Non-destructive testing method, system, device and medium for metal flange corrosion based on short wave eddy current

Abstract

The application discloses a short-wave eddy current-based metal flange corrosion nondestructive detection method, system, device and medium, relates to the technical field of electromagnetic nondestructive detection of metal components, and comprises the following steps: generating a wide-frequency continuous adjustable excitation signal to sweep the metal flange, collecting complex impedance to determine the optimal frequency and adaptively switching the impedance matching network, calibrating the probe through the air zero point and the standard test block thickness, continuously exciting the frequency and synchronously collecting the eddy current secondary field signal, sequentially performing analog anti-aliasing filtering, digital band-pass filtering and adaptive coherent noise cancellation based on an independent reference channel, performing time-frequency analysis on the noise-reduced signal to extract signal characteristics, inputting the standard corrosion sample block calibration inversion model to calculate the corrosion depth and volume loss rate, and dividing the safety level and graphically warning according to the multi-level threshold value. The application realizes 100 millimeter penetration and micron-level resolution, realizes grading warning through self-calibration of the probe and quantitative inversion output, and meets single-person live working by cooperating with full insulation protection.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic nondestructive testing technology for metal components, specifically to a method, system, equipment, and medium for nondestructive testing of corrosion of metal flanges based on short-wave eddy currents. Background Technology

[0002] Metal flanges are critical connecting components in various important facilities such as power transmission towers, petrochemical pipelines, and offshore platforms. Detecting their corrosion status is crucial for maintaining structural safety and lifespan. Eddy current testing has attracted much attention from researchers due to its absence of coupling media, sensitivity to surface or near-surface defects, and rapid scanning capabilities. In recent years, eddy current testing technology has been mainly applied in areas such as multi-frequency eddy current, pulsed eddy current, and flexible array probes. Some research results have achieved preliminary identification of coated flanges. However, conventional eddy current testing frequencies are usually in the range of hundreds of kilohertz to low megahertz. Although the skin depth can penetrate relatively thick walls, its resolution for early micro-corrosion of ferromagnetic flanges is relatively poor. The mapping relationship between signal amplitude changes and corrosion depth is greatly affected by the non-uniformity of magnetic permeability, making it difficult to meet the precise quantitative requirements of the power and heavy industry for flange residual wall thickness. In addition, low-frequency eddy current and pulsed eddy current methods have difficulties in effectively suppressing electromagnetic noise in the environment and probe lift-off effects, resulting in limited repeatability and a signal-to-noise ratio that needs further improvement.

[0003] Existing eddy current testing solutions for in-service metal flanges generally fail to address the adaptive adaptation of the excitation frequency and impedance matching network to the electromagnetic characteristics of the tested object. This results in inefficient injection of excitation energy into the workpiece, and a lack of rapid on-site compensation for zero-point drift introduced by parasitic coupling in the probe coil. At the signal processing level, conventional solutions often employ a single filtering link, failing to establish independent reference channels and adaptive cancellation mechanisms for time-varying coherent noise such as power frequency magnetic fields and coupling from nearby conductors, thus limiting the extraction capability of weak corrosion signals. In terms of defect quantification, existing technologies largely rely on single-parameter inversion, failing to incorporate eddy current amplitude attenuation, phase shift, and regional inhomogeneity into a unified corrosion quantification model. Furthermore, they lack a graded early warning criterion that correlates corrosion depth with volume loss rate based on standards and specifications, making it difficult for on-site inspection personnel to directly obtain safety level information that can guide maintenance decisions. Summary of the Invention

[0004] In view of the above-mentioned existing problems, the present invention provides a method, system, equipment and medium for non-destructive testing of corrosion of metal flanges based on short-wave eddy current, in order to solve the problems of low resolution, weak anti-interference ability and poor portability of corrosion detection systems for metal flanges in the prior art.

[0005] To address the aforementioned technical problems, a non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents is proposed, including: By generating a wideband, continuously adjustable alternating excitation signal, the metal flange is frequency-scanned, and complex impedance data at each frequency point is collected to determine the optimal excitation frequency. The impedance matching network is then adaptively switched. The transceiver-integrated single-crystal eddy current probe undergoes air zero-point self-calibration and calibration with a standard eddy current test block of known thickness. The eddy current secondary field signal is simultaneously acquired, and signal filtering and adaptive coherent noise cancellation are performed. Time-frequency analysis is conducted on the noise-reduced signal to extract signal features. The signal features are input into the inversion model to calculate the corrosion depth and volume loss rate. Based on the ratio of the corrosion depth to the design wall thickness, the volume loss rate, and the non-uniformity, the corrosion safety level is classified according to multi-level thresholds, and early warning information is output graphically.

[0006] As a preferred embodiment of the non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents described in this invention, the adaptive switching impedance matching network includes sweeping the frequency with a fixed step size within a preset frequency band to obtain the complex impedance response of the probe coil at each frequency point. The normalized impedance deviation rate at each frequency point is calculated as a corrosion sensitivity index, and the signal-to-noise ratio is evaluated. Candidate frequencies with the highest corrosion sensitivity index and signal-to-noise ratio index exceeding the preset threshold are selected. If there are multiple candidates, the optimal excitation frequency is determined based on the correspondence between skin depth and expected corrosion depth. Based on the optimal excitation frequency, the connection and switching of the reactive element combination in the reconfigurable impedance matching network are controlled, and the network topology parameters are changed so that the energy transfer efficiency parameter between the excitation source and the probe load reaches the preset matching condition.

[0007] As a preferred embodiment of the non-destructive testing method for metal flange corrosion based on short-wave eddy current described in this invention, the step of performing air zero-point self-calibration and known thickness standard eddy current test block benchmark calibration includes placing the probe in a space without the metal being tested, transmitting continuous excitation at the optimal excitation frequency, and acquiring the original zero-point signal generated by the inherent parasitic coupling between coils in the receiving channel. The adaptive compensation process is run to generate a cancellation component with the same amplitude and opposite phase as the current original zero-point signal. This component is then injected into the receiving link and finely adjusted it iteratively until the parasitic coupling residual is lower than the cancellation convergence threshold. The compensation parameters are then latched as zero-point calibration data. The probe is coupled to a standard eddy current calibration block with a nominal remaining thickness. The optimal excitation frequency is maintained, and the eddy current response of the block is obtained. Based on the mapping relationship between the current response characteristics and the known thickness, a thickness measurement benchmark relationship is constructed, and benchmark calibration is completed.

[0008] As a preferred embodiment of the non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents described in this invention, the execution signal filtering and adaptive coherent noise cancellation include: exciting the flange with the optimal excitation frequency and simultaneously sampling the induced voltage on the receiving coil through multiple channels. The induced voltage is sequentially subjected to low-pass anti-aliasing processing and digital bandpass frequency selection processing. An independent noise reference coil, which is coaxially nested with the main detection coil and does not contact the workpiece, is enabled to obtain the environmental electromagnetic interference as a reference signal. The interference component coherent with the reference signal is subtracted from the main signal through an adaptive noise cancellation mechanism to obtain a pure eddy current response. A joint time-frequency transformation is performed on the pure eddy current response. The fundamental amplitude attenuation rate and fundamental phase lag with the same frequency as the excitation are extracted in the transformation domain. The equivalent time delay parameters of the eddy current response are determined by the time-frequency energy distribution and converted into the equivalent residual wall thickness. Statistical dispersion calculation is performed on the measurement results of multiple points within the set detection area to obtain the local non-uniformity.

[0009] As a preferred embodiment of the non-destructive testing method for metal flange corrosion based on short-wave eddy current described in this invention, the calculation of corrosion depth and volume loss rate includes inputting the fundamental amplitude attenuation rate, the wall thickness reduction amount of the equivalent remaining wall thickness relative to the reference wall thickness, and the local non-uniformity into a multi-parameter corrosion quantification model constructed from the calibration data of standard corrosion samples, and outputting the corrosion depth value and the percentage of metal volume loss.

[0010] As a preferred embodiment of the non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents described in this invention, the step of classifying corrosion safety levels according to multi-level thresholds includes comparing the ratio of corrosion depth to design wall thickness, percentage of metal volume loss, and local non-uniformity with three-level judgment thresholds according to preset safety level classification criteria. When all indicators are below the first-level threshold, it is considered a normal state; when any indicator falls between the first-level threshold and the second-level threshold, it is considered a warning state; when any indicator reaches or exceeds the second-level threshold, it is considered a dangerous state. The driver displays a visually distinctive interface and uses dynamic prompts to output tiered warning information corresponding to the judgment status.

[0011] As a preferred embodiment of the non-destructive testing method for metal flange corrosion based on short-wave eddy current described in this invention, the method of outputting early warning information in a graphical manner includes: real-time monitoring of waveform display status parameters; when the peak amplitude exceeds the upper or lower limit preset ratio of the current vertical display range, automatically adjusting the front-end signal gain control parameters and synchronously adjusting the waveform vertical mapping ratio. When the time domain coordinates corresponding to the equivalent remaining wall thickness exceed the current horizontal time base range, the time axis scale range is automatically switched to widen the time axis. When the target analysis feature window deviates from the display center area by more than the set offset criterion, the data reading trigger delay is automatically adjusted to move the feature window to the display center area; Adjustments to signal gain control parameters, time axis scale range, and data read trigger delay only affect the display mapping layer; the original sampled data sequence remains unchanged.

[0012] The beneficial effects of this preferred technical solution are as follows: short-wave eddy current adaptive frequency selection and impedance matching are used to improve the sensitivity of defects; stable reference and clean signal are obtained through dual calibration and multi-level coherent noise reduction; corrosion amount is inverted by a multi-parameter model calibrated by a standard corrosion sample block and three-level early warning is performed; intelligent display, outdoor power supply protection and hardware synchronization enable on-site real-time guidance for operation and maintenance.

[0013] As a preferred embodiment of the short-wave eddy current-based non-destructive testing system for metal flange corrosion described in this invention, it is characterized by comprising a short-wave eddy current excitation and adaptive matching module, a transceiver integrated single-crystal probe and dual calibration module, a multi-channel signal acquisition and multi-level noise suppression module, and a time-frequency analysis and quantitative corrosion assessment module.

[0014] The shortwave eddy current excitation and adaptive matching module is used to generate a wideband continuously adjustable alternating excitation signal. By pre-sweeping the frequency to measure the complex impedance response of the flange under test at each frequency point, the optimal excitation frequency is automatically determined based on the principle of maximizing the normalized impedance deviation rate and meeting the signal-to-noise ratio standard. Skin depth is used for auxiliary optimization. According to the selected frequency, the reconfigurable π-type impedance matching network is controlled to switch the combination of reactive components, so that the standing wave ratio between the excitation source and the probe coil is lower than 2.0.

[0015] The transceiver integrated single-crystal probe and dual calibration module are used to employ a dual-coil structure with coaxial nesting and electrical isolation of the transmitting and receiving coils. Before testing, air zero-point self-calibration and standard eddy current test block benchmark calibration are performed sequentially. The inherent parasitic coupling between the coils is eliminated through an adaptive cancellation algorithm, the zero-point compensation parameters are latched, and the mapping relationship between the eddy current response and the wall thickness is established using a test block of known thickness to calibrate the testing benchmark.

[0016] The multi-channel signal acquisition and multi-level noise suppression module is used to drive the probe to continuously excite the flange at the optimal excitation frequency, and to synchronously acquire the induced voltage of the receiving coil through multiple channels. The voltage is then passed through analog low-pass anti-aliasing filter and digital band-pass filter in sequence. An independent noise reference coil is used to pick up environmental electromagnetic interference. The common-mode noise component is subtracted from the main signal through an adaptive coherent noise cancellation algorithm to obtain a pure eddy current response signal.

[0017] The time-frequency analysis and corrosion quantitative assessment module is used to perform short-time Fourier transform on the pure eddy current response, extract feature parameters such as fundamental amplitude attenuation rate, phase shift, equivalent remaining wall thickness, and regional inhomogeneity, and input the feature vector into a multi-parameter corrosion quantification model calibrated by a standard corrosion sample block to inversely derive the corrosion depth and metal volume loss rate, and classify the corrosion state into three levels: normal, warning, and dangerous according to a preset three-level judgment threshold, and output the graded warning in a graphical manner.

[0018] A computer device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of a method for nondestructive testing of corrosion of metal flanges based on short-wave eddy currents.

[0019] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for nondestructive testing of corrosion of metal flanges based on short-wave eddy currents.

[0020] The beneficial effects of this invention are as follows: This invention utilizes short-wave eddy current broadband pre-sweep and adaptive impedance matching to rapidly inject excitation energy into ferromagnetic flanges, compensating for the sensitivity decrease caused by insufficient impedance mismatch in conventional fixed-frequency detection. Simultaneously, the use of high-frequency skin effect significantly enhances the resolution of early micro-corrosion. In the dual calibration stage, the adaptive cancellation mechanism effectively suppresses the probe's inherent parasitic coupling, minimizing probe contamination and corrosion at extremely low levels. Furthermore, a more traceable wall thickness quantification benchmark is established using standard test blocks, fully ensuring measurement repeatability and consistency. The invention integrates a three-stage cascaded noise suppression mechanism with an independent reference channel coherent cancellation mechanism. It deeply suppresses time-varying interference such as power frequency magnetic field and adjacent coupling, and reliably recovers very weak corrosion signals submerged in strong background noise; after simultaneously obtaining amplitude attenuation rate, phase shift, equivalent remaining wall thickness and local non-uniformity by using short-time Fourier transform, it introduces a multi-parameter model calibrated by standard corrosion sample block to invert corrosion depth and volume loss rate. This not only eliminates the risk of misjudgment caused by non-uniform permeability during single-parameter regression, but also divides it into three stages according to the procedure threshold, and unfolds the forecast with differentiated visual symbols, which can be directly used as the basis for decision-making in the field operation and maintenance process; adaptive changes on the display waveform only change the mapping relationship. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1This is a flowchart illustrating the overall process of a non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents, provided in one embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram illustrating the process of classifying corrosion safety levels in a non-destructive testing method for metal flange corrosion based on short-wave eddy currents, as provided in an embodiment of the present invention.

[0024] Figure 3 The flowchart shows a system scheme for a non-destructive testing system for metal flange corrosion based on short-wave eddy current, according to an embodiment of the present invention. Detailed Implementation

[0025] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0026] Example 1, referring to Figure 1 As an embodiment of the present invention, a non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents is provided, comprising: S100: By generating a wideband continuously adjustable alternating excitation signal, the metal flange is frequency swept, complex impedance data at each frequency point is collected, the optimal excitation frequency is determined, and the impedance matching network is adaptively switched.

[0027] S200: Performs air zero-point self-calibration and known thickness standard eddy current test block reference calibration on the transceiver single crystal eddy current probe.

[0028] S300: Synchronously acquires eddy current secondary field signals, performs signal filtering and adaptive coherent noise cancellation, performs time-frequency analysis on the denoised signal, and extracts signal features.

[0029] S400: Input the signal characteristics into the inversion model, calculate the corrosion depth and volume loss rate, and classify the corrosion safety level according to the ratio of the corrosion depth to the design wall thickness, the volume loss rate and the non-uniformity, and output the early warning information in a graphical manner.

[0030] It should be noted that this invention improves the resolution of minute corrosion through short-wave eddy current adaptive frequency selection and reconfigurable impedance matching. It achieves reliable extraction of traceable wall thickness benchmarks and weak signals through dual calibration and three-level coherent noise suppression. The time-frequency multi-feature inversion and standardized three-level early warning output can directly guide corrosion classification decisions in operation and maintenance. Combined with intelligent display adjustment, long-term power supply, high-protection structure, and microsecond-level synchronization bus, it forms a complete and reliable eddy current non-destructive testing solution for metal flanges. Example 2, refer to Figure 1 and Figure 2 This is a second embodiment of the present invention, which provides a non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents, including: In step S100, the adaptive switching impedance matching network includes steps S101 to S104: S101: Sweep the frequency with a fixed step size within the preset frequency band to obtain the complex impedance response of the probe coil at each frequency point.

[0031] The transceiver integrated single-crystal eddy current probe is coupled to the surface of the metal flange under test. The short-wave eddy current excitation unit performs a fixed-step frequency sweep within a preset frequency band of 3 to 15 MHz. In this embodiment, the fixed-step frequency sweep specifically adopts a 100 kHz step linear scan with a total of 120 frequency points and a single sweep period of no more than 50 ms.

[0032] The DSP sequentially switches the excitation frequency and acquires the complex impedance response of the probe coil at each frequency point, obtaining a complex impedance sequence, expressed by the formula: in, For the complex impedance at each frequency, For resistance components, For the reactance component, The excitation frequency.

[0033] S102: To determine the optimal excitation frequency, calculate the normalized impedance deviation rate at each frequency point as a corrosion sensitivity index and evaluate the signal-to-noise ratio index.

[0034] In this embodiment, the corrosion sensitivity index is specifically adopted as the normalized impedance deviation rate. ,in, The reference impedance of the pre-stored standard test block; signal-to-noise ratio requirements. .

[0035] S103: Select candidate frequencies with the highest corrosion sensitivity index and signal-to-noise ratio index exceeding the preset threshold. If there are multiple candidates, select the optimal excitation frequency based on the correspondence between skin depth and expected corrosion depth.

[0036] The main screening criterion is to select the frequency point with the largest normalized impedance deviation rate, and the signal-to-noise ratio of the current frequency point is not less than 20dB. When multiple frequency points meet the conditions, further optimization is performed based on the skin depth.

[0037] It should be noted that the formula for calculating skin depth is as follows: in, Frequency (unit: Hz) The relative permeability of the metal flange being measured is... The measured electrical conductivity of the metal flange is given in S / m. This example is for a steel flange used in iron towers. Take 100, σ is 5.8 × 10 6 S / m, select The frequency point is approximately twice the expected corrosion depth, ultimately locking in the optimal excitation frequency, such as 7MHz.

[0038] S104: Based on the optimal excitation frequency, control the access and switching of the reactive element combination in the reconfigurable impedance matching network, change the network topology parameters, and make the energy transmission efficiency parameter between the excitation source and the probe load reach the preset matching condition.

[0039] In this embodiment, the reconfigurable impedance matching network specifically adopts a relay-switched π-type LC four-position network, and the reactance component combination includes an input parallel capacitor, a series capacitor, an output parallel capacitor, and a fixed compensation inductor of 1.5μH.

[0040] It should be noted that the four capacitor combinations are as follows: 3-5MHz range (470pF in parallel for input, 100pF in series, 220pF in parallel for output), 5-7MHz range (220pF in parallel for input, 47pF in series, 100pF in parallel for output), 7-10MHz range (100pF in parallel for input, 22pF in series, 47pF in parallel for output), and 10-15MHz range (47pF in parallel for input, 10pF in series, 22pF in parallel for output).

[0041] It should also be noted that the DSP drives the relay to switch to the corresponding position according to the optimal excitation frequency, adjusts the network topology parameters, and makes the output impedance of the excitation source and the load impedance of the probe conjugate matched, so that the energy transmission efficiency parameter, i.e., the standing wave ratio VSWR, is ≤2.0, ensuring that the excitation power is efficiently injected into the flange.

[0042] In step S200, the execution of air zero-point self-calibration and known thickness standard eddy current test block reference calibration includes steps S201~S203: S201: Place the probe in a space without the metal being measured, transmit continuous excitation at the optimal excitation frequency, and collect the original zero-point signal generated by the inherent parasitic coupling between coils in the receiving channel.

[0043] Before testing, the probe must be calibrated to eliminate inherent errors and establish a benchmark. Perform air zero-point self-calibration. Suspend the probe in the air. At this time, there is no eddy current effect of the measured metal. The signal of the receiving coil only contains the inherent parasitic coupling between the transmitting and receiving coils. Transmit a continuous sine wave at the optimal excitation frequency determined in step S100 and collect the current original zero-point signal.

[0044] S202: Run adaptive compensation processing to generate a cancellation component with the same amplitude and opposite phase as the current original zero-point signal, inject it into the receiving link, and finely adjust it iteratively until the parasitic coupling residual is lower than the cancellation convergence threshold. Then, latch the compensation parameters as zero-point calibration data.

[0045] In this embodiment, the adaptive compensation processing method specifically adopts the 32nd order LMS adaptive algorithm with a step size of 0.005. It generates an anti-relative cancellation component with the same amplitude and opposite phase as the original zero-point signal and injects it into the front end of the receiving link. The calculation process is to gradually reduce the parasitic coupling residual by iteratively adjusting the amplitude resolution of 0.1dB and the phase step of 1°.

[0046] The cancellation convergence thresholds include: residual fundamental component ≤ 0.1% of full scale, and continuous 3-cycle variation < 0.01%.

[0047] It should be noted that when the residual fundamental component amplitude decays to below 0.1% of full scale and the residual change for three consecutive adaptive cycles is less than 0.01%, the cancellation convergence threshold is considered to have been reached. The DSP latches the compensation parameters (amplitude, phase, matching range) at this moment and writes them to the flash memory as zero-point calibration. All subsequent measurement signals will subtract the current zero-point reference.

[0048] S203: Couple the probe to a standard eddy current calibration block with nominal remaining thickness, maintain the optimal excitation frequency, obtain the eddy current response of the block, construct the thickness measurement reference relationship based on the mapping relationship between the current response characteristics and the known thickness, and complete the reference calibration.

[0049] After zero-point calibration is completed, a standard eddy current test block with known thickness is used for reference calibration. The probe is coupled to the standard eddy current calibration test block. The test block is made of the same material as the flange and has a nominal remaining thickness of 4.00 mm. Coupling agent is applied to the surface to ensure close contact between the probe and the test block.

[0050] Maintaining the optimal excitation frequency selected in step S100, the eddy current response of the test block is acquired, and the fundamental amplitude A0 and phase are extracted. The equivalent time delay τ of the eddy current response is obtained through time-frequency analysis. A mapping relationship is established between the current response characteristics and the known thickness of 4.00 mm to construct a thickness measurement benchmark relationship.

[0051] In this embodiment, the equivalent wall thickness displayed on the screen is calibrated to 4.00mm ± 0.01mm by changing the internal mapping coefficient, while the reference amplitude A0 and reference phase at the current moment are also adjusted. The reference wall thickness t0=4.00mm is stored in the flash memory as a feature reference vector, and will be switched to test mode after calibration is completed.

[0052] In step S300, the execution of signal filtering and adaptive coherent noise cancellation includes steps S301 to S306: S301: Excites the flange at the optimal excitation frequency and simultaneously samples the induced voltage on the receiving coil via multiple channels.

[0053] The calibrated probe is coupled to the flange area to be inspected under constant pressure and continuously excited at the optimal excitation frequency determined in step S100. The induced voltage on the receiving coil is synchronously acquired by the multi-channel DSP at a sampling rate of 1 MSa / s.

[0054] S302: Performs low-pass anti-aliasing processing and digital bandpass frequency selection processing on the induced voltage sequentially.

[0055] The acquired induced voltage is processed by low-pass anti-aliasing, specifically a passive LC low-pass filter with a cutoff frequency of 15MHz in this embodiment, to suppress high-frequency noise and ADC aliasing.

[0056] The signal enters the digital bandpass frequency selection process, where the DSP performs FIR bandpass filtering of 3 to 15 MHz in the passband and ≥40 dB in the stopband, filtering out low-frequency interference such as power frequency 50 Hz, harmonics, and mechanical vibration.

[0057] S303: Enable an independent noise reference coil that is coaxially nested with the main detection coil and does not contact the workpiece. Obtain the environmental electromagnetic interference as a reference signal. Subtract the interference component that is coherent with the reference signal from the main signal through an adaptive noise cancellation mechanism to obtain a pure eddy current response.

[0058] The independent noise reference coil only picks up ambient electromagnetic interference as a reference signal. The reference signal is matched to the main channel frequency response within ±1dB by the front-end gain and then fed into the adaptive noise cancellation mechanism. In this embodiment, the adaptive noise cancellation mechanism specifically adopts a 32nd-order LMS adaptive filter with a convergence step size μ=0.005. It generates an inverse compensation component in real time with the reference signal as input and injects it into the main receiving link to superimpose it with the original signal, so that the common-mode noise is attenuated by no less than 35dB. After the above three-stage processing, a pure eddy current response signal is obtained.

[0059] S304: Perform a joint time-frequency transformation on the pure eddy current response to extract the fundamental amplitude attenuation rate and fundamental phase hysteresis that are in the same frequency as the excitation in the transform domain.

[0060] In this embodiment, the joint time-frequency transformation specifically employs the short-time Fourier transform (STFT) with a window length of 1024 points. Feature parameters strongly correlated with corrosion defects are extracted from the transformation results, and the attenuation rate η of the fundamental frequency at the same frequency as the excitation is extracted. The calculation formula is expressed as follows: in, The amplitude of the fundamental frequency component at the current measurement point. The reference amplitude obtained in step S200.

[0061] The formula for extracting the fundamental phase lag is expressed as: in, For the current phase, As the reference phase, The amplitude of the extracted fundamental frequency component is attenuated.

[0062] S305: The equivalent time delay parameter of the eddy current response is determined by the time-frequency energy distribution and converted into the equivalent residual wall thickness.

[0063] The equivalent time delay parameter τ of the eddy current response is identified by the peak position of the time-frequency energy distribution. Based on the thickness measurement reference relationship established in step two, the equivalent remaining wall thickness t is calculated. In this embodiment, the wall thickness conversion relationship is expressed as follows: in, To calibrate the corresponding time delay of the test block, For the equivalent time delay parameter, The equivalent remaining wall thickness is t0 = 4.00 mm.

[0064] The formula for calculating the wall thickness reduction is expressed as follows: in, This is the amount of wall thickness reduction.

[0065] S306: Perform statistical dispersion calculation on the multi-point measurement results within the set detection area to obtain the local non-uniformity.

[0066] Perform multi-point measurements within a 30mm diameter circle centered on the current measuring point, collect several sets of wall thickness values, calculate the statistical dispersion, and obtain the local non-uniformity as the standard deviation. The calculation formula is expressed as follows: in, For local non-uniformity, This represents the equivalent remaining wall thickness at the i-th measuring point. This is the arithmetic mean of the equivalent remaining wall thickness at all measuring points within the detection area. The total number of valid measurement points within the detection area; the fundamental amplitude attenuation rate, wall thickness reduction, and local non-uniformity together constitute the corrosion characteristic vector, which is output to the quantitative evaluation unit.

[0067] like Figure 2 In step S400, the classification of corrosion safety levels according to multi-level thresholds includes steps S401~S403: S401: Input the fundamental amplitude attenuation rate, the wall thickness reduction of the equivalent remaining wall thickness relative to the reference wall thickness, and the local non-uniformity into the multi-parameter corrosion quantification model constructed from the calibration data of the standard corrosion sample block, and output the corrosion depth value and the percentage of metal volume loss.

[0068] The feature vector extracted in step S300 is input into the multi-parameter corrosion quantification model. The model is constructed from the calibration data of the standard corrosion sample block through multiple regression. In this embodiment, for steel flanges, the model adopts a linear expression: in, For equivalent corrosion depth, , and These are the coefficients determined through fitting experiments on standard corrosion samples.

[0069] The formula for calculating the percentage of metal volume loss, i.e., the corrosion volume loss rate, is expressed as: in, This represents the corrosion volume loss rate.

[0070] S402: Based on the preset safety level classification criteria, the ratio of corrosion depth to design wall thickness, percentage of metal volume loss, and local non-uniformity are compared with the three-level judgment thresholds respectively.

[0071] When all indicators are below the first-level threshold, the state is considered normal; when any indicator falls between the first-level threshold and the second-level threshold, the state is considered warning; when any indicator reaches or exceeds the second-level threshold, the state is considered dangerous.

[0072] After obtaining d and V, the corrosion state is determined according to the preset safety level classification criteria and the three-level judgment threshold. In this embodiment, the flange design wall thickness of 10mm is used as a reference, and the first-level threshold is set as follows: corrosion depth 1mm (10% of wall thickness), volume loss rate 10%, and non-uniformity 1.0mm; the second-level threshold is set as follows: corrosion depth 2mm (20% of wall thickness), volume loss rate 20%, and non-uniformity 1.5mm.

[0073] It should be noted that the judgment logic includes the following: if d < 1 mm and V < 10% and ξ < 1.0 mm, it is judged as a normal state; otherwise, if d < 2 mm and V < 20% and ξ < 1.5 mm, it is judged as a warning state; otherwise, it is judged as a dangerous state.

[0074] S403: Drives the display interface to output hierarchical warning information corresponding to the judgment status using differentiated visual identifiers and dynamic prompt symbols.

[0075] The embedded ARM processor drives a 7-inch true-color touchscreen to output hierarchical information. In the normal state, a green visual identifier is displayed statically; in the warning state, a yellow identifier is displayed with a dynamic flashing prompt symbol; in the danger state, a red identifier flashes rapidly to sound an alarm and take corresponding measures.

[0076] Furthermore, in step S400, the graphical output of the warning information includes steps S411-S413: S411: Real-time monitoring of waveform display status parameters. When the peak amplitude exceeds the upper or lower limit preset ratio of the current vertical display range, the front-end signal gain control parameters are automatically adjusted, and the waveform vertical mapping ratio is adjusted synchronously. The waveform display status parameters include the peak amplitude of the eddy current waveform, the time domain coordinates corresponding to the equivalent remaining wall thickness, and the position offset of the target analysis feature window in the display area.

[0077] When the peak amplitude exceeds 90% of the current vertical display range, the front-end signal gain control parameters are automatically adjusted. In this embodiment, the digital gain range is 0 to 90 dB, and the gain is reduced in 0.5 dB steps. The waveform vertical mapping ratio is adjusted simultaneously (from 1V / div to 2V / div) to compress the waveform and prevent clipping. When the peak value is less than 30% of the range, the gain is automatically increased and the mapping ratio is amplified to utilize the screen resolution.

[0078] S412: When the time domain coordinates corresponding to the equivalent remaining wall thickness exceed the current horizontal time base range, automatically switch the time axis scale range and widen the time axis.

[0079] When the time-domain coordinates corresponding to the equivalent remaining wall thickness t exceed the coverage range of the current horizontal time base, the time axis scale range is automatically switched from 10μs / div to 20μs / div or 50μs / div, so that the far-field eddy current response characteristics are fully displayed.

[0080] S413: When the target analysis feature window deviates from the display center area by more than the set offset criterion, the data reading trigger delay is automatically adjusted to move the feature window to the display center area.

[0081] When the target analysis feature window deviates from the center of the screen by more than a preset proportion, the data reading trigger delay is automatically adjusted, that is, the trigger delay time base is shifted, and the current feature window is moved to the center third of the screen area for easier reading.

[0082] Adjustments to signal gain control parameters, time axis scale range, and data read trigger delay only affect the display mapping layer; the original sampled data sequence remains unchanged.

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0084] Example 3, referring to Figure 3 The third embodiment of the present invention provides a non-destructive testing system for corrosion of metal flanges based on short-wave eddy currents, including a short-wave eddy current excitation and adaptive matching module, a transceiver integrated single-crystal probe and dual calibration module, a multi-channel signal acquisition and multi-level noise suppression module, and a time-frequency analysis and quantitative corrosion assessment module.

[0085] The shortwave eddy current excitation and adaptive matching module is used to generate a wideband continuously adjustable alternating excitation signal. By pre-sweeping the frequency to measure the complex impedance response of the flange under test at each frequency point, the optimal excitation frequency is automatically determined based on the principle of maximizing the normalized impedance deviation rate and meeting the signal-to-noise ratio standard. Skin depth is used for auxiliary optimization. According to the selected frequency, the reconfigurable π-type impedance matching network is controlled to switch the combination of reactive components, so that the standing wave ratio between the excitation source and the probe coil is lower than 2.0.

[0086] The transceiver integrated single-crystal probe and dual calibration module are used to employ a dual-coil structure with coaxial nesting and electrical isolation of the transmitting and receiving coils. Before testing, air zero-point self-calibration and standard eddy current test block benchmark calibration are performed sequentially. The inherent parasitic coupling between the coils is eliminated through an adaptive cancellation algorithm, the zero-point compensation parameters are latched, and the mapping relationship between the eddy current response and the wall thickness is established using a test block of known thickness to calibrate the testing benchmark.

[0087] The multi-channel signal acquisition and multi-level noise suppression module is used to drive the probe to continuously excite the flange at the optimal excitation frequency, and to synchronously acquire the induced voltage of the receiving coil through multiple channels. The voltage is then passed through analog low-pass anti-aliasing filter and digital band-pass filter in sequence. An independent noise reference coil is used to pick up environmental electromagnetic interference. The common-mode noise component is subtracted from the main signal through an adaptive coherent noise cancellation algorithm to obtain a pure eddy current response signal.

[0088] The time-frequency analysis and corrosion quantitative assessment module is used to perform short-time Fourier transform on the pure eddy current response, extract feature parameters such as fundamental amplitude attenuation rate, phase shift, equivalent remaining wall thickness, and regional inhomogeneity, and input the feature vector into a multi-parameter corrosion quantification model calibrated by a standard corrosion sample block to inversely derive the corrosion depth and metal volume loss rate, and classify the corrosion state into three levels: normal, warning, and dangerous according to a preset three-level judgment threshold, and output the graded warning in a graphical manner.

[0089] It should be noted that the entire method is implemented in a portable detection system powered by a rechargeable lithium battery pack. In this embodiment, a 12V / 10Ah lithium iron phosphate battery pack is used, equipped with a battery management circuit. The battery management circuit provides overvoltage protection (SMBJ15A TVS clamping), overcurrent protection (JK30-250 self-resetting fuse, holding current 2.5A), overtemperature protection (10K3950 NTC thermistor monitoring, 60℃ shutdown MOS switch FS8205A), short circuit protection (FS8205A built-in detection, response time <100μs), and reverse connection protection (SS34 Schottky diode). The STM32F030 microcontroller performs remaining power monitoring and automatic sleep shutdown control to ensure continuous operation for no less than 8 hours on a full charge.

[0090] The system host casing is made of ABS flame-retardant engineering plastic, which has the ability to protect against solid foreign objects and dripping water, and the protection level reaches IP52; the single crystal probe adopts a fully insulated and sealed structure, namely epoxy resin vacuum encapsulation, and the electrical interface is equipped with a waterproof and dustproof sealing ring. The insulation resistance of the whole machine is ≥50MΩ, and the dielectric strength meets 1000VAC / 0.5mA / 1min.

[0091] The system internally transmits raw acquired waveform data via a high-speed serial data bus (SPI bus in this embodiment, with a rate of approximately 80MB / s); and transmits commands such as frequency setting, gain configuration, and status query via an asynchronous serial control bus (RS485 bus in this embodiment, with a baud rate of 115200bps).

[0092] To ensure the timing synchronization of excitation, acquisition and processing, a dedicated hardware synchronization trigger circuit is set up. The main control MCU outputs a 10kHz SYNC pulse signal (rising edge valid) to synchronize the excitation signal transmission, ADC sampling start and DSP processing start time. The synchronization error of the three is less than 1μs, realizing microsecond-level coordination across the entire link.

[0093] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0094] Example 4, the fourth embodiment of the present invention, differs from the previous three embodiments in that: If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0095] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0096] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0097] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

Claims

1. A non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents, characterized in that: include, By generating a wideband continuously adjustable alternating excitation signal, the metal flange is frequency swept, complex impedance data at each frequency point is collected, the optimal excitation frequency is determined, and the impedance matching network is adaptively switched. The transceiver-integrated single-crystal eddy current probe will undergo air zero-point self-calibration and standard eddy current test block calibration of known thickness. The secondary field signal of the eddy current is acquired synchronously, signal filtering and adaptive coherent noise cancellation are performed, and time-frequency analysis is conducted on the denoised signal to extract signal features; The signal characteristics are input into the inversion model to calculate the corrosion depth and volume loss rate. Based on the ratio of the corrosion depth to the design wall thickness, the volume loss rate, and the non-uniformity, the corrosion safety level is divided according to a multi-level threshold, and the early warning information is output in a graphical manner.

2. The non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in claim 1, characterized in that: The adaptive switching impedance matching network includes sweeping the frequency with a fixed step size within a preset frequency band to obtain the complex impedance response of the probe coil at each frequency point. The normalized impedance deviation rate at each frequency point is calculated as a corrosion sensitivity index, and the signal-to-noise ratio is evaluated. Candidate frequencies with the highest corrosion sensitivity index and signal-to-noise ratio index exceeding the preset threshold are selected. If there are multiple candidates, the optimal excitation frequency is determined based on the correspondence between skin depth and expected corrosion depth. Based on the optimal excitation frequency, the connection and switching of the reactive element combination in the reconfigurable impedance matching network are controlled, and the network topology parameters are changed so that the energy transfer efficiency parameter between the excitation source and the probe load reaches the preset matching condition.

3. The non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in claim 2, characterized in that: The process of performing air zero-point self-calibration and known thickness standard eddy current test block benchmark calibration includes placing the probe in a space without the metal being measured, transmitting continuous excitation at the optimal excitation frequency, and acquiring the original zero-point signal generated by the inherent parasitic coupling between coils in the receiving channel. The adaptive compensation process is run to generate a cancellation component with the same amplitude and opposite phase as the current original zero-point signal. This component is then injected into the receiving link and finely adjusted it iteratively until the parasitic coupling residual is lower than the cancellation convergence threshold. The compensation parameters are then latched as zero-point calibration data. The probe is coupled to a standard eddy current calibration block with a nominal remaining thickness. The optimal excitation frequency is maintained, and the eddy current response of the block is obtained. Based on the mapping relationship between the current response characteristics and the known thickness, a thickness measurement benchmark relationship is constructed, and benchmark calibration is completed.

4. The non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in claim 3, characterized in that: The execution signal filtering and adaptive coherent noise cancellation include energizing the flange at the optimal excitation frequency and simultaneously sampling the induced voltage on the receiving coil via multiple channels. The induced voltage is sequentially subjected to low-pass anti-aliasing processing and digital bandpass frequency selection processing. An independent noise reference coil, which is coaxially nested with the main detection coil and does not contact the workpiece, is enabled to obtain the environmental electromagnetic interference as a reference signal. The interference component coherent with the reference signal is subtracted from the main signal through an adaptive noise cancellation mechanism to obtain a pure eddy current response. A joint time-frequency transformation is performed on the pure eddy current response. The fundamental amplitude attenuation rate and fundamental phase lag with the same frequency as the excitation are extracted in the transformation domain. The equivalent time delay parameters of the eddy current response are determined by the time-frequency energy distribution and converted into the equivalent residual wall thickness. Statistical dispersion calculation is performed on the measurement results of multiple points within the set detection area to obtain the local non-uniformity.

5. The non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in claim 4, characterized in that: The calculation of corrosion depth and volume loss rate includes inputting the fundamental amplitude attenuation rate, the wall thickness reduction amount relative to the reference wall thickness, and the local non-uniformity into a multi-parameter corrosion quantification model constructed from the calibration data of standard corrosion samples, and outputting the corrosion depth value and the percentage of metal volume loss.

6. The non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in claim 5, characterized in that: The method of classifying corrosion safety levels according to multi-level thresholds includes comparing the ratio of corrosion depth to design wall thickness, percentage of metal volume loss, and local non-uniformity with three-level judgment thresholds according to preset safety level classification criteria. When all indicators are below the first-level threshold, it is considered a normal state; when any indicator falls between the first-level threshold and the second-level threshold, it is considered a warning state; when any indicator reaches or exceeds the second-level threshold, it is considered a dangerous state. The driver displays a visually distinctive interface and uses dynamic prompts to output tiered warning information corresponding to the judgment status.

7. The non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in claim 6, characterized in that: The graphical output of early warning information includes real-time monitoring of waveform display status parameters. When the peak amplitude exceeds the upper or lower limit preset ratio of the current vertical display range, the front-end signal gain control parameters are automatically adjusted, and the waveform vertical mapping ratio is adjusted synchronously. When the time domain coordinates corresponding to the equivalent remaining wall thickness exceed the current horizontal time base range, the time axis scale range is automatically switched to widen the time axis. When the target analysis feature window deviates from the display center area by more than the set offset criterion, the data reading trigger delay is automatically adjusted to move the feature window to the display center area; Adjustments to signal gain control parameters, time axis scale range, and data read trigger delay only affect the display mapping layer; the original sampled data sequence remains unchanged.

8. A non-destructive testing system for corrosion of metal flanges based on short-wave eddy currents, employing the non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in any one of claims 1 to 7, characterized in that, It includes a shortwave eddy current excitation and adaptive matching module, a transceiver integrated single crystal probe and dual calibration module, a multi-channel signal acquisition and multi-level noise suppression module, and a time-frequency analysis and corrosion quantitative assessment module; The shortwave eddy current excitation and adaptive matching module is used to generate a wideband continuously adjustable alternating excitation signal. By pre-sweeping the frequency to measure the complex impedance response of the flange under test at each frequency point, the optimal excitation frequency is automatically determined based on the principle of maximizing the normalized impedance deviation rate and meeting the signal-to-noise ratio standard. The reconfigurable π-type impedance matching network is controlled to switch the combination of reactive components according to the selected frequency, so that the standing wave ratio between the excitation source and the probe coil is lower than 2.

0. The transceiver integrated single-crystal probe and dual calibration module are used to perform air zero-point self-calibration and standard eddy current test block reference calibration before testing by adopting a dual coil structure in which the transmitting coil and receiving coil are coaxially nested and electrically isolated. The multi-channel signal acquisition and multi-level noise suppression module is used to drive the probe to continuously excite the flange at the optimal excitation frequency, and to synchronously acquire the induced voltage of the receiving coil through multiple channels. The voltage is then passed through analog low-pass anti-aliasing filter and digital band-pass filter in sequence. An independent noise reference coil is used to pick up environmental electromagnetic interference. The common-mode noise component is subtracted from the main signal through an adaptive coherent noise cancellation algorithm to obtain a pure eddy current response signal. The time-frequency analysis and corrosion quantitative assessment module is used to perform short-time Fourier transform on the pure eddy current response, extract the fundamental amplitude attenuation rate, phase shift, equivalent remaining wall thickness, and regional inhomogeneity, and input the feature vector into the multi-parameter corrosion quantification model calibrated by the standard corrosion sample block to inversely derive the corrosion depth and metal volume loss rate, and classify the corrosion state into three levels: normal, warning, and dangerous according to the preset three-level judgment threshold, and output the graded warning in a graphical manner.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the non-destructive testing method for corrosion of metal flanges based on short-wave eddy currents as described in any one of claims 1 to 7.

Images