A porcelain insulator ultrasonic detection robot signal compensation system and method suitable for long-distance cable transmission

Abstract

The application discloses a porcelain insulator ultrasonic detection robot signal compensation system and method suitable for long-distance cable transmission, relates to the technical field of robot ultrasonic nondestructive testing, and mainly comprises the following steps: calculating a compensation coefficient according to a reference echo signal, a reverse echo signal and a key feature, compensating a measured echo signal, correcting an original standard distance-amplitude curve, compensating an initial wave amplitude, coupling and judging a compensated initial wave amplitude by using a lag comparator, denoising a processed echo signal, calculating a detection confidence degree, comparing a peak value of the denoised echo signal containing a defect with the corrected distance-amplitude curve to obtain a comparison result, and making a hierarchical decision according to the detection confidence degree. The porcelain insulator ultrasonic detection robot signal compensation system and method suitable for long-distance cable transmission can improve the consistency of multi-length cable judgment and the detection reliability.

Description

Technical Field

[0001] This invention relates to the field of robotic ultrasonic nondestructive testing technology, and more specifically, to a signal compensation system and method for a robot for ultrasonic testing of porcelain insulators suitable for long-distance cable transmission. Background Technology

[0002] Porcelain insulators are critical insulation components used extensively in power systems, and their mechanical strength and insulation performance directly affect the safe operation of transmission lines. During long-term operation, defects such as cracks, pores, and impurities may develop inside porcelain insulators. These defects gradually propagate under the combined effects of electric field stress and mechanical loads, eventually leading to insulation breakdown or mechanical fracture. Therefore, regular online inspection of the internal condition of porcelain insulators is of great significance for preventing accidents.

[0003] Ultrasonic nondestructive testing (NDT) technology is the preferred method for inspecting porcelain insulators due to its high sensitivity to internal defects in ceramic materials. In automated inspection scenarios, the main body of the ultrasonic flaw detector is usually installed in a robot control cabinet or ground workstation, while the ultrasonic probe is mounted on the actuator at the robot's end effector. The robot carries the probe and scans along the surface of the porcelain insulator, completing the inspection of internal defects in critical parts such as the skirt and neck. Due to the complex working space of substations or transmission lines and the large reach of the robot, the physical distance between the probe and the flaw detector is often much greater than the coverage area of ​​the original 2m short cable provided by the manufacturer. Therefore, on-site operations often require the use of extended probe cables of 6m, 14m, or even longer.

[0004] However, the introduction of longer cables brings several adverse effects to ultrasonic testing systems. These effects are not simply resistive attenuation, but rather a complex physical process comprised of high-frequency distributed parameter effects of the transmission line, skin effect losses, and dielectric polarization losses. To gain a deeper understanding of the physical mechanism of cable attenuation, this analysis will begin with transmission line theory.

[0005] The coaxial probe cable used in an ultrasonic testing system can be considered a lossy transmission line, and its equivalent circuit per unit length includes series distributed resistance. Series distributed inductance Parallel distributed conductivity and parallel distributed capacitance Ultrasonic pulse signals have a wide spectrum, with a center frequency typically in the megahertz range and abundant high-frequency components. For a physical length of... The transmission characteristics of the cable are determined by the propagation constant. Decide: (1) in ω is the angular frequency. Expand into real and imaginary parts: (2) in It is the attenuation constant. Let be the phase constant. Under high-frequency conditions, when ... and When the attenuation constant is , it can be approximated as: (3) In coaxial cables commonly used for ultrasonic testing, series resistance... The resistance increases with the square root of the frequency, which is due to the skin effect. Considering the high-frequency skin effect, the AC resistance of the wire is: (4) in It is a DC resistance. Let the radius be the conductor. To reach skin depth, For conductor resistivity, Let be the magnetic permeability. From equation (4), we know that... and Proportional. Dielectric polarization loss term It also increases with increasing frequency, and its relationship with frequency is as follows: (5) in Let be the tangent of the dielectric loss angle. Substituting equations (4) and (5) into equation (3), we can obtain the dependence of the attenuation constant on the frequency: (6) in For conductor loss coefficient, This is the dielectric loss coefficient.

[0006] The cable length is At that time, the signal is at a frequency The amplitude transmission gain at that point is: (7) This indicates that cable attenuation is frequency-selective and length-dependent: the higher the frequency, the more severe the attenuation; the longer the cable, the more severe the attenuation. After the ultrasonic echo signal is transmitted through the cable, its spectrum will be distorted, the high-frequency components will be significantly weakened, resulting in a reduction in the amplitude of the time-domain waveform, pulse broadening, and edge blurring.

[0007] Let the time-domain waveform of the ultrasonic echo signal output at the probe end be... Its Fourier transform is Then, after passing through a length of After transmission via cable, the signal reaching the input terminal of the flaw detector is: (8) Its peak amplitude in the time domain Peak amplitude compared to the reference 2m cable The ratio is: (9) when When the value is 6m or 14m, the ratio is significantly less than 1, which is the fundamental physical reason for the decrease in defect echo amplitude observed in field testing.

[0008] For defect determination, ultrasonic testing commonly uses the distance-amplitude (DAC) curve as a quantitative standard for defects. The DAC curve specifies the minimum allowable echo amplitude at different sound paths; an echo below this curve is considered as having no defect or insufficient defect equivalent. When the defect echo decreases overall due to cable transmission attenuation, a valid defect that was originally above the DAC curve may fall below the curve, resulting in a missed detection.

[0009] Currently, the common approach in the field is to simply increase the receiving gain of the ultrasonic flaw detector. However, this approach has a deeper physical flaw. According to the noise model of an ultrasonic testing system, the total noise of the flaw detector's receiving circuit consists of probe thermal noise, preamplifier noise, and cable coupling noise. Increasing the gain at the receiving end... Input signal and input equivalent noise Synchronously amplified: (10) in This is due to the amplifier's own additional noise. In long-distance cable transmission scenarios, The noise has been attenuated by the cable, while the thermal noise and external electromagnetic interference introduced by the cable have increased. At this point, simply increasing Although the output amplitudes of the signal and noise are increased proportionally, the signal-to-noise ratio is not fundamentally improved, and may even be worsened by amplifier-added noise and additional noise introduced by cables. As the cable length increases, even if the signal is amplified to the same peak level as the reference case, its base noise is much higher than the reference case, causing the tiny defect signal to be submerged by noise. This is the root cause of the inevitable failure of the "gain compensation method".

[0010] Furthermore, the attenuation of the initial wave amplitude caused by extended cables can interfere with the automatic coupling judgment function of the ultrasonic testing system. During automated robotic scanning, the system typically monitors the initial wave amplitude to evaluate the acoustic coupling status between the probe and the workpiece surface: a low initial wave amplitude indicates insufficient couplant or insufficient probe clamping, triggering the robotic arm to increase the positive pressure. However, if the decrease in initial wave amplitude is due to cable transmission attenuation rather than genuine poor coupling, the system will misjudge and drive the robotic arm to apply incorrect pressure. Excessive pressure can not only damage expensive ultrasonic probes but may also crush porcelain insulator skirts, causing irreversible equipment damage.

[0011] To address the aforementioned technical challenges, there is an urgent need for a systematic method that can accurately compensate for the nonlinear transmission attenuation introduced by extended cables of different lengths, synchronously correct defect judgment criteria and coupling state evaluation indicators, and provide a quantitative assessment of the reliability of the compensated signal, in order to meet the needs of the engineering application of ultrasonic testing of ceramic insulator robots. Summary of the Invention

[0012] The purpose of this invention is to provide a signal compensation system and method for ultrasonic testing robots of porcelain insulators suitable for long-distance cable transmission, which can improve the consistency of multi-length cable identification and the reliability of testing.

[0013] This invention provides a signal compensation method for an ultrasonic testing robot for porcelain insulators suitable for long-distance cable transmission, comprising the following steps: S100: Obtain the reference echo signal of the reference cable on the standard test block and the comparison echo signal of the extended cable under the same working conditions, as well as their key characteristics. Calculate the compensation coefficient based on the reference echo signal, comparison echo signal, and key characteristics, and construct a cable calibration library. S200: Obtain the current cable length and measured echo signal, and use the cable calibration library to compensate the measured echo signal to obtain the compensated echo signal; S300: The original standard distance-amplitude curve is corrected using a compensation coefficient to obtain the corrected distance-amplitude curve; S400: Based on the compensated echo signal, the amplitude of the initial wave is compensated, and the amplitude of the compensated initial wave is coupled and judged using a hysteresis comparator to obtain the processed echo signal. S500: Denoise the processed echo signal and calculate the detection confidence level. Compare the peak value of the denoised echo signal containing the defect with the corrected distance-amplitude curve to obtain the comparison result, and make a graded decision based on the detection confidence level.

[0014] Furthermore, the key features include the initial wave amplitude, the defect wave peak value, the noise mean, and the signal-to-noise ratio.

[0015] Furthermore, the compensation coefficient is a piecewise compensation coefficient, and the method for obtaining the piecewise compensation coefficient is as follows: Calculate the local compensation coefficients for each discrete sampling point in the near-field initial wave region, the mid-field defect region, and the far-field bottom wave region; Based on the local compensation coefficients, a continuous piecewise compensation function is obtained using cubic spline interpolation.

[0016] Furthermore, the formula for calculating the compensated echo signal is as follows:

[0017]

[0018] in, The compensated echo signal; The length of the current cable in the cable calibration library The corresponding compensation coefficient; This is the measured echo signal; Indicates the inverse Fourier transform; Indicates Fourier transform; For compensation system functions.

[0019] Furthermore, the method for correcting the original standard distance-amplitude curve using the compensation coefficient is as follows: Inverse compensation is performed on the discrete decision points on the original standard distance-amplitude curve using the compensation coefficient to obtain the threshold values ​​of the discrete decision points on the inversely compensated standard distance-amplitude curve, as shown in the formula:

[0020] in, Represents discrete decision points The threshold after inverse compensation; Discrete decision points on the original standard distance-amplitude curve The threshold; Discrete decision points The corresponding compensation coefficient; Based on the threshold of the discrete decision points on the standard distance-amplitude curve after inverse compensation, the corrected distance-amplitude curve is obtained by piecewise fitting of a quadratic curve. The calculation formula for piecewise fitting using a quadratic curve is as follows: , ; in, To represent the corrected distance-amplitude curve; The fitting coefficients are determined by the least squares method from three adjacent points; and Indicates the first and A discrete decision point.

[0021] Furthermore, the calculation formula for compensating the initial wave amplitude is as follows:

[0022] in, The initial wave amplitude after compensation; The compensation coefficient is the one corresponding to the initial wave. Initial amplitude.

[0023] Furthermore, the hysteresis comparator is configured as follows: The current coupling state is confirmed to be unqualified, and This flips the coupling state to a qualified state. Confirm that the current coupling state is acceptable, and This flips the coupling state to an unqualified state; in, The lag zone, The initial wave amplitude after compensation; The lower limit threshold; The upper limit threshold; This is the lag coefficient.

[0024] Furthermore, the noise reduction method is as follows: The processed echo signal is subjected to discrete wavelet transform to obtain wavelet coefficients at each scale, as shown in the formula:

[0025] in, These are wavelet coefficients; The processed echo signal; For wavelet basis functions, For scale parameters, These are translation parameters; Soft thresholding is applied to the wavelet coefficients at each scale to obtain the processed wavelet coefficients, as shown in the formula:

[0026]

[0027] in, These are the processed wavelet coefficients; It is a symbolic function; These are wavelet coefficients; For the first Wavelet thresholds corresponding to each scale; For the first Estimated standard deviation of scale noise; For the first The number of wavelet coefficients at each scale; Based on the processed wavelet coefficients, the denoised echo signal is obtained by using inverse wavelet transform.

[0028] Furthermore, the formula for calculating the detection confidence level is as follows:

[0029] ,

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036] in, To test the confidence level; For signal-to-noise ratio normalization; To normalize the level of compensation; To normalize the amplitude volatility; For waveform similarity; , , , These are the weighting coefficients for signal-to-noise ratio normalization, compensation strength normalization, amplitude volatility normalization, and waveform similarity. For signal-to-noise ratio, To achieve the minimum signal-to-noise ratio, For maximum signal-to-noise ratio; The average compensation coefficient for the defect range; This represents the maximum value of the compensation coefficient. , This marks the beginning and end of the defect interval; This is the compensation coefficient; For amplitude volatility; The number of scans; The amplitude of the defect echo in the i-th scan; This represents the average amplitude of the defect echo; The echo signal after noise reduction; This is the reference waveform.

[0037] This invention also provides a signal compensation system for an ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission, comprising the following modules: The calibration library module is used to: acquire the reference echo signal of the reference cable on the standard test block and the comparison echo signal of the extended cable under the same working conditions, as well as their key characteristics, and to determine the reference echo signal based on the reference echo signal. Compare echo signals and key features to calculate compensation coefficients and construct a cable calibration library; The signal compensation module is used to: obtain the current cable length and the measured echo signal, and use the cable calibration library to compensate the measured echo signal to obtain the compensated echo signal; The threshold correction module is used to: correct the original standard distance-amplitude curve using a compensation coefficient to obtain the corrected distance-amplitude curve; The coupling judgment and correction module is used to: compensate the initial wave amplitude based on the compensated echo signal, and use a hysteresis comparator to perform coupling judgment processing on the compensated initial wave amplitude to obtain the processed echo signal. The detection result output module is used to: denoise the processed echo signal and calculate the detection confidence level; compare the peak value of the denoised echo signal containing the defect with the corrected distance-amplitude curve to obtain the comparison result; and make a classification decision based on the detection confidence level.

[0038] The signal compensation system and method for ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission provided by this invention have the following beneficial effects: This invention addresses the technical problems of missed defect detection and misjudgment of coupling state caused by signal attenuation when using extended probe cables in existing ultrasonic testing robots for porcelain insulators. Using the original short cable as a reference, a reference signal is collected on a standard test block. Comparative signals are collected by connecting extended cables of different lengths, extracting the initial wave amplitude, defect peak value, noise mean, and signal-to-noise ratio. Segmented compensation coefficients are established and stored in a calibration library. After identifying the current cable length, compensation parameters are called to correct the measured echo, simultaneously correcting the DAC judgment curve and the initial wave coupling threshold. A hysteresis comparator is introduced to prevent oscillations that could lead to misjudgment of coupling state. After filtering, a multi-factor weighted confidence score is calculated. Based on the confidence score, defect judgment results and re-inspection suggestions are output. This fundamentally solves the problems of missed defect detection and misjudgment of coupling caused by transmission attenuation of extended cables, ensuring consistency in judgment across multiple cable lengths and improving testing reliability.

[0039] Compared with existing technologies, this invention offers three key advantages: First, precise compensation: Based on transmission line theory, a cable attenuation model is established, revealing the dependence of the attenuation constant on frequency and length. The transmission functions of different cables are obtained through calibration, and a compensation filter is designed to approximate the inverse transmission function, restoring the original spectrum of the echo signal. This avoids the signal-to-noise ratio degradation caused by simply increasing gain, fundamentally solving the problem of signal weakening in long-distance transmission. Second, consistent judgment: Using the original short cable as a unified benchmark, compensation parameters for various extended cables are calibrated offline, and a calibration library is constructed. After online compensation, the acquired signals of cables of arbitrary length are mapped to the benchmark state, ensuring high consistency between DAC defect assessment and equivalent assessment across cables, eliminating the risk of judgment standard drift. Third, elimination of coupling misjudgment and oscillation prevention: An independent compensation coefficient is used to correct the initial wave amplitude in the initial wave interval, ensuring that the correction value truly reflects the coupling state between the probe and the porcelain insulator surface. A hysteresis comparator is introduced in the coupling judgment, effectively avoiding force-controlled oscillation of the robotic arm caused by small signal fluctuations, and preventing damage to the probe and workpiece from incorrect pressure application. Fourth, high-precision compensation by region: construct a piecewise compensation function related to the sound path, divide the signal into the initial wave region, the defect region and the bottom wave region, and generate a smooth compensation curve by cubic spline interpolation to accurately match the nonlinear attenuation caused by the spectral difference in each region and effectively deal with the dispersion effect. Attached Figure Description

[0040] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart of the signal compensation method for an ultrasonic testing robot for porcelain insulators suitable for long-distance cable transmission provided by the present invention; Figure 2 This is a structural block diagram of the ultrasonic testing robot signal compensation system for porcelain insulators suitable for long-distance cable transmission provided by the present invention.

[0041] Figure 3 This is a flowchart of the calibration stage provided by the present invention.

[0042] Figure 4 This is a flowchart of the detection stage provided by the present invention. Detailed Implementation

[0043] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0044] Figure 1 A schematic diagram of a signal compensation method for an ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission, according to this embodiment, is shown. In this embodiment, the signal compensation method for an ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission includes the following steps: S100: Obtain the reference echo signal of the reference cable on the standard test block and the comparison echo signal of the extended cable under the same working conditions, as well as their key characteristics. Calculate the compensation coefficient based on the reference echo signal, comparison echo signal, and key characteristics, and construct a cable calibration library. In one exemplary embodiment, the key features include the initial wave amplitude, the defect wave peak value, the noise mean, and the signal-to-noise ratio; In one exemplary embodiment, the formula for calculating the initial wave amplitude is:

[0045] in, This is the initial wave amplitude; This indicates taking the maximum value; Represents the Hilbert transform; The sampling point number; Indicates sampling point The contrast echo signal; and Indicates the sequence number of the starting and ending sampling points; In one exemplary embodiment, the defect peak value The calculation formula is:

[0046] in, The peak value of the defect wave; This indicates taking the maximum value; Indicates sampling point The contrast echo signal; and Indicates the sequence number of the starting and ending sampling points; In one exemplary embodiment, the noise mean The calculation formula is: , ; in, This is the noise mean. This represents the number of sampling points within the noise zone. This is the contrast echo signal of the i-th sampling point within the noise region; The mean of the contrast echo signals at the sampling points within the noise region; In one exemplary embodiment, the signal-to-noise ratio is calculated using the following formula:

[0047] in, Signal-to-noise ratio; The peak value of the defect wave; This is the noise mean. In one exemplary embodiment, the compensation coefficient is an overall amplitude compensation coefficient, and the formula for calculating the overall amplitude compensation coefficient is as follows:

[0048] in, This is the overall amplitude compensation coefficient; The peak value of the defect in the reference echo signal; To compare the peak values ​​of the defective echo signals; In one exemplary embodiment, the compensation coefficient is a piecewise compensation coefficient, and the method for obtaining the piecewise compensation coefficient is as follows: Calculate the local compensation coefficients for each discrete sampling point in the near-field initial wave region, the mid-field defect region, and the far-field bottom wave region; Based on the local compensation coefficients, a continuous piecewise compensation function is obtained using cubic spline interpolation.

[0049] In one exemplary embodiment, the cable calibration library includes compensation factors and cable lengths; S200: Obtain the current cable length and measured echo signal, and use the cable calibration library to compensate the measured echo signal to obtain the compensated echo signal; In one exemplary embodiment, the formula for calculating the compensated echo signal is:

[0050] in, The compensated echo signal; The length of the current cable in the cable calibration library The corresponding compensation coefficient; This is the measured echo signal; In one exemplary embodiment, the formula for calculating the compensated echo signal is:

[0051]

[0052] in, The compensated echo signal; The length of the current cable in the cable calibration library The corresponding compensation coefficient; This is the measured echo signal; Indicates the inverse Fourier transform; Indicates Fourier transform; For compensation system functions.

[0053] S300: The original standard distance-amplitude curve is corrected using a compensation coefficient to obtain the corrected distance-amplitude curve; In one exemplary embodiment, the method for correcting the original standard distance-amplitude curve using a compensation coefficient is as follows: By using a compensation coefficient to perform inverse compensation on the discrete decision points on the original standard distance-amplitude curve, the threshold of the discrete decision points on the standard distance-amplitude curve after inverse compensation is obtained, as shown in the formula:

[0054] in, Represents discrete decision points The threshold after inverse compensation; Discrete decision points on the original standard distance-amplitude curve The threshold; Discrete decision points The corresponding compensation coefficient; Based on the threshold of the discrete decision points on the standard distance-amplitude curve after inverse compensation, the corrected distance-amplitude curve is obtained by piecewise fitting of a quadratic curve. The calculation formula for piecewise fitting using a quadratic curve is as follows: , ; in, To represent the corrected distance-amplitude curve; The fitting coefficients are determined by the least squares method from three adjacent points; and Indicates the first and A discrete decision point.

[0055] S400: Based on the compensated echo signal, the amplitude of the initial wave is compensated, and the amplitude of the compensated initial wave is coupled and judged using a hysteresis comparator to obtain the processed echo signal. In one exemplary embodiment, the calculation formula for compensating the initial wave amplitude is as follows:

[0056] in, The initial wave amplitude after compensation; The compensation coefficient is the one corresponding to the initial wave. Initial amplitude; In one exemplary embodiment, the hysteresis comparator is configured as follows: The current coupling state is confirmed to be unqualified, and This flips the coupling state to a qualified state. Confirm that the current coupling state is acceptable, and This flips the coupling state to an unqualified state; in, The lag zone, The initial wave amplitude after compensation; The lower limit threshold; The upper limit threshold; This is the lag coefficient.

[0057] S500: Denoise the processed echo signal and calculate the detection confidence level. Compare the peak value of the denoised echo signal containing the defect with the corrected distance-amplitude curve to obtain the comparison result, and make a graded decision based on the detection confidence level.

[0058] In one exemplary embodiment, the noise reduction method is as follows: The processed echo signal is subjected to discrete wavelet transform to obtain wavelet coefficients at each scale, as shown in the formula:

[0059] in, These are wavelet coefficients; The processed echo signal; For wavelet basis functions, For scale parameters, These are translation parameters; Soft thresholding is applied to the wavelet coefficients at each scale to obtain the processed wavelet coefficients, as shown in the formula:

[0060]

[0061] in, These are the processed wavelet coefficients; It is a symbolic function; These are wavelet coefficients; For the first Wavelet thresholds corresponding to each scale; For the first Estimated standard deviation of scale noise; For the first The number of wavelet coefficients at each scale; Based on the processed wavelet coefficients, the denoised echo signal is obtained by using inverse wavelet transform.

[0062] In one exemplary embodiment, the formula for calculating the detection confidence level is:

[0063] ,

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070] in, To test the confidence level; For signal-to-noise ratio normalization; To normalize the level of compensation; To normalize the amplitude volatility; For waveform similarity; , , , These are the weighting coefficients for signal-to-noise ratio normalization, compensation strength normalization, amplitude volatility normalization, and waveform similarity. For signal-to-noise ratio, To achieve the minimum signal-to-noise ratio, For maximum signal-to-noise ratio; The average compensation coefficient for the defect range; This represents the maximum value of the compensation coefficient. , This marks the beginning and end of the defect interval; This is the compensation coefficient; For amplitude volatility; The number of scans; The amplitude of the defect echo in the i-th scan; This represents the average amplitude of the defect echo; The echo signal after noise reduction; This is the reference waveform.

[0071] In one exemplary embodiment, the hierarchical decision-making process is as follows: If the detection confidence level is not less than the high confidence threshold, output a definitive conclusion of defect confirmation or no defect. If the detection confidence level is confirmed to be less than the high confidence threshold but not less than the low confidence threshold, a slowdown re-inspection instruction is triggered, and the process returns to step S200. If the detection confidence level is confirmed to be less than the low confidence threshold, an alarm will be triggered and the system will be shut down.

[0072] This embodiment provides a signal compensation system for a ceramic insulator ultrasonic testing robot suitable for long-distance cable transmission, comprising the following modules: The calibration library module is used to: acquire the reference echo signal of the reference cable on the standard test block and the comparison echo signal of the extended cable under the same working conditions, as well as their key characteristics, and to determine the reference echo signal based on the reference echo signal. Compare echo signals and key features to calculate compensation coefficients and construct a cable calibration library; The signal compensation module is used to: obtain the current cable length and the measured echo signal, and use the cable calibration library to compensate the measured echo signal to obtain the compensated echo signal; The threshold correction module is used to: correct the original standard distance-amplitude curve using a compensation coefficient to obtain the corrected distance-amplitude curve; The coupling judgment and correction module is used to: compensate the initial wave amplitude based on the compensated echo signal, and use a hysteresis comparator to perform coupling judgment processing on the compensated initial wave amplitude to obtain the processed echo signal. The detection result output module is used to: denoise the processed echo signal and calculate the detection confidence level; compare the peak value of the denoised echo signal containing the defect with the corrected distance-amplitude curve to obtain the comparison result; and make a classification decision based on the detection confidence level.

[0073] In some embodiments, the above-described signal compensation method for ultrasonic testing robots of porcelain insulators suitable for long-distance cable transmission can also be implemented in the following ways.

[0074] In this embodiment, the signal compensation method for the ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission includes a calibration stage and a testing stage; The calibration phase includes: S1. Determine the reference cable: Use the original short cable as the reference cable and collect the reference echo signal on the standard test block. ,in Indicates the time or sound path sampling point; S2. Acquire echoes from extended cables: Connect the extended cables separately and collect and compare the echo signals under the same operating conditions. , This refers to the cable length. S3. Extract key features: Extract the initial amplitude of the echo signal. Defect peak value noise mean and signal-to-noise ratio ; S4. Establish compensation coefficient: Based on the echo amplitude value under the reference cable, establish the current cable length. Corresponding single compensation coefficient Alternatively, establish path-dependent piecewise compensation coefficients. ; S5. Establish a cable calibration library: Store the parameter set corresponding to each cable length into the calibration library module; The detection phase includes: S6. Identify current cable length: Determine the current cable length using the cable length identification module. ; S7. Call compensation parameters: based on the current cable length The compensation model is retrieved from the calibration library to correct the measured echo signal. ; S8. Correct the DAC decision curve: Use a compensated signal and the original DAC curve for judgment; or correct the DAC curve without compensating the signal. Then make a judgment; S9. Correcting the initial wave coupling index: compensating for the initial wave amplitude. and based on Determine the coupling state; S10. Filtering and Signal Quality Evaluation: Perform filtering on the compensated signal and calculate the detection confidence level. ; S11. Output Results and Re-inspection Suggestions: Output the defect judgment results, when the confidence level is... If the value falls below the set threshold, a re-inspection recommendation will be triggered.

[0075] Specifically, step S3 involves extracting the initial wave amplitude. The signal envelope is extracted using Hilbert transform, and the maximum value of the envelope is searched to obtain the following: ; The defect peak value Extraction within the defect wave path length interval: ; The noise mean Calculated from the root mean square value of the defect-free region of the signal: , ; The signal-to-noise ratio .

[0076] Specifically, the segmented compensation coefficient mentioned in step S4 The method for establishing it is: to set the sound path Divided into near-field wave origin region Midfield Weaknesses Far-field bottom wave region .

[0077] The boundary condition determination formulas for each interval are as follows: ; ;in, The width of the ultrasonic excitation pulse. The longitudinal wave velocity in a porcelain insulator. The sound path in the aftershock zone of the probe. The echo path at the bottom surface of the porcelain insulator is given. After calculating the compensation coefficients at discrete points within each segmented interval, a continuous piecewise compensation function is constructed using cubic spline interpolation. .

[0078] Specifically, the cubic spline interpolation constructs a piecewise compensation function. During the process, for Discrete path nodes spline function In the interval The above satisfies: The system of equations for solving its coefficients is as follows: ; in , To find the second derivative, we need to consider the natural boundary conditions. Solving this system of equations yields the continuous compensation function. .

[0079] Specifically, the correction of the measured echo signal in step S7 is achieved through frequency domain compensation: ; Among them, the compensation system function Based on the piecewise compensation coefficient Obtained through Fourier transform: .

[0080] Specifically, the method for correcting the DAC curve in step S8 is as follows: For the standard DAC curve... Discrete decision points on Inverse compensation is performed using piecewise compensation coefficients: For the corrected discrete points Piecewise fitting of quadratic curves is used to maintain the smoothness of the DAC curve: , Fit coefficients It is determined by the least squares method using three adjacent points.

[0081] Specifically, the determination of coupling state in step S9 employs hysteresis comparator logic to prevent the robotic arm from oscillating in a critical coupling state; an upper limit threshold is set. Lower threshold and lag zone : When the initial amplitude is compensated From below the lower threshold, and satisfying When the coupling is deemed successful, it is considered qualified. Decrease from above the upper limit threshold, and satisfy At that time, poor coupling was determined; hysteresis band ,in This is the lag coefficient.

[0082] Specifically, the detection confidence level described in step S10 Determined by the weighted multi-factor evaluation formula: ; Wherein, the compensation strength normalization is used The value is taken as the average compensation coefficient of the defect interval. Signal-to-noise ratio normalization ; Normalization of compensation levels Amplitude volatility Its normalization Waveform similarity The weighting coefficients satisfy ,and .

[0083] Specifically, the signal compensation method for ultrasonic testing robots of porcelain insulators suitable for long-distance cable transmission also includes performing a wavelet threshold denoising step on the compensated signal during the detection phase: [The text abruptly ends here, likely due to an incomplete sentence or a missing section.] Perform discrete wavelet transform: ;in For wavelet basis functions, For scale parameters, For translation parameters; for wavelet coefficients at each scale Soft thresholding is used: ; threshold , For the first Estimated standard deviation of scale noise; The denoised signal is then reconstructed using inverse wavelet transform. .

[0084] In some embodiments, the above-described signal compensation system for ultrasonic testing robots of porcelain insulators suitable for long-distance cable transmission can also be implemented in the following ways.

[0085] In this embodiment, the signal compensation system for the ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission includes: An ultrasonic flaw detector is used to generate ultrasonic excitation and output echo signals. An ultrasonic probe, mounted at the end of the robot, is used to contact the porcelain insulator and emit / receive ultrasonic waves; Extend the probe cable to connect the ultrasonic flaw detector to the ultrasonic probe; The robotic actuator, carrying the ultrasonic probe, performs the actions of approaching, pressing, and scanning. The signal acquisition module is used to acquire ultrasonic echo signals; The cable length identification module is used to determine the length or compensation setting of the extended probe cable currently in use. The calibration library module is used to store compensation coefficients, noise levels, signal-to-noise ratios, and confidence parameters for different cable lengths; The signal compensation module is used to perform amplitude compensation or segmented correction on the measured echo signal according to the called compensation parameters; The threshold correction module is used to correct the DAC curve and the warning threshold. The coupling judgment and correction module is used to correct the initial wave coupling threshold. The test result output module is used to output the compensated defect judgment results, signal quality confidence level, and re-inspection suggestions; The data flow and interface logic between the modules execute the method as described in any one of claims 1 to 9.

[0086] Specifically, the cable length identification module includes a readable storage chip installed at the interface end of the extended probe cable, and the storage chip internally stores the cable length code and calibration parameter version number; The read / write timing of the readable storage chip adopts a single-bus protocol or other equivalent communication protocol, including reset pulse-acknowledgment pulse-command byte-data byte. The ultrasonic flaw detector or signal acquisition module communicates with the storage chip via a data cable and reads the calibration library data corresponding to the cable length encoding.

[0087] Specifically, the mechanical model of the robot actuator carrying the ultrasonic probe to press the porcelain insulator is as follows: ; in For the applied positive pressure, The equivalent elastic modulus of the coupling agent layer. This represents the displacement of the probe as it is pressed into the coupling agent layer. This is the equivalent damping coefficient of the coupling agent layer. For the insertion speed; for the coupling agent thickness With the initial wave amplitude Satisfying empirical relationships: ; in The zero-gap reference amplitude, This is the coupling layer attenuation factor; the coupling judgment and correction module sets the upper limit of the clamping force based on this mechanical model. To avoid misjudging the coupling state and causing This can cause damage to the probe or porcelain insulator.

[0088] In some embodiments, the above-described signal compensation method and system for ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission can also be implemented in the following ways.

[0089] This embodiment applies to porcelain insulator inspection robots equipped with various industrial-grade ultrasonic flaw detectors. The robot's actuator is typically a multi-degree-of-freedom serial robotic arm, with a probe gripper and flexible pressure adapter installed at the end. In substations or transmission line sites, the robotic arm needs to move over a wide area on porcelain insulator strings to complete a full-coverage scan of each insulator skirt and neck. Because the robotic arm's workspace can span several meters to over ten meters, extended probe cables must be used between the probe and the flaw detector located on the control panel or the robot body.

[0090] Reference Figure 2 The system in this embodiment includes an ultrasonic flaw detector, an ultrasonic probe, an extended probe cable, a robot actuator, a signal acquisition module, a cable length identification module, a calibration library module, a signal compensation module, a judgment threshold correction module, a coupling judgment correction module, and a detection result output module.

[0091] The ultrasonic flaw detector is an industrial testing instrument equipped with a standard ultrasonic channel, which outputs unprocessed radio frequency A-scan signals. The flaw detector connects to the signal acquisition module via a high-speed data interface. The core of the signal acquisition module is a high-performance microcontroller that receives the digitized A-scan data output by the flaw detector through batch data transmission.

[0092] The format of an A-scan data frame is as follows: frame header, cable identification code, number of sampling points, sampling rate information, channel number, data payload area, checksum, and frame trailer. The checksum uses a Cyclic Redundancy Check (CRC) polynomial algorithm to ensure the integrity of high-speed data transmission.

[0093] The cable length identification module communicates with a memory chip installed at the extension probe cable interface via a single-bus protocol or other equivalent communication protocol. A non-volatile read / write memory chip is integrated within the flaw detector connector of the extension cable, containing a cable length code, factory calibration parameter version number, and a unique serial number. After system power-on, the microcontroller simulates communication timing via general-purpose input / output (GPIO) pins to read the memory data. The read length code uniquely identifies the currently used cable length (e.g., 0x06 represents a 6m cable, 0x0E represents a 14m cable). The system then indexes the corresponding set of compensation parameters in the calibration library module based on this code.

[0094] The calibration library module uses non-volatile memory chips to store parameters. The stored data structure is as follows: the first sector stores the number of cable lengths; subsequent sectors store a set of calibration parameters for one type of cable, specifically including: cable length code, segment interval boundaries. and Sampling point index, near-field initial wave region compensation coefficient The table includes: piecewise compensation coefficients, cubic spline interpolation parameter matrix, reference noise level, reference signal-to-noise ratio, reference waveform reference data, and coupling judgment threshold. and All parameters are written offline during the calibration phase and accessed in read-only mode during the testing phase.

[0095] The mechanical control of the robot actuator employs an impedance control strategy. The robot controller monitors the contact force between the probe and the porcelain insulator surface in real time via force sensors and performs closed-loop adjustment based on a preset target value for positive pressure. An ultrasonic coupling agent is filled between the probe tip and the porcelain insulator surface, forming a coupling layer of variable thickness. The relationship between the probe clamping force and the coupling layer thickness follows a mechanical equilibrium as follows: (11) in The equivalent elastic modulus of the coupling agent layer. This represents the displacement caused by the probe being pressed in. This is the equivalent damping coefficient during the extrusion flow of the coupling agent. This represents the probe insertion speed.

[0096] Initial amplitude The relationship with the coupling layer thickness was obtained through extensive experimental calibration and satisfies the exponential decay model: (12) in The reference initial wave amplitude under ideal coupling conditions. This is the attenuation factor of ultrasound waves in the coupling agent. When poor coupling leads to… When it is too big, If the value falls below the lower threshold, a re-compression process will be triggered.

[0097] The calibration phase is performed offline in a controlled laboratory environment, and the process is as follows: Figure 3 As shown. The core objective of calibration is to establish a complete set of compensation parameters for each type of extended cable to be used.

[0098] Step S1: Identify the reference cable and acquire the reference signal.

[0099] Select the 2m coaxial probe cable provided with the ultrasonic flaw detector as the reference cable. Set the object to be inspected as a standard ultrasonic test block (e.g., CSK-IA type), with a known-sized and positioned transverse through-hole as an artificial defect. Connect the probe to the flaw detector via the reference cable, and set the flaw detector to single-crystal straight probe mode, with the excitation voltage... Pulse width Receive gain Filter bandwidth The probe is aligned with a transverse through-hole at a specific depth on the test block, and a constant positive pressure is applied. And maintain a stable coupling agent thickness. Under this condition, the acquisition length is... The A-scan signal frames from each sampling point are used as the reference echo signal. ,in This refers to the sampling point number. Continuous sampling. Frames are averaged to reduce random noise, and the mean signal is stored in the calibration library as the final reference signal.

[0100] Step S2: Collect the comparison echo signal of the extended cable.

[0101] Keep all instrument parameters, probe position, coupling conditions, and clamping force unchanged from step S1, only replacing the cable with the 6m extended cable to be calibrated. Data acquisition. Point A scan signal frame On average Frame. Replace with a 14m extension cable in the same manner and collect data. This process strictly adheres to the "single variable principle," meaning that only the cable length changes, ensuring that signal differences stem entirely from differences in cable transmission characteristics.

[0102] Step S3: Extract key features.

[0103] Feature parameters are extracted from each group of signals acquired in steps S1 and S2. Taking a 6m cable signal as an example... For example, the extraction process is as follows: Initial amplitude Extraction: First, perform a Hilbert transform on the signal to obtain the envelope. For real signals... Its analytical signal is: (13) in This represents the Hilbert transform operator. The envelope signal is the modulus of the analytic signal: (14) Within the pre-set initial sound path range Maximum value of inner search envelope: (15) Defect peak value Extraction: Determine the defect echo path range based on the known defect depth. Within this interval, directly search for the maximum absolute value of the original signal: (16) noise mean Extraction: Select a pure noise segment and calculate the total noise level within that segment. Root mean square value of each sampling point: (17) in This represents the average value of the noise segment.

[0104] Signal-to-noise ratio calculation: (18) Similarly, feature parameters of the 2m reference signal are extracted. , , , And the characteristic parameters of the 14m extended cable signal.

[0105] Step S4: Establish the compensation coefficients and piecewise compensation functions.

[0106] First, calculate the overall amplitude compensation coefficient. For a 6m cable, define the reference amplitude ratio: (19) This coefficient reflects the attenuation ratio of the waveform amplitude due to defects introduced by the 6m cable. The calculation is similar. .

[0107] However, cable transmission attenuation is frequency-dependent, and a single overall compensation coefficient is insufficient to accurately recover the signal across the entire sound path. Therefore, this invention constructs a piecewise compensation function. .

[0108] Segmentation strategy: dividing the sound path Divided into three characteristic regions: near-field wave origin region Midfield Weakness Far-field bottom wave region .

[0109] boundary and The determination is based on the ultrasonic physical process: (20) (twenty one) in To excite the pulse width, The longitudinal wave velocity in a porcelain insulator. The sound path in the aftershock zone of the probe. The sampling point corresponds to the peak value of the bottom wave.

[0110] Within each segment interval, the local amplitude ratio is extracted at the corresponding sound path positions of the reference signal and the extended cable signal. Within the mid-field defect zone, the local compensation coefficient is calculated point-by-point, centered on the defect peak value. (twenty two) get Compensation coefficients for discrete nodes Then, a piecewise compensation function with continuous first and second derivatives and smoothness is constructed using cubic spline interpolation. In the interval formed by every two adjacent nodes Above, spline function Defined as a cubic polynomial: (twenty three) set up , Let be the second derivative at the node. From the continuity condition of the cubic spline, we can derive the following about . The three-moment equation system: (twenty four) This system of equations is for common Equations, containing One unknown to Supplementing natural boundary conditions The equation is closed and solvable. After solving using the chasing method, the spline coefficients are substituted back from the following equation: (25) (26) (27) (28) The final piecewise compensation function It has continuous first and second derivatives throughout the entire sound path range, which can smoothly compensate and correct the signal amplitude at any sound path position.

[0111] Step S5: Establish a cable calibration library.

[0112] The calibration parameters for 6m and 14m cables are packaged into separate parameter sets. and The parameters are sequentially written into the memory of the calibration library module, with a header attached to each parameter set, containing the cable length identifier, parameter version number, calibration date, and CRC check value.

[0113] The execution process of the detection phase is as follows: Figure 4 As shown, this stage runs in real time during the robot's on-site inspection.

[0114] Step S6: Identify the current cable length.

[0115] After the system is powered on, the cable length identification module reads the length code from the extended cable interface storage chip through the communication protocol and stores it in a global variable. .

[0116] Step S7: Call the compensation parameters and correct the measured echo.

[0117] Based on the identified cable length Read the corresponding piecewise compensation function from the calibration library module. The system acquires the A-scan signal of the current measurement point in real time. Compensation will be provided.

[0118] Direct time-domain compensation is equivalent to point-by-point multiplication: (29) This method requires minimal computation and is suitable for real-time processing by microcontrollers.

[0119] Frequency domain compensation can also be used: (30) Where the compensation system function for: (31) The advantage of the frequency domain method lies in its seamless integration with subsequent filtering algorithms, but its computational cost is higher than that of the time domain method. To meet real-time requirements, this invention preferably adopts a time domain compensation scheme, but retains frequency domain compensation as an optional alternative for high-precision processing.

[0120] Step S8: Correct the DAC decision curve.

[0121] To reduce online computational overhead, this embodiment employs an equivalent strategy of "directly correcting the DAC curve without compensating the signal." Let the original standard DAC curve (distance-amplitude curve) be used. Defined above A discrete decision point, for the current cable length The threshold value of the corrected DAC curve at each decision point is: (32) This formula is based on a rigorous derivation of the equivalence of the determination before and after compensation: Equivalent to To maintain the smoothness of the DAC curve, for Perform piecewise fitting of the quadratic curve: (33) The fitting coefficients are determined by using the least squares method from three adjacent points. The corrected DAC curve obtained in this way is smooth and can fully reflect the path dependence characteristics of the compensation coefficients.

[0122] Step S9: Correct the initial wave coupling index and introduce anti-oscillation logic.

[0123] First, compensate for the initial wave amplitude: (34) However, the compensated initial wave amplitude may still fluctuate within the critical range due to surface contaminants or mechanical vibrations. To avoid frequent reversals in the coupling state judgment that could cause oscillations in the robotic arm force control system, this invention introduces a hysteresis comparator logic.

[0124] Set upper limit threshold Lower threshold and lag zone : (35) in This is the lag factor, which can be set in the system configuration. The state transition logic of the lag comparator is as follows: when the current coupled state is "unqualified", it is necessary to... Rise to more than Only then is it flipped to "qualified"; when the coupling state is currently "qualified", it needs to be changed to "qualified". Drop to below Only then does it flip to "unqualified". This hysteresis comparator effectively eliminates frequent state switching caused by small fluctuations, ensuring the stable operation of the robotic arm force control system.

[0125] Step S10: Filtering and confidence evaluation.

[0126] After compensation, for Selective bandpass filtering and wavelet thresholding can be performed to further improve signal quality. Appropriate wavelet basis functions are selected to denoise the signal. Discrete wavelet transform: (36) in For discrete wavelet basis functions, For scale parameters, For translation parameters. For detail coefficients. Apply soft thresholding: (37) The threshold uses a general threshold estimation method: (38) in For estimating the median absolute deviation (MAD) of the noise standard deviation, This represents the number of scaling coefficients. The denoised signal is reconstructed using inverse wavelet transform. .

[0127] To quantitatively assess the reliability of the test results, a weighted confidence evaluation function is constructed. The definitions of each dimension's indicators are as follows: Signal-to-noise ratio normalization measures the intensity level of the current signal relative to the recorded extreme value. (39) The compensation strength normalization characterizes the intensity of compensation applied due to cable attenuation. A larger compensation level increases the risk of raising the noise floor and lowers the reliability. Here... The value is taken as the average compensation coefficient for the defect interval: (40) (41) Amplitude volatility reflects continuous Stability of defect echo amplitude during secondary scanning: (42) Its normalization process: (43) Waveform similarity is calculated by determining the normalized cross-correlation coefficient between the denoised and compensated signal and the reference waveform to evaluate the waveform fidelity. (44) The final weighted confidence score is synthesized by weighting the above four factors: (45) Among them, each weight satisfies In particular, the signal-to-noise ratio (SNR), as the most crucial metric, is typically given the highest weight, for example... The remaining weights are allocated based on engineering experience. This design ensures that the confidence assessment system has the highest sensitivity to noise pollution, while also taking into account the impact of compensation strength, measurement stability, and waveform fidelity.

[0128] Step S11: Output results and re-inspection suggestions.

[0129] The system is based on the defect wave peak value and the corrected DAC curve. The comparison results, combined with confidence levels Entering a hierarchical decision-making process: when When the confidence level is sufficient, output a definitive conclusion of "defect confirmed" or "no defect"; when If the confidence level is insufficient, a speed reduction and re-inspection command is triggered. The robot automatically reduces the scanning speed to a preset percentage (e.g., 50%) of the original speed to increase the scanning resolution and the amount of data collected, and re-executes steps S7 to S10 for a second evaluation. when If the confidence level is severely insufficient, the system will alarm and shut down, prompting the operator to check the cable connection status, coupling condition, and system noise level, and recommending that the cable be replaced with a reference short cable for manual verification. and The confidence threshold can be configured by the system based on the balance between the risk of missed detection and the false alarm rate.

[0130] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A signal compensation method for an ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission, characterized in that, Includes the following steps: S100: Obtain the reference echo signal of the reference cable on the standard test block and the comparison echo signal of the extended cable under the same working conditions, as well as their key characteristics. Calculate the compensation coefficient based on the reference echo signal, comparison echo signal, and key characteristics, and construct a cable calibration library. S200: Obtain the current cable length and measured echo signal, and use the cable calibration library to compensate the measured echo signal to obtain the compensated echo signal; S300: The original standard distance-amplitude curve is corrected using a compensation coefficient to obtain the corrected distance-amplitude curve; S400: Based on the compensated echo signal, the amplitude of the initial wave is compensated, and the amplitude of the compensated initial wave is coupled and judged using a hysteresis comparator to obtain the processed echo signal. S5 00: Denoise the processed echo signal and calculate the detection confidence level. Compare the peak value of the denoised echo signal containing the defect with the corrected distance-amplitude curve to obtain the comparison result. Make a graded decision based on the detection confidence level.

2. The method according to claim 1, characterized in that, The key features include initial wave amplitude, defect peak value, noise mean, and signal-to-noise ratio.

3. The method according to claim 1, characterized in that, The compensation coefficient is a piecewise compensation coefficient, and the method for obtaining the piecewise compensation coefficient is as follows: Calculate the local compensation coefficients for each discrete sampling point in the near-field initial wave region, the mid-field defect region, and the far-field bottom wave region; Based on the local compensation coefficients, a continuous piecewise compensation function is obtained using cubic spline interpolation.

4. The method according to claim 1, characterized in that, The formula for calculating the compensated echo signal is as follows: in, The compensated echo signal; The length of the current cable in the cable calibration library The corresponding compensation coefficient; This is the measured echo signal; Indicates the inverse Fourier transform; Indicates Fourier transform; For compensation system functions.

5. The method according to claim 1, characterized in that, The method for correcting the original standard distance-amplitude curve using a compensation coefficient is as follows: Inverse compensation is performed on the discrete decision points on the original standard distance-amplitude curve using the compensation coefficient to obtain the threshold values ​​of the discrete decision points on the inversely compensated standard distance-amplitude curve, as shown in the formula: in, Represents discrete decision points The threshold after inverse compensation; Discrete decision points on the original standard distance-amplitude curve The threshold; Discrete decision points The corresponding compensation coefficient; Based on the threshold of the discrete decision points on the standard distance-amplitude curve after inverse compensation, the corrected distance-amplitude curve is obtained by piecewise fitting of a quadratic curve. The calculation formula for piecewise fitting using a quadratic curve is as follows: ， ； in, To represent the corrected distance-amplitude curve; The fitting coefficients are determined by the least squares method from three adjacent points; and Indicates the first and A discrete decision point.

6. The method according to claim 1, characterized in that, The calculation formula for compensating the initial wave amplitude is as follows: in, The initial wave amplitude after compensation; The compensation coefficient is the one corresponding to the initial wave. Initial amplitude.

7. The method according to claim 1, characterized in that, The hysteresis comparator is configured as follows: The current coupling state is confirmed to be unqualified, and This flips the coupling state to a qualified state. Confirm that the current coupling state is acceptable, and This flips the coupling state to an unqualified state; in, The lag zone, The initial wave amplitude after compensation; The lower limit threshold; The upper limit threshold; This is the lag coefficient.

8. The method according to claim 1, characterized in that, The noise reduction method is as follows: The processed echo signal is subjected to discrete wavelet transform to obtain wavelet coefficients at each scale, as shown in the formula: in, These are wavelet coefficients; The processed echo signal; For wavelet basis functions, For scale parameters, These are translation parameters; Soft thresholding is applied to the wavelet coefficients at each scale to obtain the processed wavelet coefficients, as shown in the formula: in, These are the processed wavelet coefficients; It is a symbolic function; These are wavelet coefficients; For the first Wavelet thresholds corresponding to each scale; For the first Estimated standard deviation of scale noise; For the first The number of wavelet coefficients at each scale; Based on the processed wavelet coefficients, the denoised echo signal is obtained by using inverse wavelet transform.

9. The method according to claim 1, characterized in that, The formula for calculating the detection confidence level is as follows: ， in, To test the confidence level; For signal-to-noise ratio normalization; To normalize the level of compensation; To normalize the amplitude volatility; For waveform similarity; , , , These are the weighting coefficients for signal-to-noise ratio normalization, compensation strength normalization, amplitude volatility normalization, and waveform similarity. For signal-to-noise ratio, To achieve the minimum signal-to-noise ratio, For maximum signal-to-noise ratio; The average compensation coefficient for the defect range; This represents the maximum value of the compensation coefficient. , This marks the beginning and end of the defect interval; This is the compensation coefficient; For amplitude volatility; The number of scans; The amplitude of the defect echo in the i-th scan; This represents the average amplitude of the defect echo; The echo signal after noise reduction; This is the reference waveform.

10. A signal compensation system for an ultrasonic testing robot of porcelain insulators suitable for long-distance cable transmission, characterized in that, The system includes the following modules: The calibration library module is used to: acquire the reference echo signal of the reference cable on the standard test block and the comparison echo signal of the extended cable under the same working conditions, as well as their key characteristics, and to determine the reference echo signal based on the reference echo signal. Compare echo signals and key features to calculate compensation coefficients and construct a cable calibration library; The signal compensation module is used to: obtain the current cable length and the measured echo signal, and use the cable calibration library to compensate the measured echo signal to obtain the compensated echo signal; The threshold correction module is used to: correct the original standard distance-amplitude curve using a compensation coefficient to obtain the corrected distance-amplitude curve; The coupling judgment and correction module is used to: compensate the initial wave amplitude based on the compensated echo signal, and use a hysteresis comparator to perform coupling judgment processing on the compensated initial wave amplitude to obtain the processed echo signal. The detection result output module is used to: denoise the processed echo signal and calculate the detection confidence level; compare the peak value of the denoised echo signal containing the defect with the corrected distance-amplitude curve to obtain the comparison result; and make a classification decision based on the detection confidence level.

Images