CN114295084B - Adaptive TOF calculation method based on improved skip tongue line function LMS algorithm and thickness measurement technology adopting method - Google Patents
Adaptive TOF calculation method based on improved skip tongue line function LMS algorithm and thickness measurement technology adopting method Download PDFInfo
- Publication number
- CN114295084B CN114295084B CN202111631261.XA CN202111631261A CN114295084B CN 114295084 B CN114295084 B CN 114295084B CN 202111631261 A CN202111631261 A CN 202111631261A CN 114295084 B CN114295084 B CN 114295084B
- Authority
- CN
- China
- Prior art keywords
- skip
- tof
- adaptive
- tongue
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Length Measuring Devices Characterised By Use Of Acoustic Means (AREA)
Abstract
The invention discloses a self-adaptive TOF calculation method based on an improved skip tongue function LMS algorithm and a thickness measuring technology adopting the method, wherein the calculation method is characterized in that ultrasonic echo signals are collected, an improved skip tongue function is combined with the LMS algorithm to obtain the self-adaptive TOF calculation method based on the improved skip tongue function LMS algorithm, an expected signal is estimated and obtained by calculating an autocorrelation mean value of an input signal, and then a step length adjusting factor and a waveform control coefficient of the skip tongue function are improved by utilizing the autocorrelation mean value, so that the convergence speed of weights is accelerated, the performance of the algorithm is improved, and finally an impulse response sequence is solved by utilizing the iteration of the self-adaptive algorithm to calculate the self-adaptive TOF; the thickness of the measured piece can be calculated by combining the sound velocity of the measured piece. The self-adaptive TOF calculation method has stable performance, simple calculation and high processing speed, and can obviously improve the accuracy and efficiency of ultrasonic thickness measurement when being used in the calculation of ultrasonic thickness measurement.
Description
Technical Field
The invention relates to a TOF calculation method, in particular to a self-adaptive TOF calculation method based on an improved skip tongue function LMS algorithm and a thickness measurement technology adopting the method.
Background
The crushing process is an important process which is indispensable in mineral processing, the lining plate is a key part bearing load in crushing, the performance and the service life of the lining plate directly influence the working efficiency and the production cost of the crusher, the abrasion state of the lining plate is known in time, the problem that the crusher breaks due to the lining plate to cause operation failure can be avoided, the ore breaking quality of the crusher is improved, and the economic loss is reduced for enterprises.
The ultrasonic thickness measurement detection technology has great development in the middle of the 20 th century, has the advantages of low cost, high precision, large detection range and the like, and is favored in various fields. The ultrasonic thickness measurement technology is used for detecting the abrasion condition of the lining plate of the crusher, so that the measurement efficiency and the measurement precision can be improved, the use cost of crushing equipment can be saved, and conditions are created for timely replacement of the lining plate.
Ultrasonic thickness measurement is based on the principle of ultrasonic pulse reflection, in which when an ultrasonic pulse emitted by a probe reaches the interface of a material through the object to be measured, the pulse is reflected back to the probe, and the thickness of the material to be measured is determined by precisely measuring the ultrasonic sound (TOF) of the ultrasonic wave propagating in the material. According to the ultrasonic thickness measurement principle: the thickness of the measured workpiece is equal to the ultrasonic sound velocity multiplied by TOF, so the thickness measuring accuracy depends on the sound velocity and TOF, in actual ultrasonic thickness measurement, the ultrasonic propagation speed is basically unchanged in the measured workpiece and can be obtained by referring to the related data calibration of the measured workpiece, and therefore, the accurate calculation of the TOF is the key of ultrasonic thickness measurement.
The patents related to the technology mainly comprise: a method (CN 201310476895.1) for improving ultrasonic thickness measurement precision by utilizing curve fitting is disclosed, which is characterized in that sampling precision is compensated by utilizing a curve fitting algorithm, so that the ultrasonic thickness measurement precision is improved. The method utilizes ultrasonic signals to approximate a straight line in a small time period above and below a zero point, the coordinate of a characteristic point on a fitting curve is calculated through fitting the straight line, and the coordinate of the characteristic point is sent to a thickness measuring module for thickness measurement. The method can improve the ultrasonic thickness measurement precision under the condition that the sampling clock frequency is unchanged, but the thickness measurement precision is not obviously improved under other scenes.
An electromagnetic ultrasonic sound time measurement method and system (CN 202110431187.0) is disclosed, wherein the time interval of two groups of adjacent echo signals in a truncated signal is calculated to be used as the electromagnetic ultrasonic sound time measurement method of sound time through the rising edge and the falling edge in the truncated signal. The method utilizes the discrete threshold voltage to discretize the received signal, reduces the difficulty of signal processing and saves the cost, but has complex processing flow and longer processing time.
An electromagnetic ultrasonic thickness measuring method (CN 201910152431.2) based on frequency domain analysis discloses an electromagnetic ultrasonic thickness measuring method which utilizes frequency domain analysis to analyze an induction electric signal so as to measure thickness, transforms thickness information from a period which is difficult to read in a time domain to a peak value which is easy to read in the frequency domain, increases the use and lift-off on the basis of not changing an original circuit, and improves the detection precision. However, this method requires preparation of a test piece of known thickness of the same material and a test piece to be tested, and the preparation is troublesome.
Based on the analysis, on the traditional TOF calculation method, a calculation method which has the advantages of simple calculation, stable performance, high processing speed and the like and can obviously improve the calculation accuracy of sound is designed, and the method is a problem to be solved by the technicians in the field.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a TOF calculation method which is simple in calculation, stable in performance, high in processing speed and high in calculation precision, and the thickness is measured accurately by using the method.
In order to solve the technical problems, the invention provides a self-adaptive TOF calculation method based on an improved skip tongue function LMS algorithm and a thickness measurement technology adopting the method, and ultrasonic echo signals are collected firstly; an improved skip tongue function (LMS) algorithm is provided by combining an improved skip tongue function algorithm with a minimum mean square error (LMS) algorithm, an expected signal is estimated and obtained by calculating an autocorrelation mean value of an input signal, a step length adjusting factor and a waveform control coefficient of the skip tongue function are improved by utilizing the autocorrelation mean value, the step length is adjusted by the improved skip tongue function, and the convergence speed of weights is accelerated, so that the performance of the algorithm is improved; and then, an impulse response sequence is iteratively solved by using a self-adaptive algorithm, acoustic time data are calculated, and more accurate thickness data are calculated through sound.
The TOF adaptive computing method is a method for iteratively solving an impulse response sequence using an adaptive algorithm. The calculation principle is that firstly, the problem of TOF calculation is converted into a parameter estimation or pattern recognition problem of finite impulse response, and then an adaptive algorithm is used for solving an impulse response sequence.
The LMS algorithm is an improved algorithm of the steepest descent algorithm, is an optimized extension after the fast descent method is applied in the wiener filtering theory, and is an adaptive iterative algorithm which is proposed by Widrow and Hoff at the earliest to solve the system identification problem. The method has the characteristics of low computational complexity, good convergence in an environment with stable signals, unbiased convergence of expected values to a wiener solution, stability when the algorithm is realized by using limited precision, and the like, and is the algorithm with the best stability and the widest application in the current self-adaptive algorithm. The conventional LMS algorithm has a problem of whether the selection of the step size is proper or not, and the selection of the step size affects the steady state error and convergence speed of the LMS algorithm. The step length is a constant, when the step length is larger, the convergence speed of the LMS algorithm is high, but the steady-state error is larger; the smaller the step size, the smaller the steady state error, but the convergence speed is reduced, so that the contradiction exists between the two.
In order to solve the contradiction, an LMS algorithm based on a skip tongue line is provided, and compared with the traditional LMS algorithm, the LMS algorithm based on the skip tongue line function has the advantages of higher convergence speed and smaller steady-state error, and the defects of the traditional LMS are overcome to a certain extent. However, the influence of the parameter and the selection uncertainty leads to that the algorithm has larger iteration error and step length in the initial iteration stage, the convergence speed of the algorithm is higher at the moment, and the error is always larger until the later iteration stage, so that the step length is always larger, and therefore, the LMS algorithm based on the skip tongue line is further improved. The improved calculation method comprises the following specific steps:
step one, collecting ultrasonic echo signals and preprocessing the ultrasonic echo signals
Intercepting two adjacent echoes from the denoised ultrasonic echo signals, and then zero-filling the cut-off part between the two echoes to obtain front and rear echo signals x respectively 1 (k) And x 2 (k) A. The invention relates to a method for producing a fibre-reinforced plastic composite Then two adjacent ultrasound echo signals may be represented as:
where λ is the ultrasonic attenuation coefficient and D represents the values of two adjacent echoes TOF.
Fourier transforming two adjacent ultrasonic echo signals to obtain:
then let λe -jωD ＝W(e jω ) Then substituting the obtained product into a formula to obtain the following formula:
X 2 (e jω )＝W(e jω )·X 1 (e jω )
according to the time domain convolution theorem: convolution in the time domain corresponds to the product in the frequency domain, and the above equation can be:
x 2 (k)＝x 1 (k)*w(k)
where, is a convolution symbol, W (e jω ) Representing the impulse response between two ultrasound echoes, obtained by calculation of:
the formula uses a sinc function, and the mathematical expression isThe graph characteristics of the combined sinc function can be known: the value of D is calculated from the abscissa of the sinc function, which is symmetrical about the maximum value, and the value of TOF is obtained by calculating the maximum value abscissa of the finite impulse response w (k). Thus, W (e) jω ) The computational accuracy of the TOF is determined, so the performance of the algorithm used for the computation is particularly important.
Step two, introducing an LMS algorithm of a skip tongue line, wherein the LMS algorithm expression based on the skip tongue line is as follows:
e(k)＝d(k)-X T (k)·ω(k)
ω(k+1)＝ω(k)+2μ(k)e(k)X(k)
wherein e (k) represents the output error and d (k) represents the periodThe observation signal, T, represents the sampling period, ω (k) is the update weight at time k, X T (k) = (x (k), x (k+1),..x (k+n-1)) is the input signal, μ (k) represents the iteration step;
a is a step length adjusting factor which controls the value range of the skip tongue function, 0< a <1;
b is a coefficient for determining the waveform size of the skip tongue function, and b is more than 1.
Step three, improving LMS algorithm of skip tongue function and step mu (k) algorithm
The expected signal is estimated and obtained by calculating the autocorrelation mean value of the input signal, the step length adjusting factor a and the waveform control coefficient b of the skip tongue function are improved by utilizing the autocorrelation mean value, and the step length is adjusted by the improved skip tongue function, so that the convergence speed of the weight is increased, and the performance of the algorithm is improved. The improved algorithm expression is as follows:
p(k)＝mp(k-1)+N(1-m)e(k)e(k-1)
β(k)＝δβ(k-1)+γp 2 (k
where p (k) is the instantaneous autocorrelation mean of the error signal, where N is the interference factor of the error signal on the desired signal as a step size μ (k) adjustment function, m is the sampling frequency adjustment parameter, β (k) and α (k) are transformed from a and b, respectively, according to p (k), and δ and γ are fixed constants that adjust the shape of the skip tongue function.
Step four, self-adaptive TOF calculation based on improved skip tongue function LMS algorithm
Firstly calculating the output error e (k) of k iterations, secondly calculating the step length adjusting function p (k), then solving the update step length mu (k), and continuously and iteratively updating W (k) according to the update step length until the last iteration is completed. At this time, the abscissa t corresponding to the maximum value of the impulse response W (k) can be solved 0 Then willt 0 And multiplying the sampling period T to obtain the adaptive TOF value.
A self-adaptive TOF thickness measurement technology based on an improved skip tongue line function LMS algorithm comprises the following specific thickness measurement steps:
step one, obtaining ultrasonic sound velocity of a tested piece
And obtaining the ultrasonic sound velocity s of the measured medium through measurement calculation or an inquiry manual.
Step two, when calculating ultrasonic sound of the measured piece
Calculating ultrasonic time t according to the self-adaptive TOF calculation process based on improved skip tongue function LMS algorithm TO F。
Step three, calculating the thickness of the measured piece
Acquiring ultrasonic sound velocity s of medium of measured piece and sound time t consumed by ultrasonic propagation of measured piece TO After F, the thickness L of the measured piece is defined byThe calculation is performed by using echo signals, so that half of the time is taken when the thickness is calculated.
Compared with the prior art, the invention has the beneficial effects that:
the improved self-adaptive TOF calculation method has simple calculation, stable performance and high processing speed, can obviously improve the calculation accuracy of acoustic time when being used in the acoustic time calculation of ultrasonic thickness measurement, and specifically comprises the following steps:
(1) The expected signal is estimated and obtained by calculating the autocorrelation mean value of the input signal, and the step length adjusting factor and the waveform control coefficient of the skip tongue function are improved, so that the algorithm performance is more stable, the step length can be updated by adjusting the skip tongue function according to the autocorrelation of the signal, the convergence speed of the weight is increased, and the processing speed of the algorithm is further improved. The algorithm calculation process does not need extra preparation or additional equipment facilities, and the preparation process is simple.
(2) The step length is adaptively adjusted through the autocorrelation of the input signals, so that the problem that the selection of the step length parameters of the traditional LMS algorithm can influence the calculation result of the algorithm is solved, the result calculated by the new calculation method is not influenced by the original step length parameters, and the calculation precision of the calculation method is improved. When the method is applied to ultrasonic thickness measurement and calculation, TOF of ultrasonic echo signals on a measured piece can be accurately calculated in a complex environment.
(3) Because the improved self-adaptive TOF calculation method has faster convergence speed and calculation accuracy, when the method is used for measuring the thickness of the wearing part, the wearing amount can be obtained according to the thickness change of the wearing part before and after wearing, the wearing amount is used as the input of a prediction model, the wearing stage of the wearing part is used as the output, the wearing prediction model of the wearing part is built, the measurement efficiency and the measurement accuracy can be improved, the time intercept point of the wearing part from continuous wearing to rupture can be effectively predicted through accurate calculation of the real-time thickness of the wearing part, conditions are created for timely replacement of the wearing part, economic loss caused by the premature scrapping treatment of the wearing part can be saved, and meanwhile, the running failure of equipment caused by the rupture of the wearing part can be avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for the description of the embodiments or the prior art will be briefly described, and it is apparent that the drawings in the following description are only one embodiment of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a self-adaptive TOF calculation method for improving the LMS algorithm of a skip tongue function
FIG. 2 is a spectrum diagram of an original signal and a denoised signal according to an embodiment of the present invention
FIG. 3 is a diagram showing TOF standard deviation of each measuring point calculated by different algorithms in the embodiment of the invention
FIG. 4 is a graph of thickness measurement errors at various measurement points for different algorithms in accordance with an embodiment of the present invention
FIG. 5 is a TOF absolute error diagram in an embodiment of the invention
FIG. 6 is a diagram of TOF error signals and mean square error in an embodiment of the invention
FIG. 7 is a graph showing MSD convergence in accordance with an embodiment of the present invention
Detailed Description
In order to make the technical means, the creation characteristics, the achievement of the purpose and the effect of the present invention easy to understand, the technical solutions in the embodiments of the present invention are clearly and completely described below to further illustrate the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, not all versions.
In order to better illustrate the self-adaptive TOF calculation method based on the improved skip tongue function LMS algorithm and the application of the thickness measuring technology adopting the method in reality, a lining plate after a week of the crusher is adopted as a measured piece, the whole process of thickness measurement by adopting the self-adaptive TOF calculation method is shown, and the calculation flow of the self-adaptive TOF calculation method based on the improved skip tongue function LMS algorithm is shown in figure 1.
Because the uncertainty of the measuring field environment causes that the influence of noise interference on thickness measurement is random, in actual ultrasonic thickness measurement calculation, the thickness calculation is generally performed by utilizing TOF of the first few adjacent echoes of an ultrasonic echo signal, and the specific thickness measurement process is as follows:
firstly, acquiring an ultrasonic signal of a lining plate, carrying out noise reduction treatment on the acquired signal, intercepting two adjacent echoes from the ultrasonic echo signal after noise reduction treatment, and then zero-filling the cut-off part between the two echoes to respectively acquire front and rear echo signals x, wherein the ultrasonic signals before and after the noise reduction treatment are as shown in fig. 2 1 (k) And x 2 (k) A. The invention relates to a method for producing a fibre-reinforced plastic composite Two adjacent ultrasound echo signals may be represented as:
according to the time domain convolution theorem, the relation can be changed as follows: x is x 2 (k)＝x 1 (k)*w(k)；
Wherein:
secondly, introducing an LMS algorithm of a skip tongue line, initializing algorithm parameters, and enabling an algorithm expression to be as follows:
e(k)＝d(k)-X T (k)·ω(k
ω(k+1)＝ω(k)+2μ(k)e(k)X(k)
wherein ω (k) is an update weight at time k;
a is a step length adjusting factor which controls the value range of the skip tongue function, 0< a <1;
b is a coefficient for determining the waveform size of the skip tongue function, and b is more than 1.
The initial iteration step μ (1) =1.5 is selected while the impulse response W (k) of the first iteration is set to zero, i.e., W (1) = [0,..0] T The TOF between the two echoes is set to 6000 sampling periods.
Secondly, updating a calculation step size mu (k) by using an improved skip tongue function LMS algorithm, and improving a and b by calculating an instantaneous autocorrelation mean value of an error signal, wherein the expression is as follows: p (k) =mp (k-1) +n (1-m) e (k) e (k-1).
Wherein, the instantaneous autocorrelation mean value of the p (k) error signal;
n is the interference factor of the error signal on the desired signal, and the value n=10;
m is a sampling frequency adjusting parameter, and the value m=0.005.
From p (k), a and b are transformed into α (k) and β (k), respectively, expressed as:
β(k)＝δβ(k-1)+γp 2 (k)
wherein, delta and gamma are fixed constants for adjusting the shape of the tongue line function of the skip, and the values of delta=0.995 and gamma=0.9e-4 are respectively taken. The step length after improvement is as follows:
in the formula, the expected signal is estimated and obtained by calculating the autocorrelation mean value of the input signal, the step length adjusting factor a and the waveform control coefficient b of the skip tongue function are improved by utilizing the autocorrelation mean value, and the step length is adjusted by the improved skip tongue function, so that the convergence speed of the weight is increased, and the calculation speed is increased.
Next, solving the abscissa t corresponding to the maximum value of the impulse response W (k) 0 Then t is 0 Multiplied by the sampling period T to obtain the acoustic time TOF.
In order to make the calculation result more accurate, the method is adopted to calculate three acoustic TOF values between adjacent echoes, and the average value of the three acoustic TOF values is taken as the TOF value calculated by the thickness of the lining plate.
Finally, the ultrasonic sound velocity s of the lining plate is obtained through the inquiry manual, and the calculated average TOF value t is utilized TOF From the formulaThe thickness of the measuring point lining plate can be calculated and obtained.
In order to verify the reliability of the adaptive TOF calculation method and the thickness measurement technology using the calculation method, 8 points in the lining plate are respectively selected as verification points, and the measured thickness values of the 8 points are shown in Table 1.
TABLE 1 actual measurement of the thickness of each measurement point of the Lining plate
The thickness measurement is carried out by adopting a common cross-correlation method in reality, a traditional LMS algorithm (the value of a parameter mu is 0.0035), an LMS algorithm based on a skip tongue line (the value of a parameter a is 0.006 and the value of a parameter b is 100) and a self-adaptive TOF calculation method (the parameter is consistent with that used in the thickness measurement process) for improving the skip tongue line function LMS algorithm in the invention, so that the values of calculation results of three different adjacent echoes are obtained, and analysis and comparison are carried out. The average relative error results of the three thickness measurement averages are shown in table 2, the TOF standard deviations of the measurement points A1, A2, A3 and A4 are shown in fig. 3, and the errors of the three thickness measurement values of each measurement point of the four algorithms are shown in fig. 4.
Table 2 four methods calculate the relative error of the thickness measurement at each measuring point of the liner
As can be seen from Table 2, the difference between the measured thickness relative errors calculated for the same measuring point of the lining plate by adopting different methods is obvious, wherein the relative error calculated by the cross-correlation method is the largest, and the average relative error value of the calculating method is the smallest. As can be seen from fig. 3, in the four TOF calculation methods, the standard deviation of the cross correlation method is similar to that of the conventional LMS algorithm, the standard deviation of the LMS algorithm based on the skip tongue line is smaller than that of the former two methods, and the standard deviation of the calculation method provided by the present invention is the smallest, which again proves that the calculation method provided by the present invention has higher accuracy in calculating the TOF. According to fig. 4, in all the measuring points, the thickness measuring error of the LMS algorithm based on the skip tongue function is smaller than that of the cross correlation method and the conventional LMS algorithm, but the calculating method of the present invention is smaller than that of the LMS algorithm based on the skip tongue function, and the thickness results of the same measuring point of the lining board obtained by the calculating method of the present invention obtained by different adjacent echoes are more stable.
When the measured thickness value is applied to the calculation of the abrasion loss of the lining plate, the abrasion loss of the lining plate is measured after the abrasion is subtracted from the initial thickness of the lining plate, and the abrasion loss of the lining plate obtained by the self-adaptive TOF thickness measurement method of the improved skip tongue line function LMS algorithm and the actual manual measurement according to the invention is shown in the table 3:
TABLE 3 wear and relative error for each measuring point of the liner
As can be seen from Table 3, the abrasion loss calculated by the ultrasonic echo thickness measuring method and the abrasion loss obtained by manual actual measurement have smaller phase difference, which shows that the abrasion loss calculated by the thickness measuring method accords with the actual manual detection value, can be used for predicting the time cut-off point from continuous abrasion of the lining plate to rupture, and creates conditions for timely replacement of the lining plate.
In order to further verify the stability and the calculation precision of the self-adaptive TOF calculation method, 50 times of simulation are respectively carried out by adopting different TOF calculation methods, the absolute error of the TOF obtained by calculation is shown in figure 5, the absolute error of the TOF calculated by the cross-correlation method is the largest, compared with the absolute error of the TOF calculated by the cross-correlation method, the LMS algorithm with a fixed step length and the LMS algorithm based on a skip tongue line is smaller, the calculation error of the calculation method is the smallest, and in the three variable step length algorithms, the calculation method has smaller absolute error of the TOF, so that the integral performance of the calculation method is better than other methods.
The algorithm is simulated and calculated, the parameters of the lining plate in thickness measurement are continuously utilized, the error signals and the mean square error results of the ultrasonic echo signals are calculated by adopting different algorithms as shown in fig. 6, and as can be known from fig. 6, the error signals and the mean square error calculated by the fixed-step LMS algorithm are smaller than those calculated by the cross-correlation method, so that the calculation method of the self-adaptive sound adopting the LMS algorithm is smaller than the calculation error of the traditional sound adopting the cross-correlation method, and the calculation method of the invention is smaller than those of the error signals and the mean square error calculated by other three algorithms.
Since the value of TOF is obtained from the maximum value of the impulse response sequence between two adjacent pulse echoes when the adaptive algorithm is used for TOF calculation, correspondingly, the accuracy of the adaptive TOF calculation mainly depends on the accuracy of the impulse response sequence obtained by solving after algorithm convergence. The precision of the impulse response sequence can be evaluated through Mean Square Displacement (MSD), the calculated MSD of different algorithms is shown in figure 7, and according to figure 7, the calculated MSD of the calculation method of the invention is obviously smaller than that of other two algorithms, which shows that the precision of the calculated impulse effect sequence is higher; meanwhile, the convergence speed of the calculation method is higher, which indicates that the calculation speed of the calculation method is higher.
Having described the main technical features and fundamental principles of the present invention and related advantages, it will be apparent to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, but may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The above detailed description is, therefore, to be taken in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Furthermore, it should be understood that although the present disclosure describes embodiments in terms of various embodiments, not every embodiment is described in terms of a single embodiment, but rather that the descriptions of embodiments are merely provided for clarity, and that the descriptions of embodiments in terms of various embodiments are provided for persons skilled in the art on the basis of the description.
Claims (3)
1. An adaptive TOF calculation method based on an improved skip tongue function LMS algorithm is characterized by comprising the following steps:
step one, intercepting two adjacent echoes from the denoised ultrasonic echo signals, and then zero-filling the cut-off part between the two echoes to obtain front and rear echo signals x respectively 1 (k) And x 2 (k) The method comprises the steps of carrying out a first treatment on the surface of the Then two adjacent ultrasound echo signals are represented as:
wherein lambda is an ultrasonic attenuation coefficient, and D represents the values of TOF of two adjacent echoes;
fourier transforming two adjacent ultrasonic echo signals to obtain:
then let λe -jωD ＝W(e jω ) Then substituting it intoIn (2), obtaining: x is X 2 (e jω )＝W(e jω )·X 1 (e jω )
According to the time domain convolution theorem, obtaining: x is x 2 (k)＝x 1 (k)*w(k)；
Wherein W (e) jω ) Representing the impulse response between two ultrasound echoes, consisting of w (k) =λsinc [ pi (k-D)]Calculating to obtain;
step two, introducing an LMS algorithm of a skip tongue line, wherein the expression is as follows:
e(k)＝d(k)-X T (k)·ω(k)
ω(k+1)＝ω(k)+2μ(k)e(k)X(k)
wherein e (k) represents the output error, d (k) represents the desired signal, T represents the sampling period, ω (k) is the update weight at time k, X T (k) = (x (k), x (k+1),..x (k+n-1)) is the input signal, μ (k) represents the iteration step;
a is a step length adjusting factor, and the value range of the skip tongue line function is controlled, wherein a is more than 0 and less than 1;
b is a coefficient for determining the waveform size of the skip tongue line function, and b is more than 1;
step three, improving an LMS algorithm and a step mu (k) algorithm of a skip tongue line function, wherein the improved expression is as follows:
p(k)＝mp(k-1)+N(1-m)e(k)e(k-1)
β(k)＝δβ(k-1)+γp 2 (k)
wherein p (k) is the instantaneous autocorrelation mean of the error signal, here as a step mu (k) adjustment function, N is the interference factor of the error signal on the desired signal, m is the sampling frequency adjustment parameter, beta (k) and alpha (k) are respectively transformed from a and b according to p (k), and delta and gamma are fixed constants for adjusting the shape of the skip tongue function;
calculating an output error e (k) and a step length adjusting function p (k) of k iterations, solving an update step length mu (k), and continuously iterating and updating W (k) according to the update step length until the last iteration is completed;
solving for the abscissa t corresponding to the maximum value of the impulse response W (k) 0 Will t 0 Multiplied by the sampling period T is the value of the adaptive TOF.
2. A thickness measurement technique using the calculation method according to claim 1, comprising the steps of:
1. acquiring ultrasonic sound velocity s of a measured piece;
2. calculating ultrasonic time t using the calculation method of claim 1 TOF ；
3. The thickness of the measured piece is measured byAnd (5) calculating to obtain the product.
3. The thickness measurement technique according to claim 2, wherein the ultrasonic sound velocity of the measured piece is obtained by measurement calculation or an inquiry manual.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111631261.XA CN114295084B (en) | 2021-12-28 | 2021-12-28 | Adaptive TOF calculation method based on improved skip tongue line function LMS algorithm and thickness measurement technology adopting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111631261.XA CN114295084B (en) | 2021-12-28 | 2021-12-28 | Adaptive TOF calculation method based on improved skip tongue line function LMS algorithm and thickness measurement technology adopting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114295084A CN114295084A (en) | 2022-04-08 |
| CN114295084B true CN114295084B (en) | 2023-09-26 |
Family
ID=80972337
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111631261.XA Active CN114295084B (en) | 2021-12-28 | 2021-12-28 | Adaptive TOF calculation method based on improved skip tongue line function LMS algorithm and thickness measurement technology adopting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114295084B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5513531A (en) * | 1993-09-01 | 1996-05-07 | Combustion Engineering, Inc. | Ultrasonic system for measurement of thin layers |
| CN102707122A (en) * | 2012-06-15 | 2012-10-03 | 西南交通大学 | Detection method for variable step length LMS (Least Mean Square) harmonic current based on versiera |
| EP2757391A2 (en) * | 2013-01-22 | 2014-07-23 | Schlumberger Technology B.V. | Automatic processing of ultrasonic data |
| CN111175696A (en) * | 2020-04-10 | 2020-05-19 | 杭州优智联科技有限公司 | Single-base-station ultra-wideband AOA (automatic optical inspection) positioning method based on frequency modulated continuous waves |
| CN112054782A (en) * | 2020-08-20 | 2020-12-08 | 中国人民解放军陆军勤务学院 | Variable step size factor construction method for LMS adaptive filtering |
-
2021
- 2021-12-28 CN CN202111631261.XA patent/CN114295084B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5513531A (en) * | 1993-09-01 | 1996-05-07 | Combustion Engineering, Inc. | Ultrasonic system for measurement of thin layers |
| CN102707122A (en) * | 2012-06-15 | 2012-10-03 | 西南交通大学 | Detection method for variable step length LMS (Least Mean Square) harmonic current based on versiera |
| EP2757391A2 (en) * | 2013-01-22 | 2014-07-23 | Schlumberger Technology B.V. | Automatic processing of ultrasonic data |
| CN111175696A (en) * | 2020-04-10 | 2020-05-19 | 杭州优智联科技有限公司 | Single-base-station ultra-wideband AOA (automatic optical inspection) positioning method based on frequency modulated continuous waves |
| CN112054782A (en) * | 2020-08-20 | 2020-12-08 | 中国人民解放军陆军勤务学院 | Variable step size factor construction method for LMS adaptive filtering |
Non-Patent Citations (3)
| Title |
|---|
| 基于相关特性的改进箕舌线变步长LMS算法 -30;路翠华;李国林;周洪庆;徐君明;数据采集与处理;30(4);全文 * |
| 基于相关箕舌线的变步长自适应滤波算法;邢务强;郭华;;微电机(08);全文 * |
| 基于箕舌线函数的变步长归一化最小均方算法;韩允;解传军;刘宝华;胡瑞卿;;现代电子技术(19);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114295084A (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Research on particle filter tracking method based on Kalman filter | |
| CN110455490B (en) | Method and device for calculating supersonic velocity and hypersonic velocity wind tunnel flow field turbulence | |
| CN107450061B (en) | Self-adaptive sound time calculation method in ultrasonic thickness measurement | |
| CN111222088B (en) | Improved method for estimating weighted power harmonic amplitude of flat-top self-convolution window | |
| CN114235072B (en) | Zero-crossing detection-based Coriolis flowmeter phase difference calculation method | |
| JP2002519644A (en) | Sound wave transit time measurement system | |
| CN112380705A (en) | Metal fatigue crack propagation prediction method based on nonlinear prediction filtering algorithm | |
| CN114295084B (en) | Adaptive TOF calculation method based on improved skip tongue line function LMS algorithm and thickness measurement technology adopting method | |
| CN113124948B (en) | High-precision time difference measuring method based on FPGA and cross-correlation method | |
| EP3988964A1 (en) | Method and device for time-of-flight estimate | |
| Liu et al. | Improved real-time denoising method based on lifting wavelet transform | |
| CN111260776A (en) | Three-dimensional shape reconstruction method for adaptive normal analysis | |
| Goffaux et al. | Bioprocess state estimation: some classical and less classical approaches | |
| CN112130138B (en) | Ionized layer phase pollution correction method for over-the-horizon radar | |
| CN111756353A (en) | Liquid level meter noise optimization method based on nonlinear fusion filtering | |
| CN112304376A (en) | Ultrasonic flowmeter flow measuring method based on data fusion | |
| US11572664B2 (en) | Asphalt density estimation system, and related method of reducing signal noise | |
| CN114646935A (en) | Sea clutter distribution parameter estimation method | |
| KR102036790B1 (en) | Condition Assessment System and Method for Cables via Stepped-Frequency Waveform Reflectometry | |
| Boaghe et al. | Spectral analysis for nonlinear wave forces | |
| Farhi et al. | Signal Identification of Underwater Objects Using Autoregressive Moving Average with Exogenous excitation Model | |
| Wolf et al. | Amplitude and frequency estimator for aperiodic multi-frequency noisy vibration signals of a tram gearbox | |
| Gerasimov et al. | Increasing the Efficiency of the Probing Signals Correlation Processing of an Ultrasonic Liquid Flowmeter | |
| Hansen et al. | In-Situ Lubrication Film Thickness Measurements in a Radial Piston Motor Using Adaptive Ultrasound Reflectometry | |
| CN111339673B (en) | Multi-scale noise simulation analysis method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |