Steel wire rope defect detection method and device based on multi-specification TMR sensor array

Abstract

The invention belongs to the technical field of steel wire rope defect detection, and discloses a steel wire rope defect detection method and device based on a multi-specification TMR sensor array, and the method comprises the steps: obtaining a detection signal of the multi-specification TMR sensor array, and carrying out the preprocessing; filtering by adopting a maximum value filter to obtain a filtered detection signal; if the wave crest greater than the defect threshold exists, judging that the steel wire rope region corresponding to the wave crest has a defect; calculating the difference value between the crest with the defect and the troughs on the two sides, and judging whether the defect type of the steel wire rope area corresponding to the crest with the defect is an LF type defect or an LMA type defect according to the difference value on the two sides; extracting time indexes of wave troughs on two sides of the wave crest with defects, forming defect time intervals, and combining the defect time intervals with intersections; and calculating the peak-to-peak value of the TMR sensor in the defect time interval, and quantitatively calculating the section loss rate of the defective steel wire rope region. The LF type defect and the LMA type defect are detected at the same time, robustness is good, and real-time performance and accuracy are high.

Description

Technical Field

[0001] The present invention belongs to the technical field of wire rope defect detection, and particularly relates to a wire rope defect detection method and device based on a multi-specification TMR sensor array. Background Technique

[0002] Wire ropes are widely used in various coal mine hoists, elevators, passenger aerial ropeways and other related fields. During use, various damages of different degrees will occur to the wire ropes. Generally, the damages of wire ropes can be divided into the following two major categories:

[0003] (1) Localized Fault (LF type for short). Damages occurring at local positions on the wire rope mainly include various types of broken wire damages, rust spots, local shape abnormalities, etc.

[0004] (2) Loss of Metallic Area (LMA type for short). Damages that reduce the metal cross-sectional area within a relatively long axial range of the wire rope mainly include long-distance wear, large-area rust, relatively long tensile deformation, etc.

[0005] To ensure the safe use of wire ropes, a non-destructive flaw detector needs to be used to perform non-destructive flaw detection on the state of wire ropes. During use, due to the inconsistent changes in the wire rope defect signals caused by the LF type defects and LMA type defects of the wire rope, and the slightly different criteria for judging whether a wire rope is scrapped, for the detection of wire ropes, it is first necessary to judge the damage type, and then perform quantitative analysis on the wire rope defects. Among them, the cross-sectional loss of wire rope defects directly affects the quality of wire ropes, so the quantitative detection of the cross-sectional loss of wire rope defects is the most important. At present, the wire rope non-destructive flaw detector based on the magnetic inspection method has always been recognized as the most reliable wire rope detection instrument.

[0006] For the judgment of LF type defects and LMA type defects, a relatively easy way to achieve is to detect the waveform width of relevant magnetic signals. For example, in the invention patent with the application number CN201910904849.4, by preprocessing the magnetic flux signal and the magnetic leakage signal, relevant characteristic quantities of the width are obtained, and compared and analyzed with a preset width threshold. If the width characteristic is greater than or equal to the preset width threshold, the cross-sectional loss of the measured wire rope is obtained according to the magnetic flux characteristic value; if the defect width is less than the preset width value, the cross-sectional loss of the measured wire rope is obtained according to the magnetic flux characteristic value and the magnetic leakage characteristic value. Although this method of distinguishing LF type defects and LMA type defects from threshold analysis is easy to distinguish respectively, the setting of the threshold requires strong technical experience, and it is insufficient to use only the waveform width as the basis for judging the defect type.

[0007] In the invention patent with the application number CN201910988091.7, based on the magnetic flux leakage data of the steel wire rope collected by the magnetic sensor, a fuzzy comprehensive evaluation strategy is adopted. Based on the expert experience database summarized from practical experience, comprehensively referring to the signal amplitude, signal width, and standard deviation of each radial component, accurate judgment of LF-type defects and LMA-type defects is carried out, and then precise quantitative analysis of the degree of damage to the steel wire rope is carried out. Compared with the previous analysis methods, the accuracy of this analysis method has been greatly improved, but the expert experience data is not universal under different magnetic circuits, and the principle criteria for accurately distinguishing the two types of defects need to be further explored.

[0008] To sum up, the existing magnetic flux leakage detection technology for steel wire ropes has done a lot of research in aspects such as excitation structure, signal acquisition, and damage identification, and has made great progress. However, due to factors such as the complex structure, diverse types, and variable usage scenarios of steel wire ropes, there are the following problems in the damage identification of steel wire ropes:

[0009] (1) In terms of sensor signal capture: There are currently many ways to obtain the magnetic flux leakage signals of damaged mine-used steel wire ropes, but there are still limitations. Coil sensors are greatly affected by the detection speed and have heat generation problems. Magnetoresistive sensors have high sensitivity and high requirements for the fluctuation of the lift-off distance. Hall sensors, as the most commonly used magnetic flux leakage detection sensors, the large number of their arrangements leads to a complex conditioning circuit structure and unstable electrical performance.

[0010] (2) In terms of on-line detection: Due to the complex structure, harsh working environment, and difficult detection of mine-used steel wire ropes, there is currently little research on on-line detection systems or devices for broken wire damage of mine-used steel wire ropes. And the existing on-line detection systems only process the data reported by the probe and cannot display and mark the defects during the detection process.

[0011] (3) Defect magnetic field range: When detecting larger LMA-type defects, a relatively large magnetic flux leakage detection range is required for the magnetic induction intensity range, and the selected magnetic sensor requires a relatively large linear range, while the coefficient corresponding to its sensor sensitivity is small, and the detection adaptability to smaller defects is poor. Summary of the Invention

[0012] The purpose of the present invention is to provide a steel wire rope defect detection method and device based on a multi-specification TMR sensor array, which can detect LF-type defects and LMA-type defects simultaneously, with good robustness, high real-time performance, and high accuracy.

[0013] To achieve the above purpose, the technical solutions adopted by the present invention are as follows:

[0014] First aspect: A steel wire rope defect detection method based on a multi-specification TMR sensor array, including:

[0015] Obtain the detection signals of the multi - specification TMR sensor array and perform pre - processing. The multi - specification TMR sensor array is evenly arranged circumferentially along the steel wire rope. In the multi - specification TMR sensor array, TMR sensors of different specifications are arranged alternately. The TMR sensors of different specifications include large - linear - range TMR sensors with a maximum linear range greater than the linear threshold, and small - linear - range TMR sensors with a maximum linear range less than or equal to the linear threshold;

[0016] Use a maximum - value filter to filter the pre - processed detection signals to obtain the filtered detection signals;

[0017] If there is a peak greater than the defect threshold in the filtered detection signals, it is determined that there is a defect in the steel wire rope area corresponding to this peak; otherwise, there is no defect;

[0018] Calculate the difference between the peak with a defect and the wave valleys on both sides. According to the differences on both sides, determine whether the defect type of the steel wire rope area corresponding to the peak with a defect is an LF - type defect or an LMA - type defect;

[0019] Extract the time indices of the wave valleys on both sides of the peak with a defect to form the defect time interval corresponding to the peak with a defect, and merge the defect time intervals with intersections;

[0020] According to the defect type, calculate the peak - to - peak value of the TMR sensor within the defect time interval, and quantitatively calculate the cross - sectional loss rate of the steel wire rope area with a defect based on the peak - to - peak value.

[0021] The following also provides several optional methods, which are not additional limitations to the above overall solution, but only further supplements or optimizations. Without technical or logical contradictions, each optional method can be combined with the above overall solution alone, or multiple optional methods can be combined with each other.

[0022] Preferably, the pre - processing includes outlier removal, wavelet denoising, and channel equalization processing:

[0023] The outlier removal includes: for the detection signal of a single TMR sensor, screen the outliers in the detection signal, and take the numerical average of the sampling points on both sides of the outlier to replace the outlier;

[0024] The wavelet denoising includes: for the detection signal of a single TMR sensor after outlier removal, use the wavelet threshold denoising method for processing;

[0025] The channel equalization processing includes: using the defect - surrounding method to obtain the sensitivity coefficient of each TMR sensor, and using the sensitivity coefficient to perform weighted correction on the detection signal of the corresponding TMR sensor after wavelet denoising to obtain the pre - processed detection signal.

[0026] Preferably, determining the defect type of the wire rope area corresponding to the wave peak with defects as LF type defect or LMA type defect according to the difference between the two sides includes:

[0027] Calculating the difference between the defect threshold and the detection signal when there are no defects in the detection signal preprocessed by the TMR sensor, and recording it as the classification difference;

[0028] If the differences on both sides exceed the classification difference, the defect type is LF type defect;

[0029] If the difference on one side exceeds the classification difference and the difference on the other side does not exceed the classification difference, and the adjacent wave peaks on both sides of the current wave peak have the same rule, the defect type is LMA type defect;

[0030] Otherwise, do not determine the defect type of the current wave peak.

[0031] Preferably, merging the defect time intervals with intersections includes:

[0032] If the defect types corresponding to the defect time intervals with intersections are all LF type defects, the defect type of the new defect time interval obtained after merging is LF type defect;

[0033] If the defect types corresponding to the defect time intervals with intersections are all LMA type defects, the defect type of the new defect time interval obtained after merging is LMA type defect;

[0034] If the defect types corresponding to the defect time intervals with intersections include LF type defects and LMA type defects, the defect type of the new defect time interval obtained after merging is LMA type defect.

[0035] Preferably, calculating the peak-to-peak value of the TMR sensor within the defect time interval according to the defect type includes:

[0036] If the defect type is LF type defect, calculate the peak-to-peak value of all TMR sensors within the defect time interval;

[0037] If the defect type is LMA type defect, calculate the peak-to-peak value of all TMR sensors with large linear ranges within the defect time interval.

[0038] Preferably, quantitatively calculating the cross-sectional loss rate of the wire rope area with defects based on the peak-to-peak value includes:

[0039] If the defect type is LF type defect, calculate the average value of the peak-to-peak values of all TMR sensors in the large linear range, denoted as the large linear range average value, and at the same time calculate the average value of the peak-to-peak values of all TMR sensors in the small linear range, denoted as the small linear range average value; perform weighted summation on the large linear range average value and the small linear range average value to obtain the cross-sectional loss rate under the LF type defect;

[0040] If the defect type is LMA type defect, calculate the average value of the peak-to-peak values of all TMR sensors in the large linear range, denoted as the large linear range average value; perform weighting on the large linear range average value to obtain the cross-sectional loss rate under the LMA type defect.

[0041] Preferably, the weighted summation of the large linear range average value and the small linear range average value to obtain the cross-sectional loss rate under the LF type defect is calculated as follows:

[0042]

[0043] In the formula, ΔS is the cross-sectional loss rate, is the small linear range average value, k 1 is the first coefficient, is the large linear range average value, k 2 is the second coefficient.

[0044] Preferably, the weighting of the large linear range average value to obtain the cross-sectional loss rate under the LMA type defect is calculated as follows:

[0045]

[0046] In the formula, ΔS is the cross-sectional loss rate, is the large linear range average value, k 3 is the third coefficient, k 4 is the fourth coefficient.

[0047] Second aspect: A wire rope defect detection device based on a multi-specification TMR sensor array, including a multi-specification TMR sensor array and a computer device, where the computer device includes a processor and a memory storing a number of computer instructions, and when the computer instructions are executed by the processor, the steps of the wire rope defect detection method based on the multi-specification TMR sensor array are implemented.

[0048] The wire rope defect detection method and device based on a multi-specification TMR sensor array provided by the present invention have the following beneficial effects compared with the prior art:

[0049] 1. The present invention uses TMRs with different linear ranges, which not only realizes the quantitative calculation of large-area metal cross-sectional loss defects, but also realizes the positioning and quantitative calculation of tiny local defects;

[0050] 2. Compared with other patents for LF-type defects and LMA-type defects, the present invention no longer relies on empirical values for judgment, but uses the differences in waveform characteristics between the two types of defects for judgment, which is more accurate and effective;

[0051] 3. In terms of real-time online detection, the present invention gives full play to the rapid detection ability of non-destructive testing, can quickly locate defects and describe the amount of defect loss, and the defect calculation delay is only 20 ms. Description of the Drawings

[0052] Figure 1 is a flowchart of a wire rope defect detection method based on a multi-specification TMR sensor array of the present invention;

[0053] Figure 2 is a schematic layout diagram of the multi-specification TMR sensor array of the present invention. Detailed Embodiments

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0056] In view of the problems existing in the damage identification of existing wire ropes, this embodiment proposes a wire rope defect detection method based on a multi-specification TMR sensor array. First, collect the signals of the circumferentially uniformly arranged multi-specification TMR sensor arrays; then, perform signal preprocessing such as outlier removal, wavelet denoising, and channel equalization on the original damage signals; then, search for wave peaks and wave valleys in the preprocessed data to obtain the waveform characteristics, waveform intervals, etc. of each defect signal and effectively distinguish the same-strand wave signals and vibration signals; finally, merge the LF-type and LMA-type defect intervals for the searched defect ranges according to different judgment conditions, and perform defect quantitative calculation using different linear coefficients depending on the extracted waveform characteristics.

[0057] Specifically, as Figure 1 shown, the wire rope defect detection method of the multi-specification TMR sensor array in this embodiment includes the following steps:

[0058] Step 1: Acquire detection signals of multi-specification TMR sensor arrays and perform preprocessing.

[0059] For non-destructive testing of wire ropes, TMR sensors are stable and reliable, have low power consumption, good linearity and wide linear range, good anti-interference, and little aging performance degradation. They can work stably under harsh working conditions such as high temperature, oil pollution, and dust. However, due to the limited circumferential coverage of the wire rope by a single TMR sensor, broken wire defects distributed at any position around the wire rope will be missed. Therefore, a TMR sensor array is evenly arranged along the circumference of the wire rope, such as Figure 2 As shown, broken wire defects at different circumferential positions can be detected by the nearest TMR sensor, thereby realizing a 360° scan of the circumference of the wire rope. In view of the large difference in the magnetic induction intensity ranges of large and small defects, this embodiment selects two TMR sensors of different specifications, namely a large linear range TMR sensor with a maximum linear range greater than the linear threshold (for example, the linear threshold is 300 Gauss), and a small linear range TMR sensor with a maximum linear range less than or equal to the linear threshold. The large linear range TMR sensor meets the magnetic induction intensity range of larger metal cross-section loss-type defects, and the small linear range TMR sensor realizes the effective detection of local small defects. The two are arranged at intervals to form a complete circumferential array.

[0060] After obtaining the detection signal of the TMR sensor, the signal preprocessing method is used to remove outliers, perform wavelet denoising and channel equalization on the original detection signal. The complex structure of the wire rope determines the frequency diversity of its broken wire signal. The frequency point fluctuates all the time, and the broken wire damage signal is transient and non-stationary. The traditional single filter is also easy to weaken the broken wire damage signal when removing noise, which can easily reduce the quantitative identification accuracy of the broken wire. In recent years, the wavelet transform has been used to reduce the noise of the broken wire damage signal of the wire rope, and the noise reduction effect is good. Channel equalization is to deal with the inconsistency of different TMR sensors, the performance differences between sensors and the system errors caused by the electrical characteristics of the corresponding signal output circuit. The equalization coefficient can achieve consistency at the software level.

[0061] Step 1.1, obtain the detection signals of the large linear range TMR sensor and the small linear range TMR sensor, regard a single TMR sensor as a single channel, and record the detection signal of the single channel as D i .

[0062] Step 1.2: Use the signal preprocessing method to remove outliers, perform wavelet denoising and channel equalization on the detection signal.

[0063] Step 1.2.1. Outlier rejection includes: for the detection signal of a single TMR sensor, screening out the outliers in the detection signal, and taking the numerical average of the sampling points on both sides of the outlier to replace the outlier. After data collection, due to the influence of the device and the environment itself, some sampling points may be abnormal. For the sensor data D of a single channel i According to the set sampling window size t, the data D of each sampling point is obtained i1 , D i2 , D i3 ,…, D im ,…, D it , where m is a positive integer greater than zero and less than or equal to t, representing the sampling point serial number. Calculate the difference ΔD = D im - D i(m-1) . Calculate all the difference results. If the difference is greater than the abnormal threshold, the corresponding sampling point is regarded as an outlier, and the data D of the corresponding sampling point im Take the numerical average of the two ends of the corresponding channel for replacement D im = (D i(m-1) + D i(m+1) ) / 2, D im is the detection signal at the m-th moment of the i-th channel, D i(m-1) is the detection signal at the (m - 1)-th moment of the i-th channel, D i(m+1) is the detection signal at the (m + 1)-th moment of the i-th channel.

[0064] Step 1.2.2. Wavelet denoising includes: for the detection signal of a single TMR sensor after outlier rejection, use the wavelet threshold denoising method to process. The steel wire rope defect signals are complex and diverse. In addition to outlier rejection, the wavelet analysis method can better grasp the low-frequency time-domain characteristics of the defect signals and is very suitable for the analysis of sudden defect signals. The one-dimensional noisy magnetic flux leakage signal model of the steel wire rope collected by the hardware system is D im = X im + e im , where X im is the true damage signal, and e im is the noise signal (strand wave and high-frequency noise). Using the wavelet threshold denoising method for the noisy signal is an effective method to remove various noises in the magnetic flux leakage signal of the steel wire rope. Determine the best wavelet basis as db8, perform 3-layer decomposition, and filter out the strand wave signal and high-frequency noise e im .

[0065] Step 1.2.3. The channel equalization process includes: obtaining the sensitivity coefficient of each TMR sensor by using the defect surrounding method, and using the sensitivity coefficient to perform weighted correction on the detection signal after wavelet denoising of the corresponding TMR sensor to obtain the preprocessed detection signal. When specifically arranging the TMR sensors, the consistency of each TMR sensor is not completely the same. Therefore, in this embodiment, the circumferential fixed-size defect surrounding method is used for each TMR sensor to obtain the actual sensitivity coefficient k of the TMR sensor, and weighted correction is performed on each channel respectively, and the result is C i = X i k i , where i represents the channel number of the TMR sensor, and k i represents the correction coefficient of the i-th TMR sensor, X i represents the damage signal of the i-th TMR sensor, and C i represents the correction signal of the i-th TMR sensor.

[0066] Step 2. Use a maximum value filter to filter the preprocessed detection signal to obtain the filtered detection signal.

[0067] To achieve the function of online real-time detection, in this embodiment, sensor data is obtained in real-time segmented windows, the sampling point signals of a fixed window segment are matrixized, and peak and valley searches are performed on the matrix data. Considering the problem that defect signals may be truncated in different window segments during the online monitoring process, it is necessary to analyze the truncated waveform characteristics, accurately judge the type of defects, perform comprehensive determination after calculating the window segment data, splice the truncated defects, and perform qualitative and quantitative calculations with complete data. And this quantitative calculation method depends on the conclusion of finite element simulation and has good accuracy after verification by experimental data.

[0068] In this embodiment, a signal matrix is constructed. The size of the signal rectangle is the number of circumferentially arranged TMR sensors multiplied by the axial movement sampling window size. And the detection signals obtained by all TMR sensors within the sampling window are written into the signal matrix. Then a maximum value filter is introduced. It generates a two-dimensional structural element with a connectivity of 1, where the central pixel is connected to its upper, lower, left, and right pixels but not to the diagonal pixels. The signal matrix is filtered using the maximum value filter to obtain the positioning of all peaks and valleys, and these results contain a large number of non-defect signals.

[0069] Step 3. According to the minimum defect analysis method, set the minimum peak threshold of the defect as the defect threshold θ. If there is a peak greater than the defect threshold θ in the filtered detection signal, it is determined that there is a defect in the wire rope area corresponding to the peak, and then the subsequent steps are continued; otherwise, there is no defect, and this detection ends at this time.

[0070] For the case of judging the detection signal without window segmentation, if there are defective peaks in the detection signal of the TMR sensor, the subsequent steps 4 - 6 are executed; otherwise, the defect detection is ended and it is considered that there are no defects in the current wire rope detection area. For the case of judging the detection signal by window segmentation, if there are defective peaks in the current sampling window, the subsequent steps 4 - 6 are executed; if there are no defective peaks in the current sampling window, the judgment of the next sampling window is continued until the judgment of all sampling windows is completed and then ended.

[0071] Step 4: Calculate the difference between the defective peak and the wave valleys on both sides. According to the differences on both sides, judge whether the defect type of the wire rope area corresponding to the defective peak is an LF - type defect or an LMA - type defect.

[0072] In the single - channel detection signal obtained after pre - processing, the value of the detection signal without defects is unique (when the values are the same or there are small fluctuations, one of them is taken as the detection signal without defects). Therefore, in this embodiment, the difference between the defect threshold and the detection signal without defects in the detection signal pre - processed by the TMR sensor is calculated and denoted as the classification difference.

[0073] If the differences on both sides exceed the classification difference, the defect type is an LF - type defect.

[0074] If one of the differences on both sides exceeds the classification difference and the other does not, and the adjacent peaks on both sides of the current peak (i.e., the defective peak being judged currently) have the same rule, the defect type is an LMA - type defect. Here, the same rule means that one of the differences on both sides exceeds the classification difference and the other does not. Since the TMR sensor data is analyzed by sampling windows, the adjacent peaks on both sides of the current peak may appear in the current sampling window or in the next sampling window, that is, splicing analysis may be required between sampling windows.

[0075] Otherwise, the defect type of the current peak is not judged. In other cases, if the differences on both sides do not exceed the classification difference, it is considered that the current peak is caused by abnormal detection, and the defect judgment of the current peak can be ignored, or a warning can be marked for manual intervention judgment. If one of the differences on both sides exceeds the classification difference and the other does not, but the adjacent peaks on both sides of the current peak do not have the same rule, it is considered that the current peak is caused by abnormal detection, and the defect judgment of the current peak can be ignored, or a warning can be marked for manual intervention judgment.

[0076] Step 5: Extract the time indices of the wave valleys on both sides of the defective peak to form the defect time interval corresponding to the defective peak, and merge the defect time intervals with intersections.

[0077] To facilitate the determination of the defect type, in this embodiment, the defect results of multiple channels are merged into intervals, and the merged LF-type defects and LMA-type defects are respectively brought into different quantitative calculation methods. Specifically, if there is an intersection between the defect time intervals of two defects, the union result is output as the new merged defect time interval. For example, the defect time interval of a TMR sensor is [0, 50], and the defect time interval of another TMR sensor is [30, 60], then the new merged defect time interval is [0, 60], where 0, 30, 50, and 60 are all moments within a sampling window.

[0078] Since each defect time interval has a clear defect type before merging, the defect type of the new merged defect time interval is determined as follows:

[0079] If the defect types corresponding to the defect time intervals with an intersection are all LF-type defects, then the defect type of the new merged defect time interval is LF-type defect; if the defect types corresponding to the defect time intervals with an intersection are all LMA-type defects, then the defect type of the new merged defect time interval is LMA-type defect; if the defect types corresponding to the defect time intervals with an intersection include LF-type defects and LMA-type defects, then the defect type of the new merged defect time interval is LMA-type defect.

[0080] Step 6: According to the defect type, calculate the peak-to-peak value of the TMR sensor within the defect time interval, and quantitatively calculate the cross-sectional loss rate of the steel wire rope area with defects based on the peak-to-peak value.

[0081] In this embodiment, the defect time intervals of all different channels are merged to finally obtain a unique defect time interval and the corresponding defect type. In other embodiments, one or more defect time intervals can also be retained according to the arrangement positions of different TMR sensors. For example, the defect time intervals of the TMR sensors above the steel wire rope are merged, and the defect time intervals of the TMR sensors below the steel wire rope are merged.

[0082] (1) The calculation process for LF-type defects is as follows:

[0083] First, for the case where the LF-type defects do not exceed the range of the TMR sensor in the small linear range, calculate the peak-to-peak value B of all TMR sensors in the small linear range PP1 and the peak-to-peak value B of the TMR sensor in the large linear range PP2 .

[0084] Then, calculate the average value of the peak-to-peak values of all TMR sensors in the small linear range Denoted as the average value of the large linear range. Meanwhile, calculate the average value of the peak-to-peak values of all TMR sensors in the large linear range. Denoted as the average value of the small linear range, where avg represents the averaging operation.

[0085] Finally, perform a weighted sum of the average value of the large linear range and the average value of the small linear range to obtain the cross-sectional loss rate under LF-type defects:

[0086]

[0087] In the formula, ΔS is the cross-sectional loss rate, is the average value of the small linear range, and k 1 is the first coefficient, is the average value of the large linear range, and k 2 is the second coefficient.

[0088] (2) The calculation process for LMA-type defects is as follows:

[0089] Firstly, for the case where the LMA-type defect may exceed the measurement range of the small linear range TMR sensor, only calculate the peak-to-peak value B of all TMR sensors in the large linear range PP2 .

[0090] Then calculate the average value of the peak-to-peak values of all TMR sensors in the large linear range Denoted as the average value of the large linear range.

[0091] Finally, perform weighting on the average value of the large linear range to obtain the cross-sectional loss rate under LMA-type defects:

[0092]

[0093] In the formula, ΔS is the cross-sectional loss rate, is the average value of the large linear range, and k 3 is the third coefficient, and k 4 is the fourth coefficient. Among them, k 1 , k 2 , k 3 and k 4 can be obtained by calibration or fitting.

[0094] The present invention solves the problem of difficult high-precision detection of both local micro-damage and large-area metal cross-sectional damage. Meanwhile, defects within the circumferential range of the steel wire rope are no longer limited by the complex strand structure and are difficult to detect, making the detection range of the current system have good robustness for defects of different sizes and distributions. Additionally, at the algorithm level, the real-time calculation of the magnetic flux leakage signal and the judgment of the defect type realize the function of non-destructive online detection of the steel wire rope, enhancing the superiority of the product.

[0095] In another embodiment, the present invention further provides a wire rope defect detection device based on a multi-specification TMR sensor array, including a multi-specification TMR sensor array and a computer device. The computer device includes a processor and a memory storing a number of computer instructions. When the computer instructions are executed by the processor, the steps of the wire rope defect detection method based on the multi-specification TMR sensor array are implemented.

[0096] For the specific limitations of the wire rope defect detection device based on the multi-specification TMR sensor array, reference can be made to the limitations of the wire rope defect detection method based on the multi-specification TMR sensor array in the above text, which will not be elaborated here.

[0097] The memory and the processor are directly or indirectly electrically connected to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. The memory stores computer instructions that can run on the processor, and the processor realizes the method of the present invention by running the computer instructions stored in the memory.

[0098] Among them, the memory can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0099] The processor can be an integrated circuit chip with data processing capabilities. The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0100] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0101] The above-described embodiments merely represent several implementation manners of the present invention. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the appended claims.

Claims

1. A wire rope defect detection method based on a multi-specification TMR sensor array, characterized in that: The wire rope defect detection method based on a multi-specification TMR sensor array comprises: Acquire detection signals of a multi-specification TMR sensor array and perform preprocessing, wherein the multi-specification TMR sensor array is evenly arranged along the circumference of the wire rope, and TMR sensors of different specifications in the multi-specification TMR sensor array are alternately arranged, and the TMR sensors of different specifications include a large linear range TMR sensor whose maximum linear range value is greater than a linear threshold value, and a small linear range TMR sensor whose maximum linear range value is less than or equal to the linear threshold value; A maximum filter is used to filter the preprocessed detection signal to obtain a filtered detection signal; If there is a peak greater than the defect threshold in the filtered detection signal, it is determined that there is a defect in the wire rope area corresponding to the peak; otherwise, there is no defect; Calculate the difference between the defective peak and the troughs on both sides, and determine whether the defect type of the wire rope area corresponding to the defective peak is a LF defect or an LMA defect based on the difference on both sides; Extract the time indexes of the troughs on both sides of the defective peak to form the defective time interval corresponding to the defective peak, and merge the defective time intervals with intersections; According to the defect type, the peak-to-peak value of the TMR sensor within the defect time interval is calculated, and the cross-sectional loss rate of the defective wire rope area is quantitatively calculated based on the peak-to-peak value.

2. The wire rope defect detection method based on a multi-specification TMR sensor array according to claim 1 is characterized in that: The preprocessing includes outlier point removal, wavelet denoising and channel equalization processing: The outlier point elimination includes: for the detection signal of a single TMR sensor, screening outlier points in the detection signal, and taking the numerical average of the sampling points on both sides of the outlier point to replace the outlier point; The wavelet denoising comprises: processing the detection signal of a single TMR sensor after outlier points are removed by using a wavelet threshold denoising method; The channel equalization processing includes: using the defect surround method to obtain the sensitivity coefficient of each TMR sensor, and using the sensitivity coefficient to perform weighted correction on the detection signal of the corresponding TMR sensor after wavelet denoising to obtain the pre-processed detection signal.

3. The wire rope defect detection method based on a multi-specification TMR sensor array according to claim 1 is characterized in that: The method of judging, based on the difference between the two sides, whether the defect type of the wire rope region corresponding to the defective wave peak is a LF defect or a LMA defect includes: Calculate the difference between the defect threshold and the detection signal when there is no defect in the detection signal preprocessed by the TMR sensor, and record it as the classification difference; If the difference on both sides exceeds the classification difference, the defect type is LF type defect; If one side of the difference exceeds the classification difference and the other side does not exceed the classification difference, and the adjacent peaks on both sides of the current peak have the same pattern, the defect type is an LMA defect; Otherwise, the defect type of the current peak is not determined.

4. The wire rope defect detection method based on a multi-specification TMR sensor array according to claim 1 is characterized in that: The defect time intervals with overlapping merged elements include: If the defect types corresponding to the defect time intervals with intersection are all LF type defects, the defect type of the new defect time interval obtained after merging is LF type defect; If the defect types corresponding to the defect time intervals with intersections are all LMA type defects, the defect type of the new defect time interval obtained after merging is LMA type defect; If the defect types corresponding to the defect time intervals with an intersection include LF type defects and LMA type defects, the defect type of the new defect time interval obtained after the merging is LMA type defects.

5. The wire rope defect detection method based on a multi-specification TMR sensor array according to claim 1 is characterized in that: Calculating the peak-to-peak value of the TMR sensor within the defect time interval according to the defect type includes: If the defect type is LF type defect, calculate the peak-to-peak value of all TMR sensors in the defect time interval; If the defect type is an LMA defect, calculate the peak-to-peak values ​​of all large linear range TMR sensors within the defect time interval.

6. The wire rope defect detection method based on a multi-specification TMR sensor array according to claim 5 is characterized in that: The peak-to-peak value-based quantitative calculation of the cross-sectional loss rate of the defective wire rope region includes: If the defect type is an LF defect, the average value of the peak-to-peak values ​​of all TMR sensors in the large linear range is calculated, recorded as the large linear range average value, and the average value of the peak-to-peak values ​​of all TMR sensors in the small linear range is calculated, recorded as the small linear range average value; the large linear range average value and the small linear range average value are weighted summed to obtain the cross-sectional loss rate under the LF defect; If the defect type is an LMA defect, the average of the peak-to-peak values ​​of all large linear range TMR sensors is calculated and recorded as the large linear range average value; the large linear range average value is weighted to obtain the cross-sectional loss rate under the LMA defect.

7. The wire rope defect detection method based on a multi-specification TMR sensor array according to claim 6 is characterized in that: The weighted sum of the average value of the large linear range and the average value of the small linear range is performed to obtain the cross-sectional loss rate under the LF type defect, which is calculated as follows: Where ΔS is the cross-sectional loss rate, is the average value of the small linear range, k1 is the first coefficient, is the average value of the large linear range, and k2 is the second coefficient.

8. The wire rope defect detection method based on a multi-specification TMR sensor array according to claim 6, characterized in that: The large linear range average is weighted to obtain the cross-sectional loss rate under LMA type defects, which is calculated as follows: Where ΔS is the cross-sectional loss rate, is the average value of the large linear range, k3 is the third coefficient, and k4 is the fourth coefficient.

9. A wire rope defect detection device based on a multi-specification TMR sensor array, characterized in that: It comprises a multi-specification TMR sensor array and a computer device, wherein the computer device comprises a processor and a memory storing a plurality of computer instructions, and when the computer instructions are executed by the processor, the steps of the wire rope defect detection method based on the multi-specification TMR sensor array as described in any one of claims 1 to 8 are implemented.

Images