor in the differential array coil probe, the geomagnetic field disturbance signal is collected from the magnetized area as the second signal. The magnetic sensor can be a fluxgate sensor or a giant magnetoresistive sensor. The magnetic sensor only responds to static or low-frequency magnetic fields and does not respond to the high-frequency components of the alternating magnetic field applied in step 2. When the conductive primer layer is deformed, such as having cracks or unevenness on the surface, the distribution of ferromagnetic powder in the conductive primer layer changes, causing disturbances in the distribution of the geomagnetic field in a local area. The magnetic sensor collects the geomagnetic field disturbance signal, which only reflects the deformation of the conductive primer layer and does not contain information about cracks in the metal substrate. In actual detection, the differential array coil probe scans along a preset path within the magnetized area, collecting the first and second signals at each scanning position.

[0015] In step 4, a differential operation is performed on the first signal and the second signal to generate a differential signal. The first signal contains response information from both the conductive primer layer and the metal substrate; the second signal contains only information from the deformation of the conductive primer layer. Subtracting the second signal from the first signal eliminates the contribution of the conductive primer layer deformation to the first signal, while retaining the anomalous component caused by the crack in the metal substrate. The differential operation can be implemented at the analog circuit level or at the digital signal processing level. For example, the first and second signals can be converted from analog to digital and then input into a microprocessor to perform point-by-point subtraction.

[0016] In step 5, anomalous feature values ​​are extracted from the differential signal. The differential signal is a waveform or sequence that varies over time. Anomalous feature values ​​can be quantifiable parameters such as peak amplitude, phase shift, and spectral energy of the differential signal. The specific extraction method is as follows: the differential signal is smoothed and filtered to remove random noise, and then the difference between the maximum positive peak value and the maximum negative peak value of the signal is calculated as the amplitude feature; the zero-crossing time difference of the differential signal relative to the alternating magnetic field reference signal in step 2 is calculated and converted into a phase delay angle as the phase feature; a fast Fourier transform is performed on the differential signal to extract the fundamental frequency component amplitude and the higher harmonic component amplitude as the frequency response feature. The above features are combined into a feature vector as the anomalous feature value.

[0017] In step 6, the abnormal feature value is compared with a preset threshold to generate a crack identification result. The preset threshold is a value obtained by statistically analyzing standard samples without cracks through the same testing process. When the abnormal feature value is greater than or equal to the preset threshold, a crack in the metal matrix is ​​determined to exist at the current detection location; when the abnormal feature value is less than the preset threshold, no crack in the metal matrix is ​​determined to exist. The crack identification result can be displayed as an indicator light alarm, a display screen showing the crack location, or an output logic signal.

[0018] This method separates the crack signal of the metal substrate from the deformation signal of the surface layer through differential operation without removing the conductive primer layer, thus avoiding misjudging surface cracks as metal substrate cracks and achieving non-contact identification of cracks in the metal substrate itself.

[0019] In some implementations, step 1, forming the insulating layer and the conductive primer layer, specifically includes: Step 1.1: Coat the surface of the metal substrate with a thin layer of epoxy resin to obtain an insulating layer; Step 1.2: Spray conductive paint containing nano-sized iron oxide powder onto the surface of the insulating layer to obtain a conductive primer layer; Step 1.3: Curing the conductive primer layer to obtain a pretreated surface.

[0020] In step 1.1, a thin layer of epoxy resin is coated onto the surface of the metal substrate to obtain an insulating layer. The epoxy resin used is bisphenol A type epoxy resin, combined with an amine curing agent. Before coating, the surface of the metal substrate is cleaned to remove oil and oxide scale. The epoxy resin mixture is evenly coated onto the surface of the metal substrate using a brush or spray method. The coating thickness should be sufficient to completely cover the metal substrate without any missed areas, for example, controlled between tens and hundreds of micrometers, but no specific value is limited. After coating, it is placed at room temperature or heated to a set temperature for curing. After curing, an insulating layer is formed. The insulating layer should have a high volume resistivity to effectively block the electrical path between the conductive primer layer and the metal substrate.

[0021] In step 1.2, a conductive paint containing nano-sized iron oxide powder is sprayed onto the surface of the insulating layer to obtain a conductive primer layer. The nano-sized iron oxide powder has a uniform particle size distribution and good ferromagnetism. The conductive paint matrix can be an epoxy conductive paint, which has been premixed with conductive fillers such as conductive carbon black or silver powder. The nano-sized iron oxide powder and the conductive paint matrix are mixed at a volume ratio, and the volume ratio of iron oxide powder to the total volume of the mixed slurry is controlled at a low level to ensure that the conductivity of the conductive paint is not excessively weakened. After mixing, the mixture is stirred evenly with a high-speed stirrer to ensure that the iron oxide powder is evenly dispersed in the slurry. The mixed slurry is loaded into a spray gun, and the air pressure is adjusted to ensure uniform atomization. The mixture is sprayed onto the surface of the insulating layer, and the spray gun movement speed is kept uniform. One or two layers are sprayed to make the conductive primer layer surface smooth and free of drips. After spraying, a conductive primer layer is obtained, which is both conductive and contains dispersed iron oxide powder particles.

[0022] In step 1.3, the conductive primer layer is cured to obtain a pretreated surface. The curing method depends on the type of conductive paint substrate. For epoxy-based conductive paints, curing can be performed by placing it at room temperature for a specified time or by heating it to a certain temperature in an oven. During the curing process, the conductive paint substrate undergoes a cross-linking reaction, forming a solid coating, while the iron oxide powder is fixed inside the coating. After curing, the conductive primer layer is firmly attached to the surface of the insulating layer, and the surface hardness meets the requirements for subsequent probe scanning. The pretreated surface consists of a three-layer structure: a metal substrate, an insulating layer, and a conductive primer layer. This pretreated surface can directly accept subsequent alternating magnetic field application and signal acquisition without any additional processing. During the detection process, paint cracks on the surface of the conductive primer layer cause changes in the distribution of iron oxide powder, thereby generating geomagnetic field disturbance signals; while cracks in the metal substrate change the induced current signal by affecting the eddy current distribution. The two signals are separated through subsequent differential operations.

[0023] This method utilizes a thin layer of epoxy resin to provide reliable insulation, uses nano-sized iron oxide powder to enhance the magnetization response of the conductive primer layer, and ensures the adhesion and mechanical strength of the conductive primer layer through curing treatment, thereby forming a stable pretreated surface.

[0024] In some implementations, applying an alternating magnetic field in step 2 specifically includes: Step 2.1: Based on the pretreated surface, arrange the excitation coil in parallel above the conductive primer layer; Step 2.2: Pass an alternating current through the excitation coil to generate an alternating magnetic field; Step 2.3: Magnetize the conductive primer layer and the metal substrate with an alternating magnetic field to obtain the magnetized region.

[0025] In step 2.1, based on the pretreated surface, the excitation coil is arranged parallel to the conductive primer layer. The excitation coil is a hollow cylindrical coil or a rectangular planar coil, with the coil axis perpendicular to the surface of the conductive primer layer. Parallel arrangement means that the plane of the coil is parallel to the surface of the conductive primer layer, and the distance between them is the lift-off distance. The lift-off distance is fixed by the support structure of the coil frame or a non-magnetic pad, such as installing rollers at the bottom of the excitation coil housing to maintain a constant gap between the coil and the surface of the conductive primer layer. The size of the excitation coil should be selected according to the size of the testing area. For weld areas or stress concentration areas of large port equipment, the coil diameter or side length can cover the width of the area to be tested. The two leads of the excitation coil are connected to an alternating current source.

[0026] In step 2.2, an alternating current is passed through the excitation coil to generate an alternating magnetic field. The alternating current source outputs a sinusoidal current with a frequency that can be set in the low-frequency range, such as tens of hertz to several kilohertz. The specific value is adjusted according to the thickness of the conductive primer layer and the conductivity of the metal substrate. When the alternating current passes through the excitation coil, an alternating magnetic field is generated around the coil. The magnetic field strength is proportional to the current amplitude and the number of coil turns. The direction of the alternating magnetic field changes sinusoidally with time, and the magnetic field lines originate from one end of the coil, pass through the conductive primer layer and the metal substrate, and return to the other end of the coil.

[0027] In step 2.3, the conductive primer layer and the metal substrate are magnetized using an alternating magnetic field to obtain a magnetized region. The alternating magnetic field penetrates the insulating layer and acts on both the conductive primer layer and the metal substrate. The conductive primer layer contains ferromagnetic powder, and under the action of the alternating magnetic field, the magnetic domains of the ferromagnetic powder undergo periodic reorientation, generating magnetization intensity. The metal substrate is generally made of ferromagnetic steel and is also magnetized by the alternating magnetic field, generating alternating magnetic induction intensity within the metal substrate. The magnetized region of the conductive primer layer and the metal substrate is called the magnetized region, which is circular or elliptical, with its center located directly below the excitation coil. The boundary of the magnetized region is determined by the position where the magnetic field intensity decays below a certain threshold. Within the magnetized region, the magnetic states of the conductive primer layer and the metal substrate change periodically with time, generating their respective magnetic field responses. The differential array coil probe in subsequent steps should perform signal acquisition within the magnetized region to ensure a sufficient signal-to-noise ratio.

[0028] This method utilizes an excitation coil to generate an alternating magnetic field, simultaneously magnetizing the conductive primer layer and the metal substrate, thereby producing distinguishable electromagnetic responses between the two and providing a magnetized region for subsequent differential detection.

[0029] In some implementations, the first signal in step 3 is an induced eddy current signal, and the second signal is a geomagnetic field disturbance signal, which is obtained in the following ways: Step 3.1: Based on the magnetized region, the eddy current signal induced in the conductive primer layer and metal substrate by the receiving coil in the differential array coil probe is extracted as the first signal; Step 3.2: Based on the magnetized region, the geomagnetic field disturbance signal caused only by the deformation of the conductive primer layer is extracted by the fluxgate sensor in the differential array coil probe as the second signal.

[0030] In step 3.1, based on the magnetized region, the eddy current signal induced in the conductive primer layer and metal substrate by the alternating magnetic field is extracted by the receiving coil in the differential array coil probe and used as the first signal. The differential array coil probe integrates an excitation coil and a receiving coil. The receiving coil adopts a differential structure, such as two identical sub-coils connected in reverse series, or multiple coils arranged in a planar manner to form a differential pair. When the probe is placed in the magnetized region, the alternating magnetic field generated by the excitation coil in step 2 induces eddy currents in the conductive primer layer and metal substrate. The magnitude and distribution of the eddy currents are affected by the conductivity of the conductive primer layer, the conductivity of the metal substrate, and the presence of cracks in either. Cracks in the metal substrate will change the local conductivity, causing the eddy current path to deflect, thereby changing the secondary magnetic field generated by the eddy currents. The receiving coil detects the change in the secondary magnetic field and converts it into a voltage signal. This voltage signal is the induced eddy current signal, denoted as the first signal. The output signal of the receiving coil is pre-amplified and filtered before being sent to the signal processing unit. The first signal contains information from both the conductive primer layer and the metal substrate, making it impossible to directly distinguish the contributions of either.

[0031] In step 3.2, based on the magnetized region, a fluxgate sensor in the differential array coil probe extracts the geomagnetic field disturbance signal caused solely by the deformation of the conductive primer layer, as the second signal. The fluxgate sensor is a sensor that measures static or low-frequency magnetic fields, possessing high sensitivity and directionality. The fluxgate sensor is installed inside the differential array coil probe, arranged parallel to the receiving coil, but electrically isolated from it. The fluxgate sensor only responds to the geomagnetic field and static or quasi-static magnetic field disturbances caused by ferromagnetic materials, and does not respond to the higher-frequency alternating magnetic field in step 2. When deformation exists on the surface of the conductive primer layer, such as paint cracks, scratches, protrusions, or depressions, the thickness of the conductive primer layer in the deformed area changes, and the distribution density of the ferromagnetic powder changes accordingly. The uneven distribution of the ferromagnetic powder leads to local permeability changes, thereby altering the distribution of the geomagnetic field in that area and generating a geomagnetic field disturbance signal. The fluxgate sensor outputs the amplitude and direction of this disturbance signal. Due to the presence of the insulating layer, deformation or cracks in the metal substrate do not affect the distribution of ferromagnetic powder in the conductive primer layer. Therefore, the geomagnetic field disturbance signal is caused only by the deformation of the conductive primer layer itself and is unrelated to cracks in the metal substrate. The output signal of the fluxgate sensor is amplified and low-pass filtered to obtain the second signal.

[0032] This method extracts eddy current signals induced by alternating magnetic fields through a receiving coil to reflect the overall conductivity of the conductive primer layer and the metal substrate. It also extracts geomagnetic field disturbance signals caused only by the deformation of the conductive primer layer through a fluxgate sensor to isolate surface deformation interference, providing signals from two independent physical sources for subsequent differential operations.

[0033] In some implementations, the difference operation in step 4 specifically includes: Step 4.1: Subtract the first signal from the second signal point by point in the time domain to obtain the initial differential signal; Step 4.2: Perform bandpass filtering on the initial differential signal to remove power frequency interference and noise, and obtain the filtered differential signal; Step 4.3: Perform amplitude normalization on the filtered differential signal to obtain the differential signal.

[0034] In step 4.1, the first signal and the second signal are subtracted point-by-point in the time domain to obtain the initial differential signal. The first and second signals are obtained by a synchronous acquisition system and have the same time base and sampling rate. For each sampling moment, the instantaneous amplitude of the second signal is subtracted from the instantaneous amplitude of the first signal to obtain the difference at that moment. This operation is repeated for all sampling moments to generate a set of difference sequences, i.e., the initial differential signal. Since the second signal only reflects the geomagnetic field disturbance caused by the deformation of the conductive primer layer, while the first signal contains contributions from both the deformation of the conductive primer layer and the cracks in the metal substrate, the contribution of the conductive primer layer deformation is canceled out after the subtraction operation, and the initial differential signal mainly retains the anomalous component caused by the cracks in the metal substrate. The subtraction operation can be implemented in a digital signal processor or field-programmable gate array using subtraction instructions.

[0035] In step 4.2, the initial differential signal is bandpass filtered to remove power frequency interference and noise, resulting in a filtered differential signal. The initial differential signal contains environmental electromagnetic interference, such as power frequency signals and harmonics from the power grid, as well as thermal noise and shot noise from electronic components. The passband range of the bandpass filter is set according to the fundamental frequency of the alternating magnetic field. The lower passband frequency is lower than the fundamental frequency, and the upper passband frequency is higher than the fundamental frequency but lower than its second or third harmonic. For example, if the alternating magnetic field frequency is a specified frequency, the bandpass filter allows signal components near that frequency to pass through while attenuating frequency components below the lower limit and above the upper limit. Power frequency interference is usually at a low frequency and can be suppressed by the low-end cutoff frequency of the bandpass filter. The filter can be implemented using an infinite impulse response filter or a finite impulse response filter, with convolution operations performed in digital signal processing. The output signal after filtering is the filtered differential signal, with a significantly improved signal-to-noise ratio.

[0036] In step 4.3, the filtered differential signal is normalized to obtain the differential signal. The purpose of amplitude normalization is to eliminate the influence of factors such as probe lift-off distance fluctuations and excitation magnetic field strength fluctuations on the signal amplitude. The specific normalization method is as follows: calculate the average amplitude of the filtered differential signal in the crack-free reference region as the reference amplitude; then divide the signal amplitude at each sampling point by the reference amplitude to obtain the dimensionless normalized amplitude. Another method is to divide the filtered differential signal by its own root mean square value to normalize the overall energy of the signal. After amplitude normalization, the amplitude range of the differential signal is adjusted to a consistent standard range. The normalized signal is the final differential signal output in step 4, used for subsequent abnormal feature value extraction. In this differential signal, the abnormal component caused by the metal matrix crack is retained, while the amplitude differences under different detection conditions are eliminated.

[0037] This method eliminates the interference of conductive primer layer deformation on the first signal by subtracting point by point in the time domain, then removes power frequency interference and noise by bandpass filtering, and finally eliminates the influence of lift-off distance fluctuation by amplitude normalization, thus obtaining a stable and comparable differential signal.

[0038] In some implementations, step 5, extracting abnormal feature values, specifically includes: Step 5.1: Calculate the peak amplitude of the differential signal to obtain the amplitude characteristics; Step 5.2: Calculate the phase delay angle of the differential signal relative to the alternating magnetic field to obtain the phase characteristics; Step 5.3: Perform Fourier transform on the differential signal to extract the response intensity of the differential signal at the fundamental frequency of the alternating magnetic field and the response intensity at the second harmonic, and obtain the frequency response characteristics; Step 5.4: Combine the amplitude feature, phase feature, and frequency response feature into a feature vector, which serves as the anomalous feature value.

[0039] In step 5.1, the peak amplitude of the differential signal is calculated to obtain the amplitude characteristic. The differential signal is a waveform sequence that varies with time; in the region where the crack exists, the differential signal exhibits significant fluctuations. The peak amplitude is calculated as follows: within one scanning cycle or one detection window, find the maximum positive and maximum negative values ​​of the differential signal, calculate the sum of their absolute values, or directly take the difference between the maximum positive and minimum negative values. The amplitude characteristic reflects the opening size or depth of the crack in the metal matrix: the larger or deeper the crack opening, the stronger the eddy current disturbance, and the larger the differential signal amplitude. The amplitude characteristic can be expressed as a scalar value.

[0040] In step 5.2, the phase delay angle of the differential signal relative to the alternating magnetic field is calculated to obtain the phase characteristics. The alternating magnetic field is generated by the excitation current in step 2, and the phase of the excitation current is used as the reference phase. The differential signal originates from eddy currents. When eddy currents propagate in a conductive medium, there is a phase lag. The lag angle is related to the conductivity, permeability, and geometric parameters of the crack. The phase delay angle is calculated by extracting the fundamental frequency component of the differential signal, measuring the time difference between the fundamental frequency component and the reference phase, and converting the time difference into an angle difference, which ranges from -180° to 180°. The phase characteristics reflect the electrical properties of the crack: the presence of a crack changes the local impedance, causing a phase shift in the eddy current. Combining the phase characteristics with the amplitude characteristics can distinguish between different types or depths of cracks.

[0041] In step 5.3, a Fourier transform is performed on the differential signal to extract the response intensity at the fundamental frequency and the second harmonic of the alternating magnetic field, thus obtaining the frequency response characteristics. The Fourier transform converts the time-domain differential signal into a frequency-domain spectrum. A Fast Fourier Transform (FFT) algorithm is executed to obtain a series of complex amplitudes of frequency components. The amplitude at the fundamental frequency of the alternating magnetic field is extracted and denoted as the fundamental frequency response intensity; the amplitude at twice the fundamental frequency is extracted and denoted as the second harmonic response intensity. The fundamental frequency response intensity mainly reflects the linear electromagnetic response, while the second harmonic response intensity originates from the nonlinear magnetization characteristics of ferromagnetic materials. Cracks in the metal matrix can alter the nonlinearity of the local magnetic field, thus affecting the intensity of the second harmonic component. The fundamental frequency response intensity and the second harmonic response intensity are treated as two independent frequency response characteristics. If the differential signal contains abundant harmonic components, response intensities at higher harmonics can also be extracted.

[0042] In step 5.4, the amplitude feature, phase feature, and frequency response feature are combined into a feature vector, which serves as the anomalous feature value. The feature vector is a multi-dimensional array; for example, a three-dimensional array may contain the amplitude feature, phase feature, and fundamental frequency response intensity, or a four-dimensional array may contain the amplitude feature, phase feature, fundamental frequency response intensity, and second harmonic response intensity. The combination is arranged in a fixed order, for example, the amplitude feature as the first component, the phase feature as the second component, the fundamental frequency response intensity as the third component, and the second harmonic response intensity as the fourth component. The anomalous feature value is this feature vector. This feature vector integrates the amplitude, phase, and spectral information of the differential signal, and more accurately reflects the presence and severity of cracks in the metal matrix than a single-dimensional feature. The anomalous feature value is subsequently compared with a preset threshold, which is also a feature vector of the same dimension. The comparison can be made by calculating the Euclidean distance or Mahalanobis distance between the two vectors.

[0043] This method extracts peak amplitude, phase delay angle, and fundamental and second harmonic response intensities from differential signals, combining multi-dimensional physical quantities into feature vectors to provide comprehensive feature parameters for the quantitative determination of cracks in metal matrix.

[0044] In some implementations, the preset threshold is obtained in advance through the following steps: Based on a crack-free metal matrix sample, steps 1 to 5 are executed sequentially to obtain multiple abnormal feature values ​​under crack-free conditions, forming a crack-free feature set. Calculate the arithmetic mean and standard deviation of all abnormal feature values ​​in the crack-free feature set to obtain the statistical mean and statistical standard deviation; The preset threshold is obtained based on the statistical mean and statistical standard deviation.

[0045] The preset threshold is obtained based on a known crack-free metal substrate sample. This sample has the same material and surface treatment process as the large port equipment to be tested, i.e., it also undergoes the pretreatment in step 1, where an insulating layer and a conductive primer layer containing ferromagnetic powder are sequentially formed on the surface of the metal substrate to obtain a pretreated surface. The metal substrate of this sample itself does not contain any cracks. For the crack-free metal substrate sample, steps 2 to 5 are performed sequentially. Step 2: An alternating magnetic field is applied to the conductive primer layer to obtain a magnetized region. Step 3: A differential array coil probe is used to collect a first signal and a second signal from the magnetized region. Step 4: The first signal and the second signal are differentially processed to generate a differential signal. Step 5: Abnormal feature values ​​are extracted from the differential signal. The above steps are repeated multiple times to obtain multiple abnormal feature values. The number of repetitions can be determined according to the accuracy requirements, for example, multiple measurements are taken at different locations on the sample, or multiple measurements are taken at the same location, with each measurement independently obtaining an abnormal feature value. These abnormal feature values ​​are collected to form a crack-free feature set. Each element in this set is an anomalous feature value. The specific form of the anomalous feature value is a feature vector, as mentioned above, which includes amplitude features, phase features, and frequency response features.

[0046] Calculate the arithmetic mean of all abnormal feature values ​​in the crack-free feature set. The arithmetic mean is calculated by summing the values ​​for each dimension of each abnormal feature value, then dividing by the total number of abnormal feature values. The statistical mean is a vector with the same dimension as the abnormal feature values, representing the central location of the abnormal feature values ​​in the crack-free state. Simultaneously, calculate the standard deviation of all abnormal feature values ​​in the crack-free feature set. The standard deviation is calculated by squared the difference between each value in each dimension and the statistical mean, summing these values, dividing by the number of abnormal feature values, and then taking the square root. The statistical standard deviation is also a vector with the same dimension, representing the dispersion of the abnormal feature values ​​in the crack-free state.

[0047] A preset threshold is obtained based on the statistical mean and statistical standard deviation. The specific form of the preset threshold can be the statistical mean plus a specified multiple of the statistical standard deviation. For example, the threshold could be the sum of the statistical mean and one standard deviation, or the sum of the statistical mean and two standard deviations. The multiple is determined based on the detection sensitivity and false alarm rate requirements. The preset threshold and the abnormal feature value have the same dimension. For example, if the abnormal feature value is a four-dimensional feature vector, then the preset threshold is also a four-dimensional vector, where each dimension corresponds to the statistical mean's corresponding dimension plus a specified multiple of the standard deviation of that dimension. In actual step 6, the measured abnormal feature value is compared with the preset threshold. If the norm of one or all dimensions of the abnormal feature value exceeds the corresponding threshold, a crack is determined to exist. Establishing the threshold using the statistical mean and standard deviation quantifies the normal fluctuation range in a crack-free state, avoiding misjudgments caused by random signal fluctuations.

[0048] This method statistically analyzes the distribution of abnormal characteristic values ​​in crack-free metal matrix samples, and constructs a threshold using the arithmetic mean and standard deviation, thus providing a statistical basis for subsequent crack determination.

[0049] Some implementations also include a step of locating the crack: Step 7.1: Based on the scanning path of the differential array coil probe, record the coordinates of the probe at each scanning position and the abnormal feature value at that position to generate a position-feature value sequence; Step 7.2: Based on the location-feature value sequence, extract the location coordinates of abnormal feature values ​​that exceed a preset threshold to obtain the crack location coordinate set; Step 7.3: Plot the crack location coordinate set on the two-dimensional mesh map of the preprocessed surface to generate crack identification results containing crack location information.

[0050] Before scanning, a differential array coil probe is used in conjunction with a two-dimensional moving platform or a manual scanning frame. The probe moves above the pre-processed surface along a preset scanning path. The scanning path can be a serpentine line, a grid line, or a straight line along the weld direction. During the probe's movement, a scanning position is recorded at fixed spatial steps. Each scanning position corresponds to a planar coordinate system. For example, a two-dimensional rectangular coordinate system is established with a corner point on the pre-processed surface as the origin, where the horizontal and vertical coordinates represent the projection position of the probe center on the surface, respectively. At each scanning position, following steps 3 to 5, the first and second signals are acquired, differential operations are performed, and the abnormal feature values ​​at that position are extracted. The position coordinates are associated with the abnormal feature values ​​at that position, generating a position-feature value sequence. This sequence can be a table or a linked list, where each row contains the horizontal and vertical coordinates and the corresponding abnormal feature value vector.

[0051] The coordinates of locations where abnormal feature values ​​exceed a preset threshold are extracted, resulting in a crack location coordinate set. For each location in the sequence, its abnormal feature value is compared with a pre-obtained preset threshold. The comparison method can be Euclidean distance or dimension-by-dimensional comparison. If the abnormal feature value is greater than the preset threshold, a metal matrix crack is determined to exist at that location, and the coordinates of that location are recorded in the crack location coordinate set; if the abnormal feature value is not greater than the preset threshold, the location is ignored. After traversing all scanned locations, the crack location coordinate set contains the coordinates of all suspected crack locations.

[0052] The coordinate set of crack locations is plotted on a 2D mesh map of the pre-treated surface, generating a crack identification result containing crack location information. The 2D mesh map is a planar diagram with the pre-treated surface as a reference, with the horizontal and vertical axes aligned with the scanning coordinate system. Points in the entire crack location coordinate set are marked on this mesh map, for example, with red dots or crosses. For densely packed crack locations, interpolation or contour extraction can be used to form continuous crack trajectory lines. The final crack identification result can be a crack distribution image or an inspection report containing a list of crack locations. This result directly displays the actual location of cracks in the metal matrix on the surface of large port equipment, facilitating defect location and repair by maintenance personnel. The crack identification result can be output to a display screen, printer, or storage medium.

[0053] This method utilizes the correspondence between the position coordinates on the scanning path of the differential array coil probe and the abnormal feature values ​​to extract the positions exceeding the preset threshold and plot them on a two-dimensional grid, thereby achieving precise visualization of the spatial location of cracks.

[0054] In some implementations, the conductive primer layer containing ferromagnetic powder in step 1 is prepared by the following method: Based on ferromagnetic powder and conductive paint matrix, nano-sized iron oxide powder and epoxy conductive paint matrix are mixed in a volume ratio to obtain a mixed slurry, wherein the volume of iron oxide powder accounts for less than 1% of the total volume of the mixed slurry. After mixing the slurry evenly, load it into the spray gun to obtain the slurry to be sprayed; The slurry to be sprayed is evenly sprayed onto the surface of the insulating layer to obtain a conductive primer layer.

[0055] In step 1, the conductive primer layer containing ferromagnetic powder is prepared as follows: First, prepare the ferromagnetic powder and the conductive paint matrix. The ferromagnetic powder used is nano-sized iron oxide powder, which has a large specific surface area and good dispersibility. The conductive paint matrix is ​​an epoxy conductive paint matrix, which contains conductive fillers such as carbon black or silver powder and already possesses a certain conductivity. The nano-sized iron oxide powder and the epoxy conductive paint matrix are mixed at a volume ratio, and the volume of iron oxide powder is controlled to be less than 1% of the total volume of the mixed slurry. This volume ratio is set based on the following: an excessively high volume ratio of iron oxide powder will dilute the original conductive fillers in the conductive paint matrix, reducing the conductivity of the conductive primer layer. At the same time, excessive solid powder will reduce the cohesion of the coating, affecting adhesion and flexibility. A volume ratio of less than 1% ensures that the epoxy conductive paint matrix can still form a continuous conductive network, and the uniform dispersion of the nano-sized iron oxide powder is sufficient to provide a ferromagnetic response that can be detected by a magnetic sensor. During mixing, a mechanical stirrer is used for low-speed pre-stirring to avoid generating air bubbles. The mixture is then transferred to a high-speed disperser for further dispersion, breaking up any agglomerates of the ferric oxide powder and distributing them evenly throughout the slurry. The dispersion time is determined by visually eliminating any particle agglomeration. After dispersion, a mixed slurry is obtained.

[0056] After thoroughly mixing the slurry, pour it into the spray gun to obtain the slurry to be sprayed. Mixing can be done using a manual stirring rod or an electric mixer within the spray gun's reservoir, ensuring that the slurry does not settle before spraying. Use a gravity-feed or suction-type spray gun, selecting the nozzle diameter based on the viscosity of the slurry. Pour the slurry into the spray gun's reservoir, close the lid, and the preparation of the slurry to be sprayed is complete.

[0057] The slurry to be coated is evenly sprayed onto the surface of the insulating layer to obtain a conductive primer layer. Before spraying, ensure that the surface of the insulating layer is completely cured and free of contaminants. Adjust the air pressure of the spray gun to form a uniform fan-shaped spray of the atomized slurry. Keep the spray gun perpendicular to the surface of the insulating layer and maintain a constant distance. Use a cross-spraying method, that is, spray one coat in one direction first, and then spray a second coat in the perpendicular direction to ensure uniform coating thickness. Allow the coating to dry for a period of time after each spraying to avoid sagging. After spraying, the conductive primer layer covers the entire surface of the insulating layer, and the nano-sized iron oxide powder is dispersed within the coating. This conductive primer layer possesses the following characteristics: Due to the inherent conductivity of the epoxy conductive paint matrix, the surface resistivity of the conductive primer layer is low, enabling it to conduct induced eddy currents. The nano-sized iron oxide powder dispersed within it imparts ferromagnetism to the conductive primer layer, generating a magnetization response under the influence of an external magnetic field. Furthermore, when deformation such as cracks occurs in the conductive primer layer, the distribution of the iron oxide powder changes, thereby altering the local geomagnetic field distribution and generating geomagnetic field disturbance signals. Simultaneously, the volume proportion of less than 1% ensures that the conductivity of the conductive primer layer is not significantly degraded by the addition of iron oxide powder, allowing the receiving coil to still acquire sufficiently strong induced eddy current signals.

[0058] This method involves mixing nano-sized iron oxide powder with an epoxy conductive paint matrix at a volume ratio, while controlling the volume percentage of iron oxide powder to be less than 1%. This ensures that the conductive primer layer has good conductivity while also possessing ferromagnetic response capabilities, and avoids excessive powder from damaging the conductive network.

[0059] Example 2 like Figure 2 As shown, in a second aspect, the present invention proposes a damage and crack identification system for large port equipment. The system employs a damage and crack identification method for large port equipment proposed in any of the above embodiments, and the system includes: A pretreatment surface forming module is used to perform step 1: sequentially forming an insulating isolation layer and a conductive primer layer containing ferromagnetic powder on the surface of a metal substrate to obtain a pretreatment surface; The magnetization region generation module is used to perform step 2: based on the pretreated surface, an alternating magnetic field is applied to the conductive primer layer to obtain the magnetization region; The differential array coil probe signal acquisition module is used to perform step 3: based on the magnetized region, the differential array coil probe is used to acquire the first signal and the second signal from the magnetized region; Step 3 includes: Step 3.1: Based on the magnetized region, arrange the differential array coil probe above the conductive primer layer; Step 3.2: Based on the excitation coil and receiving coil in the differential array coil probe, the induced current signal is collected from the magnetized region as the first signal; Step 3.3: Based on the magnetic sensor in the differential array coil probe, collect the geomagnetic field disturbance signal from the magnetized region as the second signal. The differential operation module is used to execute step 4: perform differential operation on the first signal and the second signal to generate a differential signal; The abnormal feature value extraction module is used to perform step 5: extracting abnormal feature values ​​from the differential signal; The crack identification result generation module is used to perform step 6: compare the abnormal feature values ​​with the preset threshold to generate crack identification results.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying damage cracks in large port equipment, characterized in that, include: Step 1: An insulating isolation layer and a conductive primer layer containing ferromagnetic powder are sequentially formed on the surface of a metal substrate to obtain a pretreated surface; Step 2: Based on the pretreated surface, apply an alternating magnetic field to the conductive primer layer to obtain a magnetized region; Step 3: Based on the magnetized region, a differential array coil probe is used to acquire the first and second signals from the magnetized region; Step 3 includes: Step 3.1: Based on the magnetized region, arrange the differential array coil probe above the conductive primer layer; Step 3.2: Based on the excitation coil and receiving coil in the differential array coil probe, the induced current signal is collected from the magnetized region as the first signal; Step 3.3: Based on the magnetic sensor in the differential array coil probe, collect the geomagnetic field disturbance signal from the magnetized region as the second signal. Step 4: Perform a differential operation on the first signal and the second signal to generate a differential signal; Step 5: Extract abnormal feature values ​​from the differential signal; Step 6: Compare the abnormal feature values ​​with the preset threshold to generate crack identification results.

2. The method according to claim 1, characterized in that, Step 1, which involves forming an insulating layer and a conductive primer layer, specifically includes: Step 1.1: Coat the surface of the metal substrate with a thin layer of epoxy resin to obtain an insulating layer; Step 1.2: Spray conductive paint containing nano-sized iron oxide powder onto the surface of the insulating layer to obtain a conductive primer layer; Step 1.3: Curing the conductive primer layer to obtain a pretreated surface.

3. The method according to claim 1, characterized in that, The application of the alternating magnetic field in step 2 specifically includes: Step 2.1: Based on the pretreated surface, arrange the excitation coil in parallel above the conductive primer layer; Step 2.2: Pass an alternating current through the excitation coil to generate an alternating magnetic field; Step 2.3: Magnetize the conductive primer layer and the metal substrate with an alternating magnetic field to obtain the magnetized region.

4. The method according to claim 1, characterized in that, In step 3, the first signal is the induced eddy current signal, and the second signal is the geomagnetic field disturbance signal, which are obtained in the following ways: Step 3.1: Based on the magnetized region, the eddy current signal induced in the conductive primer layer and metal substrate by the receiving coil in the differential array coil probe is extracted as the first signal; Step 3.2: Based on the magnetized region, the geomagnetic field disturbance signal caused only by the deformation of the conductive primer layer is extracted by the fluxgate sensor in the differential array coil probe as the second signal.

5. The method according to claim 4, characterized in that, Step 4, the difference operation specifically includes: Step 4.1: Subtract the first signal from the second signal point by point in the time domain to obtain the initial differential signal; Step 4.2: Perform bandpass filtering on the initial differential signal to remove power frequency interference and noise, and obtain the filtered differential signal; Step 4.3: Perform amplitude normalization on the filtered differential signal to obtain the differential signal.

6. The method according to claim 5, characterized in that, Step 5, extracting anomalous feature values, specifically includes: Step 5.1: Calculate the peak amplitude of the differential signal to obtain the amplitude characteristics; Step 5.2: Calculate the phase delay angle of the differential signal relative to the alternating magnetic field to obtain the phase characteristics; Step 5.3: Perform Fourier transform on the differential signal to extract the response intensity of the differential signal at the fundamental frequency of the alternating magnetic field and the response intensity at the second harmonic, and obtain the frequency response characteristics; Step 5.4: Combine the amplitude feature, phase feature, and frequency response feature into a feature vector, which serves as the anomalous feature value.

7. The method according to claim 1, characterized in that, The preset threshold is obtained in advance through the following steps: Based on a crack-free metal matrix sample, steps 1 to 5 are executed sequentially to obtain multiple abnormal feature values ​​under crack-free conditions, forming a crack-free feature set. Calculate the arithmetic mean and standard deviation of all abnormal feature values ​​in the crack-free feature set to obtain the statistical mean and statistical standard deviation; The preset threshold is obtained based on the statistical mean and statistical standard deviation.

8. The method according to claim 1, characterized in that, It also includes the step of locating the crack: Step 7.1: Based on the scanning path of the differential array coil probe, record the coordinates of the probe at each scanning position and the abnormal feature value at that position to generate a position-feature value sequence; Step 7.2: Based on the location-feature value sequence, extract the location coordinates of abnormal feature values ​​that exceed a preset threshold to obtain the crack location coordinate set; Step 7.3: Plot the crack location coordinate set on the two-dimensional mesh map of the preprocessed surface to generate crack identification results containing crack location information.

9. The method according to claim 1, characterized in that, The conductive primer layer containing ferromagnetic powder in step 1 is prepared by the following method: Based on ferromagnetic powder and conductive paint matrix, nano-sized iron oxide powder and epoxy conductive paint matrix are mixed in a volume ratio to obtain a mixed slurry, wherein the volume of iron oxide powder accounts for less than 1% of the total volume of the mixed slurry. After mixing the slurry evenly, load it into the spray gun to obtain the slurry to be sprayed; The slurry to be sprayed is evenly sprayed onto the surface of the insulating layer to obtain a conductive primer layer.

10. A damage and crack identification system for large port equipment, characterized in that, The system employs a method for identifying damage cracks in large port equipment as described in any one of claims 1 to 9, and the system comprises: A pretreatment surface forming module is used to perform step 1: sequentially forming an insulating isolation layer and a conductive primer layer containing ferromagnetic powder on the surface of a metal substrate to obtain a pretreatment surface; The magnetization region generation module is used to perform step 2: based on the pretreated surface, an alternating magnetic field is applied to the conductive primer layer to obtain the magnetization region; The differential array coil probe signal acquisition module is used to perform step 3: based on the magnetized region, the differential array coil probe is used to acquire the first signal and the second signal from the magnetized region; Step 3 includes: Step 3.1: Based on the magnetized region, arrange the differential array coil probe above the conductive primer layer; Step 3.2: Based on the excitation coil and receiving coil in the differential array coil probe, the induced current signal is collected from the magnetized region as the first signal; Step 3.3: Based on the magnetic sensor in the differential array coil probe, collect the geomagnetic field disturbance signal from the magnetized region as the second signal. The differential operation module is used to execute step 4: perform differential operation on the first signal and the second signal to generate a differential signal; The abnormal feature value extraction module is used to perform step 5: extracting abnormal feature values ​​from the differential signal; The crack identification result generation module is used to perform step 6: compare the abnormal feature values ​​with the preset threshold to generate crack identification results.

Images