Vibration waveform interception method

Abstract

The invention discloses a vibration waveform interception method. The method comprises the following steps: acquiring original data of waveform data; performing segmentation operation to obtain waveform data segments; performing feature extraction on each waveform data segment, calculating a feature calculation value of each waveform data segment and extracting a feature calculation value corresponding to the segment to obtain a feature calculation value sequence and a corresponding serial number; setting a reference value of the characteristic calculation value, determining serial numbers corresponding to a primary ultrasonic starting point and an ending point and serial numbers corresponding to a secondary ultrasonic starting point and an ending point, and correcting the serial numbers corresponding to the primary ultrasonic starting point and the ending point and the serial numbers corresponding to the secondary ultrasonic starting point and the ending point; and extracting a primary ultrasonic waveform and a secondary ultrasonic waveform according to the corrected serial numbers corresponding to the primary ultrasonic starting point and the secondary ultrasonic ending point and the serial numbers corresponding to the secondary ultrasonic starting point and the secondary ultrasonic ending point. According to the method, the speed and accuracy of identifying and tracking the ultrasonic starting point and the ultrasonic ending point are improved, and the efficiency of ultrasonic welding quality detection is improved.

Description

Technical Field

[0001] The invention relates to the technical field of ultrasonic welding, and in particular to a vibration waveform interception method. Background Art

[0002] In the field of ultrasonic welding, power signal and amplitude signal are the main data sources for traditional ultrasonic welding quality monitoring. The above signals are electrical feedback signals loaded by the generator on the transducer piezoelectric ceramic. As a mechanical-electrical coupling component, the mechanical vibration output by the transducer and transmitted to the welding head / bottom mold through the amplitude modulator cannot be fully characterized by the above signals.

[0003] It has become a new trend to use contact or non-contact vibration sensors to measure the actual mechanical vibration signals of the welding head / bottom mold for ultrasonic welding quality monitoring. However, ultrasonic welding is divided into the first ultrasonic delay, the first ultrasonic, the second ultrasonic delay, the second ultrasonic, and the second ultrasonic after stages. The vibration signal sampling rate is generally above 300K, and the amount of waveform data generated in a single welding process is extremely large. If all the original data are used for analysis of ultrasonic welding quality, the amount of calculation is large and the quality detection efficiency is low.

[0004] How to accurately and quickly extract the primary ultrasonic signal and the secondary ultrasonic signal from the original vibration signal is a problem that needs to be solved. Summary of the invention

[0005] In view of this, the purpose of the present invention is to overcome the defects in the prior art and provide a vibration waveform interception method to set interception conditions, identify and track the starting point and end point of the ultrasonic wave, so as to accurately and quickly extract the effective ultrasonic vibration signal from the original vibration signal.

[0006] A vibration waveform interception method of the present invention comprises the steps of:

[0007] Get the raw data of waveform data;

[0008] Performing a segmentation operation on the waveform data to obtain waveform data segments;

[0009] Perform feature extraction on each waveform data segment, calculate the feature calculation value of each waveform data segment and extract the feature calculation value corresponding to the segment, and obtain a feature calculation value sequence and a corresponding sequence number;

[0010] The reference value of the characteristic calculation value is set to determine the serial number corresponding to the starting point and the ending point of the first ultrasound, and the serial number corresponding to the starting point and the ending point of the second ultrasound,

[0011] Correct the sequence number, correct the sequence number corresponding to the starting point and the ending point of the first ultrasound, and correct the sequence number corresponding to the starting point and the ending point of the second ultrasound;

[0012] Extracting an ultrasonic waveform according to the sequence numbers corresponding to the corrected starting point and ending point of the ultrasonic waveform;

[0013] The secondary ultrasound waveform is extracted according to the sequence numbers corresponding to the start point and the end point of the corrected secondary ultrasound.

[0014] Furthermore, in performing a segmentation operation on the waveform data to obtain waveform data segments, a preset waveform segmentation parameter N is set, and the waveform data is cut into waveform data segments of equal length starting from a preset starting point of the waveform data.

[0015] Furthermore, the characteristic calculation method may adopt any of the following methods:

[0016] Perform Fourier transform on each waveform data segment, calculate the power spectrum of each waveform data segment and extract the maximum value Pc of the power spectrum amplitude corresponding to the segment, and obtain the power spectrum sequence Pc and the corresponding sequence number;

[0017] The root mean square (RMS) value of each waveform data segment is calculated as the characteristic calculation value Rc, and the Rc value of each segment and the corresponding sequence number are recorded to obtain the root mean square sequence Rc and the corresponding sequence number.

[0018] Further, when the power spectrum amplitude is selected for calculation, its reference value is LP=MaxPc*Per, where MaxPc is the maximum value in the power spectrum sequence Pc, and Per is the reference value ratio;

[0019] When the RMS calculation is selected, the reference value is LP=MaxRc*Per, where MaxRc is the maximum value in the RMS sequence Rc, and Per is the reference value ratio.

[0020] Furthermore, the reference value ratio Per ranges from 1% to 5%.

[0021] Furthermore, the revised sequence number includes:

[0022] The corrected ultrasound starting point sequence number: Fs" = Fs-Fs';

[0023] The corrected ultrasonic end point sequence number: Fe" = Fe + Fe';

[0024] Corrected secondary ultrasound starting point sequence number: Ss" = Ss-Ss';

[0025] Corrected secondary ultrasound end point sequence number: Se" = Se + Se';

[0026] Among them, Fs' is the correction value of the sequence number of the first ultrasound start point; Fe' is the correction value of the sequence number of the first ultrasound end point; Ss' is the correction value of the sequence number of the second ultrasound start point; Se' is the correction value of the sequence number of the second ultrasound end point.

[0027] Furthermore, the correction value Fs' of the first ultrasonic starting point sequence number is 1-100, the correction value Fe' of the first ultrasonic starting point sequence number is 1-100, the correction value Ss' of the second ultrasonic starting point sequence number is 1-100, and the correction value Se' of the second ultrasonic ending point sequence number is 1-100.

[0028] Furthermore, the extracted first ultrasonic waveform is to extract the original waveform between the segments Fs"~Fe", and the extracted second ultrasonic waveform is to extract the original waveform between the segments Ss"~Se".

[0029] The beneficial effects of the present invention are as follows: a vibration waveform interception method disclosed by the present invention improves the speed and accuracy of identifying and tracking the start and end points of the ultrasonic wave by quickly intercepting the original data;

[0030] By intercepting effective ultrasonic waveforms and setting variable correction parameters, more original waveforms can be intercepted as needed. The intercepted ultrasonic waveforms are used to judge the quality of ultrasonic welding, which improves the efficiency of ultrasonic welding quality detection compared with the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 It is a waveform capture flow chart;

[0033] Figure 2 Divide the original waveform into N segments

[0034] Figure 3 This is a Pc point sequence diagram of the first embodiment;

[0035] Figure 4 The first embodiment obtains the start / end point sequence number diagram according to the reference value;

[0036] Figure 5 A schematic diagram of a primary waveform extracted from the first embodiment;

[0037] Figure 6 This is the Rc point sequence diagram of the second embodiment. DETAILED DESCRIPTION

[0038] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0039] In the description of the embodiments of the present application, the terms "upper", "lower", "front", "rear", etc., are based on the directions or positions shown in the drawings, and are only for the convenience of description and simplification of operation, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation of the present application. In addition, the terms "first", "second", etc. are only used to distinguish in the description and have no special meaning.

[0040] The existing technology uses contact or non-contact vibration sensors to measure the actual mechanical vibration signals of the welding head / bottom mold for ultrasonic welding quality. The collected vibration ultrasonic signals are divided into the first ultrasonic delay, the first ultrasonic, the second ultrasonic delay, the second ultrasonic, and the second ultrasonic after stages. The vibration signal sampling rate is generally above 300K, and the amount of waveform data generated in a single welding process is extremely large. If all the original data are used for analysis when analyzing the ultrasonic welding quality, the amount of calculation is large and the detection efficiency is low.

[0041] In order to solve the above technical problems, according to some embodiments of the present application, a vibration waveform interception method is provided, which can accurately and quickly extract an effective ultrasonic vibration signal from an original vibration signal, such as Figure 1 As shown in the waveform capture flow chart, the specific steps include:

[0042] Get the raw data of waveform data;

[0043] Performing a segmentation operation on the waveform data to obtain waveform data segments;

[0044] Perform feature extraction on each waveform data segment, calculate the feature calculation value of each waveform data segment and extract the feature calculation value corresponding to the segment, and obtain a feature calculation value sequence and a corresponding sequence number;

[0045] Setting a reference value of the characteristic calculation value, which is used to determine the serial number corresponding to the starting point and the ending point of the first ultrasound, and to determine the serial number corresponding to the starting point and the ending point of the second ultrasound;

[0046] Correct the sequence number, correct the sequence number corresponding to the starting point and the ending point of the first ultrasound, and correct the sequence number corresponding to the starting point and the ending point of the second ultrasound;

[0047] Extracting an ultrasonic waveform according to the sequence numbers corresponding to the corrected starting point and ending point of the ultrasonic waveform;

[0048] The secondary ultrasound waveform is extracted according to the sequence numbers corresponding to the start point and the end point of the corrected secondary ultrasound.

[0049] In this embodiment, the original data of the waveform data is acquired by collecting it through a data acquisition device or reading it from a stored data file; Figure 2 As shown, the original waveform includes the vibration signals of the first ultrasonic delay, the first ultrasonic period, the second ultrasonic delay, the second ultrasonic period, and the second ultrasonic period. The data volume is large, but when judging the welding quality, it is only necessary to analyze the complete first ultrasonic and second ultrasonic waveforms. Therefore, the original waveform is divided into N segments, where N=1000 is set; when setting the N value, the larger the N value, the higher the waveform capture accuracy, but the larger the calculation amount and the longer the time; therefore, taking into account both accuracy and time, it is generally recommended that the duration of each segment is 1ms, that is, N=original waveform time / duration of each segment.

[0050] First embodiment

[0051] After the segmentation is completed, Fourier transform is performed on each waveform data segment, the power spectrum of each waveform data segment is calculated, and the maximum value of the power spectrum amplitude corresponding to the segment is extracted and recorded as Pc, where c = 0, 1, 2, 3...N-1, and the power spectrum sequence Pc and the corresponding sequence number are obtained, such as Figure 3 As shown, the horizontal axis represents the Nth segment, and the vertical axis represents the power spectrum sequence Pc corresponding to the Nth segment.

[0052] extract Figure 4The maximum value MaxPc in the power spectrum sequence Pc, the reference value ratio is Per, and the reference value is LP = MaxPc * Per. After setting the reference value, a corresponding reference value line will be formed. By setting the judgment conditions, the sequence number of the starting point of the first ultrasound is Fs, the sequence number of the ending point of the first ultrasound is Fe, the sequence number of the starting point of the second ultrasound is Ss, and the sequence number of the ending point pair of the second ultrasound is Se. The Pc corresponding to the sequence numbers between the sequence number Fs of the starting point of the first ultrasound and the sequence number Fe of the ending point pair of the first ultrasound is ≥ LP, the Pc corresponding to the sequence numbers between the sequence number Fe of the ending point pair of the first ultrasound and the sequence number Ss of the starting point of the second ultrasound is < LP, and the Pc corresponding to the sequence numbers between the sequence number Ss of the starting point of the second ultrasound and the sequence number Se of the ending point pair of the second ultrasound is ≥ LP. Setting the reference value is mainly to observe whether the details of the waveform start and end are complete, that is, whether the intercepted ultrasonic waveform is complete.

[0053] The second embodiment

[0054] After segmentation is completed, calculate the root mean square (RMS) value of each waveform data segment as the feature calculation value Rc, and record the Rc value of each segment and the corresponding sequence number, where c = 0, 1, 2, 3... N - 1, to obtain the root mean square sequence Rc and the corresponding sequence numbers. As Figure 3 shown, the abscissa represents the Nth segment, and the ordinate represents the root mean square Rc obtained corresponding to the Nth segment.

[0055] Extract Figure 4 the maximum value MaxRc in the root mean square sequence Rc, the reference value ratio is Per, and the reference value is LP = MaxRc * Per. After setting the reference value, a corresponding reference value line will be formed. By setting the judgment conditions, the sequence number of the starting point of the first ultrasound is Fs, the sequence number of the ending point of the first ultrasound is Fe, the sequence number of the starting point of the second ultrasound is Ss, and the sequence number of the ending point pair of the second ultrasound is Se. The Rc corresponding to the sequence numbers between the sequence number Fs of the starting point of the first ultrasound and the sequence number Fe of the ending point pair of the first ultrasound is ≥ LP, the Rc corresponding to the sequence numbers between the sequence number Fe of the ending point pair of the first ultrasound and the sequence number Ss of the starting point of the second ultrasound is < LP, and the Rc corresponding to the sequence numbers between the sequence number Ss of the starting point of the second ultrasound and the sequence number Se of the ending point pair of the second ultrasound is ≥ LP.

[0056] The first embodiment uses frequency domain analysis to extract the maximum value of the power spectrum amplitude as a characteristic value and a reference value of the power spectrum amplitude. For a first ultrasound, when the maximum value of the power spectrum amplitude of a certain segment of data exceeds the set reference value, the serial number of the segment of data is determined as the starting point of the first ultrasound. When the maximum value of the power spectrum amplitude is lower than the reference value, the corresponding serial number is the end point of the first ultrasound. Similarly, when the maximum value of the power spectrum amplitude of a certain segment of data exceeds the set reference value again, the serial number of the segment of data is determined as the starting point of the second ultrasound. When the maximum value of the power spectrum amplitude is lower than the reference value again, the corresponding serial number is the end point of the second ultrasound.

[0057] The second embodiment uses time domain analysis to extract the RMS value of each segment of data as a characteristic value. At the beginning of an ultrasound, the RMS value will rise rapidly and exceed the preset starting RMS value reference value. At this time, the serial number of the segment of data is the starting point of the ultrasound. When the RMS value drops below the RMS reference value, the serial number is determined to be the end point of the ultrasound. Similarly, at the beginning of a second ultrasound, the RMS value will rise rapidly again and exceed the preset starting RMS value reference value. At this time, the serial number of the segment of data is the starting point of the second ultrasound. When the RMS value drops below the RMS reference value again, the serial number is determined to be the end point of the second ultrasound. Compared with the calculation method of the first embodiment, this method is simpler because the MS calculation only focuses on the value of each sampling point of the signal. When intercepting a section of the signal, it is only necessary to obtain the sampling point data in the corresponding interval, and then calculate it in sequence according to the established simple arithmetic operation process. It is not necessary to consider the many requirements and complex processing processes of the signal in the frequency domain conversion like FFT, so the interception operation and subsequent calculations are relatively less time-consuming, and the result of measuring the effective value of the signal can be obtained more quickly. By comparison, the calculation interception wavelength length of the second embodiment is 4ms, while the calculation interception wavelength length of the first embodiment is 14ms.

[0058] In a specific embodiment, the serial number at the starting point of the first ultrasound obtained by using the above-mentioned reference value is Fs, the serial number of the first ultrasound end point pair is Fe, the serial number of the starting point of the second ultrasound is Ss, and the serial number Se of the second ultrasound end point pair exists in Pc corresponding to the first ultrasound. Since the reference value line may be located between the two serial numbers, in order to make the intercepted waveform more complete, it is necessary to correct the serial number obtained based on the reference line and expand the interception range.

[0059] In a specific embodiment, the correction value is set as a variable, and the correction value can be set according to accuracy, completeness, etc. The correction formula is as follows:

[0060] The corrected ultrasound starting point sequence number: Fs" = Fs-Fs'

[0061] The corrected ultrasonic end point sequence number: Fe" = Fe + Fe'

[0062] Corrected secondary ultrasound starting point sequence number: Ss”=Ss-Ss'

[0063] Corrected secondary ultrasound end point sequence number: Se" = Se + Se'

[0064] In a specific embodiment, Fs' is the correction value of the sequence number of the first ultrasound starting point, and its value range is 1 to 100; Fe' is the correction value of the sequence number of the first ultrasound ending point, and its value range is 1 to 100; Ss' is the correction value of the sequence number of the second ultrasound starting point, and its value range is 1 to 100; Se' is the correction value of the sequence number of the second ultrasound ending point, and its value range is 1 to 100; Fs" is the corrected sequence number of the first ultrasound starting point; Fe" is the corrected sequence number of the first ultrasound ending point; Ss" is the corrected sequence number of the second ultrasound starting point; Se" is the corrected sequence number of the second ultrasound ending point.

[0065] In a specific embodiment, Fs', Fe', Ss', and Se' are all variables. In actual use, they can be set to the same correction value, or they can be set separately according to the accuracy of the judgment. The larger the numerical setting, the more intercepted data is included, that is, the more segments are selected, and the more complete the ultrasonic waveform is.

[0066] In a specific embodiment, the corrected ultrasound start point sequence number Fs" and the corrected ultrasound end point sequence number Fe" are obtained through the above steps. According to the corrected sequence number obtained, the data of the corresponding segmented original waveform is extracted, that is, the original signal data of the Fs" to Fe" segment is intercepted as an ultrasound, such as Figure 5 As shown in the figure, it is the intercepted graphic data; similarly, it is necessary to extract the data of the original waveform after the corresponding segmentation according to the corrected serial number obtained, that is, the original signal data of the Ss"~Se" segment is intercepted as the secondary ultrasound. This method can quickly identify and track the starting and ending points of the ultrasonic wave, and achieve the purpose of quickly intercepting the effective ultrasonic wave waveform. The intercepted waveform can be used for ultrasonic welding quality inspection. After the ultrasonic wave is effectively intercepted, the quality analysis is performed to improve the efficiency of quality inspection.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A vibration waveform interception method, characterized in that: The method comprises the steps of: Get the raw data of waveform data; Performing a segmentation operation on the waveform data to obtain waveform data segments; Perform feature extraction on each waveform data segment, calculate the feature calculation value of each waveform data segment and extract the feature calculation value corresponding to the segment, and obtain a feature calculation value sequence and a corresponding sequence number; Setting a reference value of the characteristic calculation value, which is used to determine the serial number corresponding to the starting point and the ending point of the first ultrasound, and to determine the serial number corresponding to the starting point and the ending point of the second ultrasound; Correct the sequence number, correct the sequence number corresponding to the starting point and the ending point of the first ultrasound, and correct the sequence number corresponding to the starting point and the ending point of the second ultrasound; Extracting an ultrasonic waveform according to the sequence numbers corresponding to the corrected starting point and ending point of the ultrasonic waveform; The secondary ultrasound waveform is extracted according to the sequence numbers corresponding to the start point and the end point of the corrected secondary ultrasound.

2. A vibration waveform interception method according to claim 1, characterized in that: In performing a segmentation operation on the waveform data to obtain waveform data segments, a preset waveform segmentation parameter N is set, and the waveform data is cut into waveform data segments of equal length starting from a preset starting point of the waveform data.

3. A vibration waveform interception method according to claim 1, characterized in that: The characteristic calculation method may adopt any of the following methods: Perform Fourier transform on each waveform data segment, calculate the power spectrum of each waveform data segment and extract the maximum value Pc of the power spectrum amplitude corresponding to the segment, and obtain the power spectrum sequence Pc and the corresponding sequence number; The root mean square (RMS) value of each waveform data segment is calculated as the characteristic calculation value Rc, and the Rc value of each segment and the corresponding sequence number are recorded to obtain the root mean square sequence Rc and the corresponding sequence number.

4. A vibration waveform interception method according to claim 3, characterized in that: When choosing to calculate the power spectrum amplitude, its reference value is LP = MaxPc*Per, where MaxPc is the maximum value in the power spectrum sequence Pc, and Per is the reference value ratio; When the RMS calculation is selected, the reference value is LP=MaxRc*Per, where MaxRc is the maximum value in the RMS sequence Rc, and Per is the reference value ratio.

5. A vibration waveform interception method according to claim 4, characterized in that: The reference value ratio Per ranges from 1% to 5%.

6. A vibration waveform interception method according to claim 5, characterized in that: The revised serial number includes The corrected ultrasound starting point sequence number: Fs" = Fs-Fs'; The corrected ultrasonic end point sequence number: Fe" = Fe + Fe'; Corrected secondary ultrasound starting point sequence number: Ss" = Ss-Ss'; Corrected secondary ultrasound end point sequence number: Se" = Se + Se'; Among them, Fs' is the correction value of the sequence number of the first ultrasound start point; Fe' is the correction value of the sequence number of the first ultrasound end point; Ss' is the correction value of the sequence number of the second ultrasound start point; Se' is the correction value of the sequence number of the second ultrasound end point.

7. A vibration waveform interception method according to claim 6, characterized in that: The correction value Fs' of the first ultrasonic starting point sequence number is 1-100, the correction value Fe' of the first ultrasonic starting point sequence number is 1-100, the correction value Ss' of the second ultrasonic starting point sequence number is 1-100, and the correction value Se' of the second ultrasonic ending point sequence number is 1-100.

8. A vibration waveform interception method according to claim 6, characterized in that: The extracted first ultrasonic waveform is to extract the original waveform between the segments Fs"~Fe", and the extracted second ultrasonic waveform is to extract the original waveform between the segments Ss"~Se".

Images