Laser fusion depth measuring method and system based on wavelet denoising and percentile filtering

Abstract

The invention discloses a laser penetration measurement method and system based on wavelet denoising and percentile filtering, and the method comprises the steps: constructing an OCT penetration detection device, and collecting the keyhole depth information of a workpiece in the laser penetration welding process, so as to obtain the keyhole depth information of the workpiece; performing filtering processing on the keyhole depth information of the workpiece through a wavelet threshold denoising method to obtain a wavelet coefficient of the keyhole depth information of the workpiece; and performing noise removal processing on the wavelet coefficient of the workpiece keyhole depth information in combination with a percentile filtering algorithm and a moving average algorithm to obtain denoised workpiece keyhole depth information. By using the method, the instantaneous change and local characteristics of the signal can be effectively captured, and the characteristic information of the signal is extracted from different scales, so that the measurement precision of the fusion depth in the laser welding process is improved. The laser penetration measurement method and system based on wavelet denoising and percentile filtering can be widely applied to the technical field of laser penetration measurement.

Description

Technical Field

[0001] The invention relates to the technical field of laser melting depth measurement, and in particular to a laser melting depth measurement method and system based on wavelet denoising and percentile filtering. Background Art

[0002] The laser welding process involves complex physical processes such as heat conduction, melting or solidification, and plasma radiation. The welding process is prone to typical welding defects such as spatter, pores, and cracks due to molten pool fluctuations and uneven heat conduction. These potential welding defects significantly weaken the mechanical properties of welded parts and increase the fatigue risk of parts, thus seriously affecting the service life of the product. In the welding process, the generation of defects is closely related to the stability of the keyhole, and the depth of penetration during the welding process is closely related to the stability of the keyhole. Therefore, the depth of penetration of the weld can reflect the welding quality to a certain extent. Laser welding is a welding method that uses a high-energy-density laser beam as a heat source. However, in actual operation, due to a variety of unpredictable factors, the finished weld may have defects such as thermal cracks, pores, and humps. Among them, insufficient weld depth is one of the main problems affecting weld quality. At present, researchers have developed a variety of monitoring technologies to evaluate the penetration depth during laser welding, including pyrometers, photodiodes, spectrometers, and optical cameras. These technologies mainly infer the penetration depth indirectly by measuring the light intensity radiated during welding. Optical cameras can continuously record the welding process and estimate the depth of the molten pool through image processing technology. Although these methods provide a certain degree of penetration information, they are easily affected by various interference factors in the welding process because they do not directly measure the penetration depth, which limits their accuracy in continuous welding and continuous monitoring.

[0003] During the laser deep melting welding process, the measurement data of the OCT system is scattered. Due to the continuous changes in the keyhole shape, the dynamic surge of the molten pool, and the metal vapor generated by the evaporation of the molten metal, not every measured value is valid melting depth information. It contains a small amount of noise signal, and the existing noise signal needs to be further filtered to obtain the actual melting depth measurement value. Traditional denoising methods mainly include spectral subtraction and Wiener filtering. These methods can reduce the impact of noise to a certain extent, but they also have some defects. For example, spectral subtraction is prone to signal distortion, and Wiener filtering has high requirements on the statistical characteristics of noise.

[0004] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide a laser penetration measurement method and system based on wavelet denoising and percentile filtering, which can effectively capture the instantaneous changes and local characteristics of the signal, extract the characteristic information of the signal from different scales, and thus improve the measurement accuracy of the penetration depth during laser welding.

[0006] The first technical solution adopted by the present invention is: a laser melting depth measurement method based on wavelet denoising and percentile filtering, comprising the following steps:

[0007] An OCT penetration depth detection device is constructed to collect the keyhole depth information of the workpiece during the laser deep penetration welding process to obtain the keyhole depth information of the workpiece;

[0008] The keyhole depth information of the workpiece is filtered by wavelet threshold denoising method to obtain the wavelet coefficient of the keyhole depth information of the workpiece;

[0009] The wavelet coefficients of the workpiece keyhole depth information are processed by combining percentile filtering algorithm and moving average algorithm to obtain the denoised workpiece keyhole depth information.

[0010] Further, the OCT penetration detection device includes an optical coherence tomography device and a laser welding device, and the optical coherence tomography device and the laser welding device are connected through coaxial coupling, wherein the optical coherence tomography device includes a broadband light source generator, a beam splitter, a measuring arm, a reference arm, a reflector and a spectrometer, wherein:

[0011] The broadband light source generator is used to emit broadband light source;

[0012] The beam splitter is used to split the broadband light source to obtain a first branch light source and a second branch light source;

[0013] The measuring arm is used to collect and process information on the keyhole depth of the workpiece according to the first branch light source to obtain a keyhole depth spectrum signal;

[0014] The reference arm is used to collect information on the upper surface of the workpiece according to the second branch light source to obtain a spectral signal of the upper surface profile of the workpiece;

[0015] The reflector is used to transmit the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal to the spectrometer;

[0016] The spectrometer is used to analyze the interference signal between the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal to obtain the keyhole depth information of the workpiece.

[0017] Further, the step of constructing an OCT penetration depth detection device to collect keyhole depth information of a workpiece during laser deep penetration welding to obtain the keyhole depth information of the workpiece specifically includes:

[0018] Construct an OCT penetration detection device;

[0019] Based on the OCT penetration depth detection device, the keyhole depth information of the workpiece during the laser deep penetration welding process is collected to obtain a keyhole depth spectrum signal and a workpiece upper surface profile spectrum signal;

[0020] The interference signal between the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal is obtained, and a fast Fourier transform is performed to obtain the keyhole depth information of the workpiece.

[0021] Further, the step of filtering the keyhole depth information of the workpiece by a wavelet threshold denoising method to obtain the wavelet coefficient of the keyhole depth information of the workpiece specifically includes:

[0022] Standardizing the keyhole depth information of the workpiece to obtain standardized keyhole depth information of the workpiece;

[0023] Determine the number of decomposition layers according to the frequency characteristics of the standardized keyhole depth information of the workpiece;

[0024] Based on the number of decomposition layers, the standardized workpiece keyhole depth information is decomposed by discrete wavelet transform algorithm to obtain the coefficient components of the workpiece keyhole depth information;

[0025] The coefficient components of the workpiece keyhole depth information are filtered by the threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information.

[0026] Further, the step of filtering the coefficient components of the workpiece keyhole depth information by a threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information specifically includes:

[0027] The coefficient components of the workpiece keyhole depth information are filtered by a soft threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information, wherein the soft threshold method is to set a threshold coefficient, set the coefficient components of the workpiece keyhole depth information less than the threshold coefficient to zero, and perform a difference calculation between the coefficient components of the workpiece keyhole depth information greater than the threshold coefficient and the threshold coefficient;

[0028] or;

[0029] The coefficient components of the workpiece keyhole depth information are filtered by a hard threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information, wherein the hard threshold method is to set a threshold coefficient, set the coefficient components of the workpiece keyhole depth information less than the threshold coefficient to zero, and keep the coefficient components of the workpiece keyhole depth information greater than the threshold coefficient unchanged.

[0030] Further, the step of combining the percentile filtering algorithm with the moving average algorithm to remove noise from the wavelet coefficients of the workpiece keyhole depth information to obtain the denoised workpiece keyhole depth information specifically includes:

[0031] Inputting the wavelet coefficient of the workpiece keyhole depth information into a percentile filter for filtering, and obtaining a filtered keyhole depth variation curve, wherein the filtered keyhole depth variation curve represents a curve composed of several percentile values;

[0032] The filtered keyhole depth variation curve is low-pass filtered using the moving average algorithm to obtain the denoised keyhole depth information of the workpiece.

[0033] Furthermore, the step of inputting the wavelet coefficients of the workpiece keyhole depth information into a percentile filter for filtering to obtain a filtered keyhole depth variation curve specifically includes:

[0034] Define the moving center flat window size τ and percentile p of the percentile filter;

[0035] Constructing a keyhole depth data set according to the keyhole depth information of the workpiece, wherein the keyhole depth data set is represented by all measurement values ​​before τ / 2 and after τ / 2;

[0036] Arranging the keyhole depth data set according to a preset sorting rule to obtain an arranged keyhole depth data set;

[0037] The position of the percentile p in the arranged keyhole depth data set is determined, the percentile value is obtained and replaced, and all elements in the arranged keyhole depth data set are traversed to obtain a filtered keyhole depth change curve.

[0038] The second technical solution adopted by the present invention is: a laser melting depth measurement system based on wavelet denoising and percentile filtering, comprising:

[0039] The first module is used to construct an OCT penetration depth detection device to collect keyhole depth information during laser deep penetration welding of the workpiece to obtain the keyhole depth information of the workpiece;

[0040] The second module is used to filter the keyhole depth information of the workpiece by using a wavelet threshold denoising method to obtain the wavelet coefficient of the keyhole depth information of the workpiece;

[0041] The third module is used to combine the percentile filtering algorithm and the moving average algorithm to remove noise from the wavelet coefficients of the workpiece keyhole depth information to obtain the denoised workpiece keyhole depth information.

[0042] The beneficial effects of the method and system of the present invention are as follows: the present invention acquires the keyhole depth information of the workpiece during the laser deep melting welding process by constructing an OCT penetration detection device, and obtains the keyhole depth information of the workpiece, which can directly detect the welding penetration and minimize the interference of the harsh external environment, and can achieve a stable working state, and then filter the keyhole depth information of the workpiece by a wavelet threshold denoising method, and then combine the percentile filtering algorithm with the moving average algorithm to remove noise from the wavelet coefficients of the keyhole depth information of the workpiece, and use the moving average algorithm to low-pass filter the percentile filtering curve for the peak interference in the percentile filtering curve, which can effectively capture the instantaneous changes and local characteristics of the signal, extract the characteristic information of the signal from different scales, and thus improve the measurement accuracy of the penetration depth during laser welding. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of the steps of the laser melting depth measurement method based on wavelet denoising and percentile filtering of the present invention;

[0044] Figure 2 It is a structural block diagram of the laser melting depth measurement system based on wavelet denoising and percentile filtering of the present invention;

[0045] Figure 3 It is a structural schematic diagram of an OCT penetration detection device provided by a specific embodiment of the present invention;

[0046] Figure 4 is a schematic diagram of wavelet threshold denoising decomposition provided by a specific embodiment of the present invention;

[0047] Figure 5 It is a schematic diagram of data collection by the OCT penetration detection device provided in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0048] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only provided for the convenience of explanation and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0049] Reference Figure 1 The present invention provides a laser melting depth measurement method based on wavelet denoising and percentile filtering, the method comprising the following steps:

[0050] S100, constructing an OCT penetration depth detection device to collect keyhole depth information of a workpiece during laser deep penetration welding, thereby obtaining keyhole depth information of the workpiece;

[0051] Specifically, first, it is necessary to construct an OCT penetration detection device, which includes an optical coherence tomography device and a laser welding device, wherein the optical coherence tomography device and the laser welding device are connected via coaxial coupling, wherein the optical coherence tomography device includes a broadband light source generator, a beam splitter, a measuring arm, a reference arm, a reflector and a spectrometer, wherein the broadband light source generator is used to emit a broadband light source; the beam splitter is used to split the broadband light source to obtain a first beam branch light source and a second beam branch light source; the measuring arm is used to collect information on the keyhole depth of the workpiece according to the first beam branch light source to obtain a keyhole depth spectrum signal; the reference arm is used to collect information on the upper surface of the workpiece according to the second beam branch light source to obtain a workpiece upper surface profile spectrum signal; the reflector is used to transmit the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal to the spectrometer; the spectrometer is used to analyze the interference signal between the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal to obtain the keyhole depth information of the workpiece.

[0052] Further, in this embodiment, if Figure 3 As shown, an OCT monitoring system is added to the original laser welding device. The system mainly includes components such as a super luminescent diode (SLD), a fiber coupler, a measuring arm, a reference arm and a spectrometer, wherein the measuring arm includes a collimator. The light emitted from the fiber interface has a divergence angle, and the emitted light presents a conical diffusion beam. Whether it is converging the light in the measuring arm or splitting the collected scattered light signal in the spectrometer, it is necessary to ensure that the emitted light at the fiber end is parallel light, so a collimating lens needs to be added to the fiber output end to collimate the light. The reference arm includes an optical reflector and a plane reflector. The optical reflector: refracts the split light beam in the beam splitter to the plane reflector, and then reflects it to the optical reflector, and interferes with the fiber coupler; the plane reflector: receives the reference laser beam emitted by the reference arm, and reflects it back completely, and returns to the fiber coupler through the optical reflector to interfere with the measurement beam reflected back by the measuring arm.

[0053] Secondly, based on the OCT penetration detection device, the keyhole depth information of the workpiece during the laser deep penetration welding process is collected to obtain the keyhole depth spectral signal and the workpiece upper surface profile spectral signal; the interference signal between the keyhole depth spectral signal and the workpiece upper surface profile spectral signal is obtained, and a fast Fourier transform is performed to obtain the keyhole depth information of the workpiece.

[0054] like Figure 5 As shown in the figure, the SLD emits a broadband light source, which is divided into two paths by the beam splitter. One part of the light enters the reference path and is reflected back to the spectrometer by the reflector; the other part of the light is guided to the processing head and combined with the processing laser beam to focus on the workpiece surface. In the OCT system, by adjusting two optical reflectors with variable angles, the measurement light is divided into two beams: one beam is coaxial with the laser welding beam, reaches the bottom of the keyhole and generates scattered light before returning; the other beam is reflected to the upper surface of the workpiece to form a "keyhole mapping", which can well determine the optimal position of the OCT measurement beam. The light returned from the measurement arm and the reference arm interferes through the fiber coupler, and the original spectral interference fringe image is obtained by the spectrometer linear array camera. The obtained spectral interference pattern is subjected to Fast Fourier Transform (FFT) to obtain the depth position information of the measurement beam. Since the two sub-beams interfere with the same reference arm, the obtained interference pattern contains both the position information of the bottom of the keyhole and the position information of the workpiece surface. Setting the workpiece surface reference facilitates direct quantification of the keyhole depth.

[0055] In summary, it should be noted that the OCT monitoring system mainly obtains the corresponding depth information by analyzing the collected interference spectrum. As the distance between the object arm and the reference arm increases, the modulation of the fringe pattern has a higher spatial frequency. Fast Fourier transform can distinguish the frequency of the measured spectrum to determine the spatial position information of the measured object. Such a monitoring system can directly measure the depth of penetration, thereby greatly improving the accuracy of continuous welding and continuous monitoring. The real-time monitoring of the laser welding process based on OCT can directly obtain the welding depth and weld cross-sectional morphology during processing, which is more convenient to control the keyhole stability during the welding process, reduce the impact of external interference on the welding process, and has significant advantages in welding quality control. Therefore, the study of penetration detection in the laser welding process based on OCT technology can solve the problem of direct depth measurement and ensure the quality of laser welding.

[0056] S200, filtering the keyhole depth information of the workpiece by a wavelet threshold denoising method to obtain a wavelet coefficient of the keyhole depth information of the workpiece;

[0057] Specifically, the keyhole depth information of the workpiece is standardized to obtain the standardized keyhole depth information of the workpiece; the number of decomposition layers is determined according to the frequency characteristics of the standardized keyhole depth information of the workpiece; based on the number of decomposition layers, the standardized keyhole depth information of the workpiece is decomposed by a discrete wavelet transform algorithm to obtain the coefficient components of the keyhole depth information of the workpiece; the coefficient components of the keyhole depth information of the workpiece are filtered by a threshold method to obtain the wavelet coefficients of the keyhole depth information of the workpiece.

[0058] It should be noted that the threshold method includes a soft threshold method and a hard threshold method;

[0059] The coefficient components of the workpiece keyhole depth information are filtered by a soft threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information, wherein the soft threshold method is to set a threshold coefficient, set the coefficient components of the workpiece keyhole depth information less than the threshold coefficient to zero, and perform a difference calculation between the coefficient components of the workpiece keyhole depth information greater than the threshold coefficient and the threshold coefficient; the coefficient components of the workpiece keyhole depth information are filtered by a hard threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information, wherein the hard threshold method is to set a threshold coefficient, set the coefficient components of the workpiece keyhole depth information less than the threshold coefficient to zero, and keep the coefficient components of the workpiece keyhole depth information greater than the threshold coefficient unchanged.

[0060] In this embodiment, there are many methods of wavelet denoising, including modulus maximum denoising, spatial correlation denoising, and threshold denoising. The embodiment of the present invention adopts a threshold denoising method, which considers that after the noisy signal is decomposed by wavelet, the signal component with a larger coefficient is generally the signal required by the system, while the component with a smaller coefficient is generally noise. Therefore, an appropriate threshold is selected to suppress or discard the noise component with a coefficient less than the threshold, and the signal component with a coefficient greater than the threshold is retained. Using such a criterion to implement denoising is wavelet threshold denoising.

[0061] like Figure 4 As shown in the figure, the original signal has been decomposed at three levels. After each decomposition process, the low coefficient component where the noise is located is removed by selecting a suitable threshold, while the signal represented by the high coefficient component is retained. Finally, the wavelet coefficient component after three threshold denoising is reconstructed by wavelet to restore the denoised time domain signal, thus realizing the denoising of the signal.

[0062] First, obtain the signal to be denoised, that is, obtain a noisy image. Standardize the signal for subsequent processing. Select a suitable wavelet basis based on the signal characteristics. Select the appropriate number of decomposition layers (usually 2-4 layers) based on the frequency characteristics and noise level of the signal. Perform wavelet decomposition: Use discrete wavelet transform (DWT) to decompose the signal into multiple levels of approximate coefficients and detail coefficients. These coefficients contain the frequency information of the signal, where low-frequency coefficients correspond to the approximate part of the signal, and high-frequency coefficients correspond to the detailed part of the signal, including noise.

[0063] Furthermore, threshold processing is applied to the wavelet coefficients, usually using soft threshold or hard threshold methods. Soft threshold processing sets coefficients less than the threshold to zero, and coefficients greater than the threshold are subtracted from the threshold, which can retain the edge information of the signal to a certain extent. Hard threshold processing sets coefficients less than the threshold directly to zero, while coefficients greater than the threshold remain unchanged. This method has a more thorough denoising effect, but may lose some signal details.

[0064] S300, combining percentile filtering algorithm and moving average algorithm to perform noise removal on the wavelet coefficients of the workpiece keyhole depth information to obtain the denoised workpiece keyhole depth information.

[0065] S310, inputting the wavelet coefficient of the workpiece keyhole depth information into a percentile filter for filtering, and obtaining a filtered keyhole depth variation curve, wherein the filtered keyhole depth variation curve represents a curve composed of several percentile values;

[0066] Specifically, a moving center flat window size τ and a percentile p of a percentile filter are defined; a keyhole depth data set is constructed according to the keyhole depth information of the workpiece, and the keyhole depth data set is represented by all measurement values ​​before τ / 2 and after τ / 2; the keyhole depth data set is arranged according to a preset sorting rule to obtain an arranged keyhole depth data set; the position of the percentile p in the arranged keyhole depth data set is determined, the percentile value is obtained and replaced, all elements in the arranged keyhole depth data set are traversed, and a filtered keyhole depth change curve is obtained.

[0067] In this embodiment, after obtaining the wavelet coefficients, a threshold needs to be determined to distinguish between signals and noise. Percentile filtering, also known as sorting filtering, is a nonlinear smoothing technique based on the sorting principle. Percentile filtering is calculated in the form of a moving center flat window. There are two main parameters that need to be set: the size of the moving center flat window τ and the percentile p. Assume that at a certain time t i , then at t i All measured values ​​L within the τ / 2 time period before and after i , whose expression is:

[0068]

[0069] In the above formula, the size of the filter window is τ, and Δt is the period between two depth measurements. The percentile filter function sorts all the measurements D(t) of the set Li, which means that at t i The measured depth value at the moment is represented by the set L i The p percentile value Dp(t i ) is replaced by the set L ip% of the measured values ​​are less than Dp(t i ), and finally traverse all the measured values ​​by moving the central flat window to obtain the percentile filtering curve, which is expressed as:

[0070] Dp(t i )=F[D(t)],t∈L i

[0071] In the above formula, F[D(t)] is equivalent to the percentile function, which is used to calculate all depth measurements within the τ window (i.e., the set L i Percentile filtering is performed on all elements in the filter.

[0072] S320, performing low-pass filtering on the filtered keyhole depth variation curve by using a moving average algorithm to obtain denoised workpiece keyhole depth information.

[0073] In this embodiment, the moving average algorithm (MA) is used to perform low-pass filtering on the percentile filter curve to prevent possible spike interference. The moving average algorithm is calculated in a central flat window manner, which can be used to remove data noise, reduce high-frequency disturbances in the data, and retain useful low-frequency trends. The parameters that need to be set for the moving average are the moving window size T. Assume that the penetration value obtained by the percentile filter algorithm at any time t is D t , then before and after time t The value set D of the penetration depth obtained by the percentile filtering algorithm within time is expressed as:

[0074]

[0075] The moving average algorithm uses the arithmetic mean of the elements in the set D. Instead of the penetration value D extracted by the percentile filtering algorithm t Finally, the moving window method is used to traverse all the penetration depth values ​​to fit the penetration depth curve.

[0076] Apply percentile filtering to each layer of high-frequency detail coefficients: Sort the high-frequency coefficients. Calculate the position of the selected percentile and set the coefficients above that position to zero or replace them with other values ​​(such as the mean or median) to suppress noise. Combine the processed low-frequency approximation coefficients and the percentile-filtered high-frequency detail coefficients and reconstruct the signal using the inverse wavelet transform (IDWT). This step will restore a denoised signal.

[0077] The laser welding system uses a single-mode continuous fiber laser, and the maximum power, center wavelength, beam quality (Beam Parameter Product, BPP), sampling rate and other parameters of the laser need to be set. The laser head and the worktable are tilted vertically to form a certain angle. The OCT penetration monitoring system uses a superluminescent diode (SLD) as the radiation light source, and the measuring beam and the processing beam share the focusing mirror, and the measuring beam and the processing beam are coaxially aligned.

[0078] OCT technology is based on the principle of optical coherence tomography. The monitoring process is not affected by the strong light, smoke and other processing environments of welding, so the measurement accuracy is extremely high. The accuracy of online monitoring of welding penetration is mainly affected by the axial resolution of the OCT system. The axial resolution refers to the minimum distance that the OCT can distinguish along the direction of weld penetration, and its calculation formula is:

[0079]

[0080] In the above formula: Δz represents the axial resolution (mm); λ 0 Indicates the wavelength of the light source (nm); It represents the half-maximum full width of the light source (nm). By continuously trying different light source wavelengths and half-maximum full widths of the light source, good axial resolution can be obtained to improve the accuracy of online monitoring of welding penetration.

[0081] In addition, the accuracy of OCT online monitoring of penetration depth is also an important parameter. The accuracy of OCT online monitoring of penetration depth is described by the error rate δ of the monitoring result, and its calculation formula is:

[0082]

[0083] In the above formula, δ represents the error rate (%); d 0 represents the actual penetration depth of the weld measured by metallography (mm); d represents the penetration depth of the weld measured by the OCT monitoring system (mm).

[0084] The smaller the error rate δ, the closer the OCT measured penetration result is to the true value, and the higher the measurement accuracy. This step requires obtaining a metallographic image after the experiment and comparing it with the monitoring result image. The error rate value can be used to obtain more accurate experimental results.

[0085] In summary, the embodiment of the present invention adopts an OCT monitoring system to detect the penetration depth of laser welding. Compared with traditional pyrometers, photodiodes, spectrometers and optical cameras, it can directly detect the welding penetration depth, and minimize the interference of the harsh external environment, and can achieve a stable working state. A method for fitting an OCT penetration curve using percentile filtering and a moving average algorithm is proposed. The penetration curve has a higher degree of consistency with the actual weld depth curve, and high-precision measurement of the laser welding penetration is achieved. The method uses a percentile filtering algorithm to extract the penetration curve of the OCT measurement data. In view of the spike interference in the percentile filtering curve, the percentile filtering curve is low-pass filtered using a moving average algorithm, and finally an OCT penetration measurement curve is fitted. The OCT penetration measurement curve is compared with the actual weld depth curve, and the average error is small. Compared with the penetration curve extracted only by the percentile filtering algorithm, the measurement accuracy of the finally fitted penetration measurement curve is greatly improved. Wavelet transform combined with percentile filtering has the following characteristics:

[0086] 1) Time-frequency localization: Wavelet function has good localization characteristics in time domain and frequency domain, and can effectively capture the instantaneous changes and local characteristics of the signal.

[0087] 2) Multi-scale analysis: Wavelet transform can perform multi-scale analysis on signals and extract characteristic information of signals from different scales.

[0088] 3) Adaptability: The wavelet function can be selected according to the characteristics of the signal and has strong adaptability.

[0089] Reference Figure 2 , a laser melting depth measurement system based on wavelet denoising and percentile filtering, including:

[0090] The first module 201 is used to construct an OCT penetration depth detection device to collect keyhole depth information of a workpiece during laser deep penetration welding to obtain keyhole depth information of the workpiece;

[0091] The second module 202 is used to filter the keyhole depth information of the workpiece by using a wavelet threshold denoising method to obtain the wavelet coefficient of the keyhole depth information of the workpiece;

[0092] The third module 203 is used to perform noise removal processing on the wavelet coefficients of the workpiece keyhole depth information by combining the percentile filtering algorithm and the moving average algorithm to obtain the workpiece keyhole depth information after noise removal.

[0093] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0094] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A laser penetration depth measurement method based on wavelet denoising and percentile filtering, characterized in that: The following steps are involved: An OCT penetration depth detection device is constructed to collect the keyhole depth information of the workpiece during the laser deep penetration welding process to obtain the keyhole depth information of the workpiece; The keyhole depth information of the workpiece is filtered by wavelet threshold denoising method to obtain the wavelet coefficient of the keyhole depth information of the workpiece; The wavelet coefficients of the workpiece keyhole depth information are processed by combining percentile filtering algorithm and moving average algorithm to obtain the denoised workpiece keyhole depth information.

2. The laser penetration depth measurement method based on wavelet denoising and percentile filtering according to claim 1 is characterized in that: The OCT penetration detection device includes an optical coherence tomography device and a laser welding device, wherein the optical coherence tomography device and the laser welding device are connected via coaxial coupling, wherein the optical coherence tomography device includes a broadband light source generator, a beam splitter, a measuring arm, a reference arm, a reflector and a spectrometer, wherein: The broadband light source generator is used to emit broadband light source; The beam splitter is used to split the broadband light source to obtain a first branch light source and a second branch light source; The measuring arm is used to collect and process information on the keyhole depth of the workpiece according to the first branch light source to obtain a keyhole depth spectrum signal; The reference arm is used to collect information on the upper surface of the workpiece according to the second branch light source to obtain a spectral signal of the upper surface profile of the workpiece; The reflector is used to transmit the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal to the spectrometer; The spectrometer is used to analyze the interference signal between the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal to obtain the keyhole depth information of the workpiece.

3. The laser penetration depth measurement method based on wavelet denoising and percentile filtering according to claim 2 is characterized in that: The step of constructing an OCT penetration depth detection device to collect keyhole depth information during laser deep penetration welding of a workpiece to obtain the keyhole depth information of the workpiece specifically includes: Construct an OCT penetration detection device; Based on the OCT penetration depth detection device, the keyhole depth information of the workpiece during the laser deep penetration welding process is collected to obtain a keyhole depth spectrum signal and a workpiece upper surface profile spectrum signal; The interference signal between the keyhole depth spectrum signal and the workpiece upper surface profile spectrum signal is obtained, and a fast Fourier transform is performed to obtain the keyhole depth information of the workpiece.

4. The laser penetration depth measurement method based on wavelet denoising and percentile filtering according to claim 3 is characterized in that: The step of filtering the keyhole depth information of the workpiece by the wavelet threshold denoising method to obtain the wavelet coefficient of the keyhole depth information of the workpiece specifically includes: Performing standardization processing on the keyhole depth information of the workpiece to obtain standardized keyhole depth information of the workpiece; Determine the number of decomposition layers according to the frequency characteristics of the standardized keyhole depth information of the workpiece; Based on the number of decomposition layers, the standardized workpiece keyhole depth information is decomposed by discrete wavelet transform algorithm to obtain the coefficient components of the workpiece keyhole depth information; The coefficient components of the workpiece keyhole depth information are filtered by the threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information.

5. The laser penetration depth measurement method based on wavelet denoising and percentile filtering according to claim 4 is characterized in that: The step of filtering the coefficient components of the workpiece keyhole depth information by a threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information specifically includes: The coefficient components of the workpiece keyhole depth information are filtered by a soft threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information, wherein the soft threshold method is to set a threshold coefficient, set the coefficient components of the workpiece keyhole depth information less than the threshold coefficient to zero, and perform a difference calculation between the coefficient components of the workpiece keyhole depth information greater than the threshold coefficient and the threshold coefficient; or; The coefficient components of the workpiece keyhole depth information are filtered by a hard threshold method to obtain the wavelet coefficients of the workpiece keyhole depth information, wherein the hard threshold method is to set a threshold coefficient, set the coefficient components of the workpiece keyhole depth information less than the threshold coefficient to zero, and keep the coefficient components of the workpiece keyhole depth information greater than the threshold coefficient unchanged.

6. The laser penetration depth measurement method based on wavelet denoising and percentile filtering according to claim 5 is characterized in that: The step of combining the percentile filtering algorithm and the moving average algorithm to remove noise from the wavelet coefficients of the workpiece keyhole depth information to obtain the denoised workpiece keyhole depth information specifically includes: Inputting the wavelet coefficient of the workpiece keyhole depth information into a percentile filter for filtering, and obtaining a filtered keyhole depth variation curve, wherein the filtered keyhole depth variation curve represents a curve composed of several percentile values; The filtered keyhole depth variation curve is low-pass filtered using the moving average algorithm to obtain the denoised keyhole depth information of the workpiece.

7. The laser penetration depth measurement method based on wavelet denoising and percentile filtering according to claim 6 is characterized in that: The step of inputting the wavelet coefficient of the workpiece keyhole depth information into the percentile filter for filtering to obtain the filtered keyhole depth variation curve specifically includes: Define the moving center flat window size τ and percentile p of the percentile filter; Constructing a keyhole depth data set according to the keyhole depth information of the workpiece, wherein the keyhole depth data set is represented by all measurement values ​​before τ / t and after τ / t; Arranging the keyhole depth data set according to a preset sorting rule to obtain an arranged keyhole depth data set; The position of the percentile p in the arranged keyhole depth data set is determined, the percentile value is obtained and replaced, and all elements in the arranged keyhole depth data set are traversed to obtain a filtered keyhole depth change curve.

8. The laser melting depth measurement system based on wavelet denoising and percentile filtering is characterized by: Includes the following modules: The first module is used to construct an OCT penetration depth detection device to collect keyhole depth information during laser deep penetration welding of the workpiece to obtain the keyhole depth information of the workpiece; The second module is used to filter the keyhole depth information of the workpiece by using a wavelet threshold denoising method to obtain the wavelet coefficient of the keyhole depth information of the workpiece; The third module is used to combine the percentile filtering algorithm and the moving average algorithm to remove noise from the wavelet coefficients of the workpiece keyhole depth information to obtain the denoised workpiece keyhole depth information.

Images