A method for estimating the number of broken wires of a steel wire rope considering the effect of lift-off value

Abstract

The application discloses a steel wire rope broken wire root number estimation method considering lift-off effect, collects a magnetic flux leakage signal of a steel wire rope surface, and removes a trend of the magnetic flux leakage signal by using multidimensional variation modal decomposition; then, a time-frequency spectrum is obtained by using synchronous extrusion transformation of wavelet transformation, a greedy algorithm with forward and backward windowing is used to extract a stock wave time-frequency ridge line, and a corresponding stock wave instantaneous frequency curve is obtained; then, the size of the lift-off value is judged through the stock wave instantaneous frequency, and a broken wire magnetic flux leakage field distribution is inverted; finally, the number of broken wires of the steel wire rope is estimated through a magnetic field radial component.

Description

Technical Field

[0001] This invention belongs to the field of nondestructive testing technology for steel wire ropes, and more specifically, it relates to a method for estimating the number of broken wires in a steel wire rope that takes into account the lift-off effect. Background Technology

[0002] Due to the unique nature of wire rope operations, the importance of non-destructive testing (NDT) for production safety in their application scenarios is self-evident. Wire rope damage can be categorized into two types: Local Fault (LF) and Loss of Metallic Area (LMA). LF damage mainly includes broken wires, pitting, and localized shape deformation, while LMA damage mainly includes wear, corrosion, and reduction in wire rope diameter.

[0003] Currently, magnetic flux leakage (MF) testing is widely used in the detection of broken wires in steel wire ropes. However, due to the vibration of the steel wire rope in actual working conditions, the change in the lift-off value between its surface and the sensor affects the distribution of the magnetic flux leakage field, thus affecting the quantitative analysis of the number of broken wires in the steel wire rope. Therefore, eliminating the interference of lift-off value changes during the MF testing of steel wire ropes is of great engineering significance for the non-destructive testing and quantitative analysis of steel wire ropes. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for estimating the number of broken wires in a wire rope that considers the lift-off value effect. This method effectively eliminates the lift-off value variation caused by factors such as wire rope vibration, improves the accuracy of wire rope broken wire estimation, and clarifies the damage state of the wire rope.

[0005] To achieve the above-mentioned objective, the present invention provides a method for estimating the number of broken wires in a steel wire rope considering the lift-off effect, characterized by comprising the following steps:

[0006] (1) Collect the leakage magnetic signal of the steel wire rope;

[0007] The leakage magnetic field signal of a saturated magnetized steel wire rope is detected using a Y-channel Hall sensor. The leakage magnetic field signal on the surface of the steel wire rope is collected at an equal-time sampling rate to obtain the leakage magnetic field signal of Y-channel length N, denoted as S(t)=[S1(t),S2(t),…,S…]. y (t),…,S Y [(t)],S y (t) represents the leakage magnetic signal of the y-th channel, where y = 1, 2, ..., Y;

[0008] (2) Detrending processing of leakage magnetic signal;

[0009] Using variational mode decomposition algorithm for each S path yThe mode decomposition of (t) is performed. After decomposition, the first component is removed, and the other components are summed to obtain the de-stressed leakage magnetic signal Z for each path. y (t);

[0010] After detrending, we obtain Z(t) = [Z1(t), Z2(t), ..., Z y (t),…,Z Y (t)];

[0011] (3) Use spline interpolation to convert Z(t) into a magnetic flux leakage image with a resolution of M×N;

[0012] The Y-channel leakage magnetic flux signal Z(t) is interpolated using spline interpolation to form M-channel leakage magnetic flux signals Z(t) = [Z1(t), Z2(t), Z3(t)...Z M [(t)], convert Z(t) into a magnetic flux leakage image Z(m,n) with a resolution of M×N, where m=1,2,3…M,n=1,2,3…N;

[0013] (4) Extract the instantaneous frequency of the stock wave;

[0014] (4.1) Using the wavelet synchronous compression transform method, first process a certain signal Z in Z(t) m (t) is subjected to wavelet transform to obtain the time-frequency matrix WT f (a,b), m=1,2,…,M, where a is the scale factor and b is the time factor;

[0015] For any (a, b), when WF f When (a,b)≠0, the instantaneous frequency of signal Z(t) can be expressed as: Where i is the complex unit;

[0016] Discretize the scale factor a as a k k = 1, 2, ...; Discretize the frequency ω to represent ω l , l = 1, 2…;

[0017] [ω] l -1 / 2Δω,ω l The wavelet coefficients within the interval [+1 / 2Δω] are superimposed to the center frequency ω. l As the instantaneous frequency, Δω is the frequency interval after linear discretization of frequency ω, thus obtaining the time-frequency amplitude spectrum of each wave signal. Among them, Δa=a k -a k-1 a k Is it satisfied by |ω(a) k ,b)-ω l All scale factors ≤ Δω / 2;

[0018] (4.2) On the obtained time spectrum, a greedy algorithm with forward and backward windowing is used to search for and extract the ridge with the largest energy on the amplitude spectrum as the instantaneous frequency curve of the stock wave.

[0019] (4.3) Extract the instantaneous frequency amplitude of the stock wave from the leakage magnetic signals of the M channels respectively, and obtain the instantaneous frequency amplitude matrix IA(m,n);

[0020] (5) Lift-off value measurement;

[0021] (5.1) For the instantaneous frequency amplitude matrix IA(m,n) of the stock wave, the nonlinear least squares method is used according to IA(m,n)=c·f(e,β,α0,α) m The fit is performed using )+ε, where IA(m,n) is the instantaneous amplitude of the stock wave 1*M at time n, c is the positional relationship constant, and f(e,β,α0,α) is the constant value of the positional relationship. m α represents the relative position angle of the sensor. m The function, where e is the eccentricity distance, β is the eccentricity angle, and α is the eccentricity angle. m Let ε be the angle between the m-th sensor and the first sensor, and let ε be the error vector.

[0022] (5.2) Given an initial value for each parameter, obtain the corresponding fitting curve according to the fitting formula in (5.1), and then calculate the mean square error as follows: Among them, w m The weight values ​​fitted for each channel;

[0023] (5.3) The mean square error is minimized using the trust region algorithm, and the estimated values ​​of each parameter in the fitting formula are calculated iteratively. These values ​​are then substituted into the geometric relationship of the lift-off value (r+h). m ) 2 =e 2 +R 2 -2eRcos(β-α m -α0), to obtain the lift-off value h of the m-th channel. m Then, the lift-off value matrix HH(m,n) of each channel is obtained, where r is the diameter of the wire rope and R is the distance from the center of the wire rope to the Hall sensor;

[0024] (6) Inversion of broken wire signal and estimation of the number of broken wires;

[0025] (6.1) Based on the lift-off value matrix HH(m,n) of each channel, and combined with the magnetic flux leakage image of the broken wire detection, the lift-off value h of the channel where the broken wire is located is obtained;

[0026] (6.2) Based on the magnetic dipole model, the radial component distribution of the magnetic field at the broken wire signal is as follows: C is the proportionality coefficient, and w is the width of the broken wire;

[0027] Given the lift-off value h and the wire breakage signal LF(q), where q = 1, 2, 3…N q N q Given the length of the broken wire signal, the radial component B of the leakage magnetic field after eliminating the lift-off value can be solved by fitting the magnetic dipole model using the least squares method. y ;

[0028] (6.3) The radial component B of the leakage magnetic field after eliminating the lift-off value at the broken wire location. y Substituting the magnetic dipole model Where, ρ ms Let μ be the magnetic surface charge density, μ0 be the relative permeability, and (x,y) be the magnetic flux density. y Coordinates of the measuring point;

[0029] (6.4) By solving the magnetic dipole model, the wire breakage depth d is obtained, and then the number of broken wires is calculated. D is the diameter of each wire in the wire rope.

[0030] The objective of this invention is achieved as follows:

[0031] This invention discloses a method for estimating the number of broken wires in a steel wire rope considering the lift-off effect. The method involves acquiring the leakage magnetic field signal from the surface of the steel wire rope and detrending the signal using multidimensional variational mode decomposition. Next, a synchronous squeezing transform of wavelet transform is used to obtain its time-frequency spectrum. Then, a greedy algorithm with forward and backward windowing is used to extract the time-frequency ridge of the strand wave, yielding the corresponding instantaneous frequency curve of the strand wave. The lift-off value is then determined based on the instantaneous frequency of the strand wave, and the distribution of the leakage magnetic field due to broken wires is inverted. Finally, the number of broken wires in the steel wire rope is estimated using the radial component of the magnetic field.

[0032] Meanwhile, the method for estimating the number of broken wires in a steel wire rope that considers the lift-off effect, as proposed in this invention, also has the following beneficial effects:

[0033] (1) The present invention can extract the lift-off value change information by the magnetic leakage signal on the surface of the wire rope, and determine the relative position of the wire rope in the detection.

[0034] (2) By judging the change in the lifting value of the wire rope, the present invention can more accurately reflect the magnetic leakage information on the surface of the wire rope and determine the number of broken wires in the wire rope. Attached Figure Description

[0035] Figure 1 This is a flowchart of a method for estimating the number of broken wires in a steel wire rope that considers the lift-off effect, according to the present invention.

[0036] Figure 2 This is a schematic diagram showing the relative positions of the Hall sensor and the steel wire rope under eccentric conditions;

[0037] Figure 3 It consists of the original leakage magnetic signal and mode decomposition components;

[0038] Figure 4 These are magnetic flux leakage images before and after detrending processing;

[0039] Figure 5 It is the process of extracting the instantaneous frequency of stock waves;

[0040] Figure 6 These are the extraction results of the lift-off values ​​at different time points;

[0041] Figure 7 These are the results of inversion from two broken wires. Detailed Implementation

[0042] The specific embodiments of the present invention will now be described with reference to the accompanying drawings to enable those skilled in the art to better understand the invention. It should be particularly noted that in the following description, detailed descriptions of known functions and designs that might obscure the main content of the invention will be omitted here.

[0043] Example

[0044] Figure 1 This is a flowchart of a method for estimating the number of broken wires in a steel wire rope that considers the lift-off effect, according to the present invention.

[0045] In this embodiment, as Figure 1 As shown, the present invention provides a method for estimating the number of broken wires in a steel wire rope considering the lift-off effect, comprising the following steps:

[0046] S1. Collect the leakage magnetic signal of the steel wire rope;

[0047] The leakage magnetic field signal of a saturated magnetized steel wire rope is detected using a Y-channel Hall sensor. The leakage magnetic field signal on the surface of the steel wire rope is collected at an equal-time sampling rate to obtain the leakage magnetic field signal of Y-channel length N, denoted as S(t)=[S1(t),S2(t),…,S…]. y (t),…,S Y [(t)],S y (t) represents the leakage magnetic signal of the y-th channel, y = 1, 2, ..., Y. In this embodiment, Y = 16. The relative positions of the steel wire rope and the Hall sensor array are as follows: Figure 2 As shown, when vibrations or other phenomena occur, the wire rope will deviate from the center position of the Hall sensor array.

[0048] S2, Detrending processing of leakage magnetic signal;

[0049] Using variational mode decomposition algorithm for each S path y The mode decomposition of (t) is performed. After decomposition, the first component is removed, and the other components are summed to obtain the de-stressed leakage magnetic signal Z for each path. y (t);

[0050] After detrending, we obtain Z(t) = [Z1(t), Z2(t), ..., Z y (t),…,Z Y (t)];

[0051] S3. Use spline interpolation to convert Z(t) into a magnetic flux leakage image with a resolution of M×N;

[0052] The Y-channel leakage magnetic flux signal Z(t) is interpolated using spline interpolation to form M-channel leakage magnetic flux signals Z(t) = [Z1(t), Z2(t), Z3(t)...Z M [(t)], convert Z(t) into a magnetic flux leakage image Z(m,n) with a resolution of M×N, where m=1,2,3…M,n=1,2,3…N;

[0053] In this embodiment, the variational mode decomposition algorithm uses a mode decomposition number of 4. For example... Figure 3 (a) Leakage magnetic signal S of a single channel Y (t), the sub-mode signals decomposed from it are arranged in ascending order of center frequency. Due to the slowly varying characteristics of the trend term signal, the signal energy is large and the frequency is low. The first sub-signal obtained from the decomposition is the trend term, such as... Figure 3 (b), Figure 3 (c) and 3(d) are mainly stock wave signal components, with the remaining high-frequency signals such as Figure 3 As shown in (e). After spline interpolation processing of the leakage magnetic signal Z(t), as shown... Figure 4 As shown in (a), the two-dimensional magnetic flux leakage image Z(t)=[Z1(t),Z2(t),Z3(t),…,Z M [t] shows the two-dimensional image Z(m,n) formed after removing the trend component, such as Figure 4 (b) clearly eliminates trend signals with different amplitudes, which is beneficial for subsequent signal processing.

[0054] S4. Extract the instantaneous frequency of the stock wave;

[0055] S4.1. Using a wavelet synchronous compression transform method, first process a certain signal Z in Z(t) m (t) is subjected to wavelet transform to obtain the time-frequency matrix WT f (a,b), m=1,2,…,M, where a is the scale factor and b is the time factor;

[0056] For any (a, b), when WF f When (a,b)≠0, the instantaneous frequency of signal Z(t) can be expressed as: Where i is the complex unit;

[0057] Discretize the scale factor a as a kk = 1, 2, ...; Discretize the frequency ω to represent ω l , l = 1, 2…;

[0058] [ω] l -1 / 2Δω,ω l The wavelet coefficients within the interval [+1 / 2Δω] are superimposed to the center frequency ω. l As the instantaneous frequency, Δω is the frequency interval after linear discretization of frequency ω, thus obtaining the time-frequency amplitude spectrum of each wave signal. Among them, Δa=a k -a k-1 a k Is it satisfied by |ω(a) k ,b)-ω l All scale factors ≤ Δω / 2;

[0059] S4.2 On the obtained time spectrum, use a greedy algorithm with forward and backward windowing to search for and extract the ridge line with the largest energy on the amplitude spectrum as the instantaneous frequency curve of the stock wave.

[0060] In this embodiment, taking one of the channel signals as an example, Figure 5 (a) is the original leakage magnetic signal to be extracted. Figure 5 (b) is the time-frequency diagram of synchronous squeezing transform based on wavelet transform. Figure 5 (c) is the final extracted instantaneous frequency curve of the stock wave.

[0061] S4.3 Extract the instantaneous frequency amplitude of the stock wave from the leakage magnetic signals of the M channels respectively, and obtain the instantaneous frequency amplitude matrix IA(m,n) of the stock wave;

[0062] S5. Lift-off value measurement;

[0063] S5.1 For the instantaneous frequency amplitude matrix IA(m,n) of the stock wave, the nonlinear least squares method is used according to IA(m,n)=c·f(e,β,α0,α) m The fit is performed using )+ε, where IA(m,n) is the instantaneous amplitude of the stock wave 1*M at time n, c is the positional relationship constant, and f(e,β,α0,α) is the constant value of the positional relationship. m α represents the relative position angle of the sensor. m The function, where e is the eccentricity distance, β is the eccentricity angle, and α is the eccentricity angle. m Let ε be the angle between the m-th sensor and the first sensor, and let ε be the error vector.

[0064] S5.2. Given initial values ​​for each parameter, obtain the corresponding fitted curve according to the fitting formula in S5.1, and then calculate the mean squared error as follows: Among them, w m The weight values ​​fitted for each channel;

[0065] S5.3. Utilize the trust region algorithm to minimize the mean squared error iteratively, calculate the estimated values ​​of each parameter in the fitting formula, and substitute them into the geometric relationship of the lift-off value (r+h). m ) 2 =e 2 +R 2 -2eRcos(β-α m -α0), to obtain the lift-off value h of the m-th channel. m Then, the lift-off value matrix HH(m,n) of each channel is obtained, where r is the diameter of the wire rope and R is the distance from the center of the wire rope to the Hall sensor;

[0066] In this embodiment, the result of the eccentricity distance parameter extraction is as follows: Figure 6 As shown, where, Figure 6 (a) and (b) are the instantaneous lift-off values ​​of each channel measured at 0.78s and 1.95s, respectively. It can be observed that the jitter causes the wire rope to deviate from the center, and the Hall sensor's relative lift-off value changes.

[0067] S6. Inversion of broken wire signal and estimation of the number of broken wires;

[0068] S6.1. Based on the lift-off value matrix HH(m,n) of each channel and combined with the magnetic flux leakage image of the broken wire detection, the lift-off value h of the channel where the broken wire is located is obtained.

[0069] S6.2, Based on the magnetic dipole model, the radial component distribution of the magnetic field at the broken wire signal is as follows: C is the proportionality coefficient, and w is the width of the broken wire;

[0070] Given the lift-off value h and the wire breakage signal LF(q), where q = 1, 2, 3…N q N q Given the length of the broken wire signal, the radial component B of the leakage magnetic field after eliminating the lift-off value can be solved by fitting the magnetic dipole model using the least squares method. y ;

[0071] S6.3, Eliminate the radial component B of the leakage magnetic field after removing the lift-off value at the broken wire location. y Substituting the magnetic dipole model Where, ρ ms Let μ be the magnetic surface charge density, μ0 be the relative permeability, and (x,y) be the magnetic flux density. y Coordinates of the measuring point;

[0072] S6.4. By solving the magnetic dipole model, the wire breakage depth d is obtained, and then the number of broken wires is calculated. D is the diameter of each wire in the wire rope.

[0073] In this embodiment, after considering the lift-off effect, the inversion results of the two broken wire signals are as follows: Figure 7As shown. The amplitude of the original broken wire signal 1 is 5.15mV, and the amplitude of the original broken wire signal 2 is 2.62mV. The instantaneous lift-off value measured for broken wire 1 is 6.31mm, and the instantaneous lift-off value measured for broken wire 2 is 5.90mm. In the experiment, the lift-off value of the channel where the two broken wires are located remained at 6.15mm, and the relative errors of the lift-off values ​​measured at the two broken wires were 2.60% and 4.07%, respectively. After inversion, the amplitude of the broken wire signals increased, with the amplitude of broken wire signal 1 being 9.59mV and the amplitude of broken wire signal 2 being 3.97mV. By substituting the amplitude of the broken wire signals considering the lift-off effect into the magnetic dipole model, the estimated number of broken wires 1 in this embodiment is 4.67, and the estimated number of broken wires 2 is 2.55. The actual number of broken wires are 4.5 and 2.5, respectively, with corresponding relative errors of 3.78% and 2.00%, respectively. This effectively determines the number of broken wires and realizes the estimation of the number of broken wires in the wire rope.

[0074] Although the illustrative specific embodiments of the present invention have been described above to enable those skilled in the art to understand the invention, it should be understood that the invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the invention as defined and determined by the appended claims, and all inventions utilizing the concept of the present invention are protected.

Claims

1. A method of estimating the number of broken wires of a steel wire rope considering the effect of lift-off value, characterized by, Comprising the following steps: (1) Collecting the magnetic flux leakage signal of the steel wire rope; The leakage magnetic field signal of a saturated magnetized steel wire rope is detected using a Y-channel Hall sensor. The leakage magnetic field signal on the surface of the steel wire rope is collected at an equal-time sampling rate to obtain the leakage magnetic field signal of Y-channel length N, denoted as S(t)=[S1(t),S2(t),…,S…]. y (t),…,S Y [(t)],S y (t) represents the leakage magnetic signal of the y-th channel, where y = 1, 2, ..., Y; (2) De-trend processing of the magnetic flux leakage signal; The mode decomposition is performed on each of the S y (t) by using a variational mode decomposition algorithm, and after the decomposition, the first component is removed, and the sum of the other components is obtained as the de-trended magnetic flux leakage signal Z y (t) of each channel. After detrending, we get Z(t) = [Z1(t), Z2(t),..., Z y (t),..., Z Y (t)]; (3) Converting Z(t) into a magnetic flux leakage image with resolution of MxN by using spline interpolation method; The Y-path magnetic flux leakage signals Z(t) are interpolated by using a spline interpolation method to form M-path magnetic flux leakage signals Z(t) = [Z1(t), Z2(t), Z3(t)…Z M (t)] and convert Z(t) into a magnetic flux leakage image Z(m, n) with a resolution of M x N, m = 1, 2, 3…M, n = 1, 2, 3…N. (4) Extracting the Stockwell transient frequency; (4.1), using wavelet synchronous squeezing transform method, first wavelet transform of a signal Z m (t) in Z(t) to get time-frequency matrix WT f (a, b), m = 1, 2, …, M, a is the scale factor, and b is the time factor. For any (a, b), the instantaneous frequency of the signal Z(t) can be expressed as f (a, b)≠0 where i is the imaginary unit. The scale factor a is discretely represented as a k , k = 1, 2,...; the frequency ω is discretely represented as ω l , l = 1, 2,...; The wavelet coefficients in the interval [ω l -1 / 2Δω,ω l +1 / 2Δω] are superimposed to the center frequency ω l As the instantaneous frequency, Δω is the frequency interval after linear discretization of the frequency ω, so as to obtain the time-frequency amplitude spectrum of each wavelet signal Where Δa=a k -a k-1 , a k is all scale factors satisfying |ω(a k ,b)-ω l |≤Δω / 2; (4.2) On the obtained time-frequency spectrum, using the greedy algorithm with forward and backward windowing to search and extract the ridge line with the maximum energy on the amplitude spectrum as the Stockwell transient frequency curve; (4.3) Extracting the Stockwell transient frequency amplitude of the magnetic flux leakage signal of each channel respectively to obtain the Stockwell transient frequency amplitude matrix IA(m,n); (5) Measuring the lift-off value; (5.1) For the stock wave instantaneous frequency amplitude matrix IA(m, n), a nonlinear least squares method is used to fit according to IA(m, n) = c f(e, b, a0, a m )+e, where IA(m, n) is the stock wave instantaneous amplitude of 1*M at time n, c is a position relationship constant, f(e, b, a0, a m ) is a function of the sensor relative position angle a m , e is the eccentric distance, b is the eccentric angle, a m is the angle between the mth sensor and the first sensor, and e is the error vector. (5.2) Given the initial value of each parameter, the corresponding fitting curve is obtained according to the fitting formula of (5.1), and then the mean square error is calculated as follows: where w m is the weight value fitted for each channel; (5.3) The estimated values of each parameter in the fitting formula are calculated iteratively by minimizing the mean square error using the confidence region algorithm, which are brought into the geometric relationship of the lift-off value (r + h m ) 2 = e 2 + R 2 - 2eRcos(β - α m - α0), to obtain the mth channel lift-off value h m , and further obtain the lift-off value matrix HH(m, n) of each channel, where r is the diameter of the steel wire rope, and R is the distance from the center of the steel wire rope to the Hall sensor. (6) Broken wire signal inversion and broken wire number estimation; (6.1) According to the lift-off value matrix HH(m,n) of each channel, combined with the broken wire detection magnetic flux leakage image, the lift-off value h of the broken wire at the channel is obtained; (6.2), the distribution of the radial component of the magnetic field at the broken wire signal is inverted according to the magnetic dipole model C is a proportional coefficient, and w is the broken wire width. With the known lift-off value h and the wire breakage signal LF(q), q = 1, 2, 3…N q , N q is the length of the wire breakage signal, the least square method is used to fit the magnetic dipole model, and the radial component B y of the magnetic flux leakage field after eliminating the lift-off value can be solved. (6.3) Eliminate the radial component of the leakage magnetic field B after the wire breakage y Bringing in the magnetic dipole model where ρ ms is the surface density of magnetic charge, μ0 is the relative permeability, (x, y) is the coordinate of the measuring point B y ; (6.4) The broken wire depth d is obtained by solving the magnetic dipole model, and the number of broken wires is further obtained D is the diameter of each steel wire of the steel wire rope.

Images