CN117102630A - Arc welding quality monitoring method, device, electronic equipment and storage medium - Google Patents
Arc welding quality monitoring method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117102630A CN117102630A CN202311381336.2A CN202311381336A CN117102630A CN 117102630 A CN117102630 A CN 117102630A CN 202311381336 A CN202311381336 A CN 202311381336A CN 117102630 A CN117102630 A CN 117102630A
- Authority
- CN
- China
- Prior art keywords
- welding
- pulse
- waveform
- voltage
- current
- 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.)
- Granted
Links
- 238000003466 welding Methods 0.000 title claims abstract description 688
- 238000000034 method Methods 0.000 title claims abstract description 132
- 238000012544 monitoring process Methods 0.000 title claims abstract description 72
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 238000013441 quality evaluation Methods 0.000 claims abstract description 42
- 230000008569 process Effects 0.000 claims abstract description 41
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 238000001303 quality assessment method Methods 0.000 claims abstract description 22
- 238000012806 monitoring device Methods 0.000 claims abstract description 11
- 238000005070 sampling Methods 0.000 claims description 37
- 230000000630 rising effect Effects 0.000 claims description 34
- 230000008859 change Effects 0.000 claims description 28
- 238000001514 detection method Methods 0.000 claims description 18
- 230000035515 penetration Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 10
- 230000009191 jumping Effects 0.000 claims description 9
- 230000015654 memory Effects 0.000 claims description 9
- 230000010365 information processing Effects 0.000 claims description 4
- 238000010891 electric arc Methods 0.000 claims description 3
- 230000002349 favourable effect Effects 0.000 abstract description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 2
- 238000004422 calculation algorithm Methods 0.000 description 24
- 238000004364 calculation method Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 13
- 230000007547 defect Effects 0.000 description 12
- 238000012545 processing Methods 0.000 description 12
- 238000003708 edge detection Methods 0.000 description 9
- 238000000605 extraction Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000012935 Averaging Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 238000009499 grossing Methods 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 239000000523 sample Substances 0.000 description 4
- 238000007405 data analysis Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 2
- 230000003321 amplification Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- 238000005493 welding type Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K9/00—Arc welding or cutting
- B23K9/095—Monitoring or automatic control of welding parameters
- B23K9/0953—Monitoring or automatic control of welding parameters using computing means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K9/00—Arc welding or cutting
- B23K9/32—Accessories
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention provides an arc welding quality monitoring method, an arc welding quality monitoring device, electronic equipment and a storage medium, which belong to the technical field of intelligent manufacturing and mainly comprise the following steps: acquiring welding waveforms during arc welding, including welding voltage waveforms and welding current waveforms; extracting a plurality of waveform features of a welding waveform; determining a weld quality assessment parameter based on at least a portion thereof; and determining the welding phenomenon during arc welding according to the welding quality evaluation parameters, the waveform characteristics and the preliminary recognition result of the welding image during arc welding. According to the invention, the characteristics of the welding waveform in the welding process are comprehensively extracted, and the welding quality evaluation parameters for welding phenomenon analysis are determined according to the extracted waveform characteristics, so that the waveform characteristics, the welding quality evaluation parameters and the analysis result of the welding image are synthesized, the welding phenomenon in the arc welding process is comprehensively monitored, the accuracy of the monitoring result can be effectively improved, and a favorable basis is provided for the adjustment of the arc welding process.
Description
Technical Field
The present invention relates to the field of intelligent manufacturing technologies, and in particular, to a method and apparatus for monitoring welding quality of arc welding, an electronic device, and a storage medium.
Background
Arc welding quality monitoring refers to the real-time or subsequent detection and assessment of the quality of a weld in an arc welding process. Its main purpose is to ensure that the welding quality meets the expected standard and to find and correct problems that may occur during the welding process in time. Arc welding quality monitoring is essential to ensure weld quality, improve process stability, reduce costs, and meet standard requirements. The welding method can effectively improve the reliability and consistency of the welding process and improve the quality and reliability of the welded product.
Currently, existing monitoring schemes only monitor welding quality by using image or characteristics of electrical signals during welding, for example: a thick plate welding quality monitoring and controlling system and a welding method thereof realize high-quality welding of thick plates through quality monitoring control of a welding process. The welding defect type is judged by mainly utilizing the characteristic value of the molten pool and the characteristic value of the welding electric signal extracted by the characteristic analysis module.
However, the existing monitoring scheme does not fully extract the characteristic value of the welding electric signal, and does not describe the characteristics of the actual welding defect and the characteristics of the electric signal in detail, and the lack of judging logic between the welding defect and the characteristic value of the signal is mainly because the welding physical phenomenon is not very known, so that the adopted characteristic extraction means and analysis method do not really reflect the current welding quality, and therefore, the impending adverse phenomenon cannot be accurately predicted.
Disclosure of Invention
The invention provides an arc welding quality monitoring method, an arc welding quality monitoring device, electronic equipment and a storage medium, which are used for solving the defects that the extraction of a welding electric signal characteristic value is incomplete or the analysis between the welding electric signal characteristic value and a welding phenomenon is unclear in the prior art, and can realize the accurate analysis of the welding phenomenon.
In a first aspect, the present invention provides a method of monitoring the quality of an arc welding process, comprising:
acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms;
extracting a plurality of waveform features of the welding waveform;
determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform features;
and determining the welding phenomenon during arc welding according to the welding quality evaluation parameters and the waveform characteristics and combining the primary recognition result of the welding image during arc welding.
According to the method for monitoring the welding quality of the arc welding, provided by the invention, the waveform characteristics comprise at least one of the following:
pulse peak current, pulse base current, pulse peak voltage, pulse base voltage, pulse frequency, pulse period average current, and pulse period average voltage.
According to the method for monitoring the welding quality of the arc welding, the extracting of the waveform characteristics of the welding waveform comprises the following steps:
acquiring a real-time welding waveform in a current sampling period;
determining each waveform key point of the real-time welding waveform; the waveform key points comprise rising edge starting time, rising edge ending time, falling edge starting time and falling edge ending time of each pulse in the welding waveform;
determining a pulse period and a pulse frequency of the real-time welding waveform based on the rising edge starting time of each pulse so as to calculate average current and average voltage in the pulse period as the average current and average voltage of the pulse period;
determining each pulse peak plateau based on a rising edge start time and a rising edge end time of each pulse to determine the pulse peak current and the pulse peak voltage based on the pulse peak plateau;
each pulse base plateau is determined based on a falling edge start time and a falling edge end time of a pulse to determine the pulse base current and the pulse base voltage based on the pulse base plateau.
According to the method for monitoring the welding quality of the arc welding, under the condition that the waveform distortion of the real-time welding waveform acquired in the current sampling period is determined, the following operations are executed:
Replacing the real-time welding waveform by using the historical welding waveform acquired in the previous sampling period of the current sampling period;
the weld quality assessment parameter is determined based on the waveform characteristics of the historical welding waveform.
According to the method for monitoring the welding quality of the arc welding, the pulse peak current and the pulse peak voltage are determined based on the pulse peak platform, and the method comprises the following steps:
determining the pulse peak current and the pulse peak voltage based on waveforms near a start point, an end point, and a middle point of the pulse peak plateau;
alternatively, the pulse peak current is determined based on a current average of all waveforms of the pulse peak plateau, and the pulse peak voltage is determined based on a voltage average of all waveforms of the pulse peak plateau;
alternatively, the pulse peak current is determined based on a current median of all waveforms of the pulse peak plateau, and the pulse peak voltage is determined based on a voltage median of all waveforms of the pulse peak plateau.
According to the method for monitoring the welding quality of the arc welding, the pulse base value current and the pulse base value voltage are determined based on the pulse base value platform, and the method comprises the following steps:
Determining the pulse base current and the pulse base voltage based on waveforms near a start point and an end point of the pulse base plateau;
or, based on the maximum probability value of all sampling points of the pulse base value platform, the current and the voltage of the sampling point corresponding to the maximum probability value are used as the pulse base value current and the pulse base value voltage.
According to the arc welding quality monitoring method provided by the invention, the welding quality evaluation parameters comprise pulse arc length voltage;
the determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform characteristics includes:
determining a dry extension resistance based on the pulse peak voltage, the pulse base voltage, the pulse peak current, and the pulse base current;
determining a dry extension voltage based on the pulse peak current and the dry extension resistance;
the pulse arc length voltage is determined based on the pulse peak voltage and the dry stretch voltage.
According to the method for monitoring the welding quality of the arc welding provided by the invention, the welding phenomenon during the arc welding is determined according to the welding quality evaluation parameter and the waveform characteristic and by combining the primary recognition result of the welding image during the arc welding, and the method comprises the following steps:
Determining a pulse average voltage variation based on an average value of the average voltage of the continuous plurality of pulse periods;
determining a pulse average current variation based on an average value of the average current of a plurality of continuous pulse periods;
and determining that the welding phenomenon is welding penetration when the primary identification result is determined to be welding penetration, wherein the pulse average voltage variation is larger than or equal to a first preset threshold value, and the pulse average current variation is larger than or equal to a second preset threshold value.
The method for monitoring the welding quality of the arc welding provided by the invention further comprises the following steps:
determining that the welding phenomenon is necking when the electric arc welding is performed under the condition that the preliminary identification result is necking, the pulse average voltage variation is larger than or equal to the first preset threshold value, and the pulse average current variation is larger than or equal to a third preset threshold value;
the third preset threshold is smaller than the second preset threshold.
According to the arc welding quality monitoring method provided by the invention, the method further comprises the step of determining the pulse arc length voltage variation, wherein the pulse arc length voltage variation is determined based on pulse arc length voltages of a plurality of continuous pulse periods;
Determining a pulse arc length voltage variation ratio of the plurality of pulse periods according to the pulse arc length voltage variation, wherein the pulse arc length voltage variation ratio is used for representing the variation amplitude of the pulse arc length voltage in the continuous plurality of pulse periods;
and when the preliminary identification result is determined to be normal, the pulse average voltage variation is larger than or equal to the fourth preset threshold value but smaller than or equal to the first preset threshold value, the pulse average current variation is smaller than or equal to the third preset threshold value but larger than or equal to a fifth preset threshold value, and the pulse arc length voltage variation ratio is larger than or equal to the sixth preset threshold value but smaller than or equal to the seventh preset threshold value, determining that the welding phenomenon is arc jumping instability during arc welding.
The method for monitoring the welding quality of the arc welding provided by the invention further comprises the following steps: and determining that the welding phenomenon is unstable when the arc welding is performed under the condition that the preliminary identification result is normal, the pulse average voltage variation is larger than or equal to the first preset threshold value, and the pulse average current variation is larger than or equal to a third preset threshold value.
The method for monitoring the welding quality of the arc welding provided by the invention further comprises the following steps: and determining that the welding phenomenon is a cylinder out of round when the preliminary identification result is normal, the pulse average voltage variation is larger than or equal to the first preset threshold, the pulse average current variation is larger than or equal to a third preset threshold and the pulse arc length voltage variation ratio is larger than or equal to an eighth threshold.
According to the arc welding quality monitoring method provided by the invention, the preliminary identification result of the welding image during arc welding is determined based on the following steps: acquiring a welding image during the arc welding; and inputting the welding image into a pre-trained welding quality detection model to acquire the preliminary identification result output by the welding quality detection model.
In a second aspect, the present invention also provides an arc welding quality monitoring device, comprising:
the welding information acquisition unit is used for acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms;
a welding information analysis unit for extracting a plurality of waveform features of the welding waveform;
A welding information processing unit for determining at least one welding quality assessment parameter based on at least a portion of the plurality of waveform features;
and the welding phenomenon determining unit is used for determining the welding phenomenon during the arc welding according to the welding quality evaluation parameter and the waveform characteristic and combining the primary recognition result of the welding image during the arc welding.
In a third aspect, the present invention provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any one of the arc welding quality monitoring methods described above when the program is executed.
In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the arc welding quality monitoring method as described in any of the above.
According to the arc welding quality monitoring method, the device, the electronic equipment and the storage medium, the characteristics of the welding waveform in the welding process are comprehensively extracted, and the welding quality evaluation parameters for welding phenomenon analysis are determined according to the extracted waveform characteristics, so that the waveform characteristics, the welding quality evaluation parameters and the analysis results of welding images are synthesized, the welding phenomenon during arc welding is comprehensively monitored, the accuracy of the monitoring results can be effectively improved, and a favorable basis is provided for adjusting the arc welding process.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the 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 an arc welding quality monitoring method provided by the invention;
FIG. 2 is a schematic diagram of a system for weld quality monitoring provided by the present invention;
FIG. 3 is a schematic view of weld puddle related information acquisition provided by the present invention;
FIG. 4 is a schematic diagram of an online OK/NG determination of a welding waveform provided by the present invention;
FIG. 5 is a schematic diagram of waveform characteristics under PFM feedback provided by the present invention;
FIG. 6 is a schematic diagram of waveform characteristics under PFM-PAM mixed feedback provided by the present invention;
FIG. 7 is a diagram showing a change in welding phenomenon corresponding to a change in welding waveform according to the present invention;
FIG. 8 is a flow chart of the welding phenomenon determination logic according to the present invention;
FIG. 9 is a second flow chart of the welding phenomenon determination logic according to the present invention;
FIG. 10 is a third flow chart of the welding phenomenon determination logic according to the present invention;
FIG. 11 is a flow chart of the welding phenomenon determination logic provided by the present invention;
FIG. 12 is a flow chart of the welding phenomenon determination logic provided by the present invention;
FIG. 13 is a schematic view of a construction of an arc welding quality monitoring apparatus provided by the present invention;
fig. 14 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that in the description of embodiments of the present invention, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. The orientation or positional relationship indicated by the terms "upper", "lower", etc. are based on the orientation or positional relationship shown in the drawings, are merely for convenience of description and to simplify the description, and are not indicative or implying that the apparatus or elements in question must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the present invention. Unless specifically stated or limited otherwise, the terms "mounted," "connected," and "coupled" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
The terms "first," "second," and the like in this specification are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present application may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type, and are not limited to the number of objects, such as the first object may be one or more. In addition, "and/or" indicates at least one of the connected objects, and the character "/", generally indicates that the associated object is an "or" relationship.
The arc welding quality monitoring method, apparatus, electronic device and storage medium provided by the embodiments of the present application are described below with reference to fig. 1 to 14.
At present, only the characteristics of images or electric signals are generally utilized to monitor welding quality, for example, the characteristic value of a welding pool or the characteristic value of a welding electric signal is simply extracted to judge welding phenomena, and mainly the types of welding defects are analyzed.
Because the prior art does not describe the characteristics of the actual welding defect and the characteristics of the electrical signal in detail, especially lacks the judging logic between the type of the welding defect and the characteristics of the electrical signal, does not really and clearly clarify what kind of change occurs in the characteristics of the electrical signal to represent what defect occurs, and the reason is mainly that the welding physical phenomenon is not very known, and the acquisition means and the analysis method do not really reflect the current welding quality, so that the impending bad welding phenomenon cannot be accurately predicted.
FIG. 1 is a flow chart of the method for monitoring the welding quality of an arc welding provided by the invention, as shown in FIG. 1, including but not limited to the following steps:
step 101: a welding waveform at the time of arc welding is obtained, the welding waveform including a welding voltage waveform and a welding current waveform.
The acquisition of the welding waveform during arc welding requires first the use of corresponding equipment to monitor the welding voltage and welding current. Typically, this is accomplished using a specialized welder with a matched monitoring device.
The invention utilizes the acquisition board card with the acquisition frequency reaching 100K to acquire the welding current of the welding machine in real time when in welding in cooperation with a welding current sensor (such as a Hall voltage sensor), and utilizes the welding voltage sensor (such as a voltage acquisition probe) to acquire the welding voltage of the welding machine in real time when in welding.
FIG. 2 is a schematic diagram of a system for welding quality monitoring according to the present invention, where the system for welding quality monitoring shown in FIG. 2 may be used to collect welding waveforms, and includes, but is not limited to, the following steps:
1) Preparing a welding machine and monitoring equipment: an arc welder capable of providing a welding current and voltage output is selected and connected to a welding current sensor (denoted I in fig. 2), a welding voltage sensor (denoted U in fig. 2), and an image sensor, which can convert the detected current and voltage into measurable electrical signals, respectively, for capturing welding images in real-time during arc welding.
It should be noted that the acquired welding voltage waveform, welding current waveform and welding image are corresponding in time sequence, that is, the welding voltage, welding current and welding image are simultaneously acquired at the same sampling time.
2) And (3) connecting monitoring equipment: the monitoring device is connected to the welder to receive and record welding voltage and current signals in real time. These devices typically include data collectors, oscilloscopes, data recorders, and the like.
3) Welding: and starting the welding machine to perform welding operation according to specific welding requirements. The acquisition of the welding waveform is performed during the welding process.
4) Data acquisition and recording: in the welding process, the monitoring equipment can acquire waveform signals of welding voltage and welding current in real time. These waveform signals may then be recorded or displayed in real time on an oscilloscope or the like.
Step 102: extracting a plurality of waveform features of the welding waveform;
after the welding waveform is obtained, data analysis and processing can be performed on the collected welding voltage waveform and welding current waveform. For example, waveform parameters such as the welding voltage waveform and the peaks, valleys, peak voltages, peak currents, etc. of the welding current waveform are extracted and calculated using data analysis software.
Step 103: at least one weld quality assessment parameter is determined based on at least a portion of the plurality of waveform characteristics.
Unlike the prior art that the welding phenomenon is determined directly through analysis of single waveform characteristics, the method comprehensively realizes accurate analysis of the welding phenomenon through analysis of the relation between the welding phenomenon and the waveform characteristics so as to be expected to pass through the waveform characteristics and the association relation between the characteristics. Accordingly, the present invention determines a welding quality assessment parameter for achieving a welding quality assessment by comprehensively analyzing at least a portion of a plurality of waveform features of an extracted welding waveform.
According to the invention, the calculation of the welding quality evaluation parameters can be realized according to the waveform characteristics related to the welding voltage waveform and the welding current waveform, and then the welding phenomenon in the current welding process can be directly judged according to the welding quality evaluation parameters.
It should be noted that, according to specific welding requirements and standard requirements, appropriate waveform characteristics may be selected to perform logic calculation to determine different welding quality evaluation parameters. Then, more accurate welding phenomenon evaluation results can be obtained by comprehensively matching welding quality evaluation parameters and part of waveform characteristics.
As an alternative embodiment, the extracted waveform features may include, but are not limited toIPA、IBA、IPV、IBV、F、 Avg_A、Avg_V、IPA_max、IBA_min、R Dry 、Uarc、UextEtc.
Wherein,IPArefers to pulse peak current;IBArefers to a pulse base current;IPVrefers to pulse peak voltage;IBVrefers to a pulse base voltage;Frefers to pulse frequency;Avg_Amean pulse period average current;Avg_Vmean pulse period average voltage;IPA_maxrefers to the peak pulse current maximum;IBA_minrefers to the pulse base current minimum.
Further, a portion of the waveform characteristics may be selected to calculate a welding quality assessment parameter closely related to the welding phenomenon, e.g., based on the pulse peak powerPressingIPVPulse base voltageIBVPeak pulse currentIPAAnd pulse base currentIBACalculating pulse arc length voltage and the likeR Dry 、Uarc、UextAnd welding quality evaluation parameters.
Wherein,R dry Refers to dry elongation resistance;Uarcrefers to pulse arc length voltage;Uextrefers to the dry elongation voltage.
Step 104: and determining the welding phenomenon during arc welding according to the welding quality evaluation parameters and the waveform characteristics and combining the primary recognition result of the welding image during arc welding.
The invention can comprehensively combine the waveform characteristics of the pulse arc length voltage, the pulse dry extension voltage, the pulse period average current and the like which can clearly reflect the welding state and the variation trend and variation of related welding quality evaluation parameters to judge the stability of the welding process and define the variation of the welding process, and different variation corresponds to different welding phenomena.
Fig. 3 is a schematic diagram of information collection related to a welding pool, as shown in fig. 3, a captured welding image may be input into a neural network model trained in advance, so that characteristics of a pool length, a pool width, a pool area, weld seam information (such as pool deviation information) and the like during welding are extracted by using the neural network model, and a primary recognition result is determined by combining the extracted welding direction and welding position. The preliminary identification result may include a result that can be obtained from a welding image, such as normal, penetration, etc.
The information of the molten pool length, the molten pool width, the molten pool area, the molten pool deviation and the like at different sampling moments in the whole sampling period can be determined by welding quality evaluation parameters and waveform characteristics and combining with a preliminary recognition result of a welding image during arc welding, and the information is presented in a signal wave mode.
Fig. 4 is an on-line OK/NG judgment schematic diagram of a welding waveform provided in the present invention, as shown in fig. 4, by describing the above embodiment, a welding waveform (i.e. the voltage in fig. 4) and a welding current waveform (i.e. the current in fig. 4) as well as a puddle length waveform, a puddle width waveform, a puddle area waveform and other signal waveforms that are helpful for welding phenomenon can be obtained, and are displayed in a centralized manner, so that it is advantageous to intuitively and comprehensively judge the welding phenomenon in the sampling period.
Finally, the relevant waveform characteristics of the welding waveform and the welding quality evaluation parameters calculated by the waveform characteristics can be synthesized, and the welding phenomenon during arc welding can be comprehensively determined by combining the primary recognition results of the welding images during arc welding.
It should be noted that, in the welding phenomenon provided by the invention, the analysis of the welding phenomenon is not only performed from the welding waveform, but also more information can be obtained from the welding image by combining the primary recognition result of the welding image during arc welding, so as to further determine the welding phenomenon during arc welding, such as the shape of the welding seam, air holes, surface defects and the like.
In a word, by integrating the welding waveform characteristics, the welding quality evaluation parameters and the preliminary recognition results of the welding images, the welding phenomenon and the welding quality during arc welding can be comprehensively evaluated. This helps to optimize welding parameters, improve weld quality, and provide a basis for quality control.
According to the invention, the characteristics of the welding waveform in the welding process are comprehensively extracted, and the welding quality evaluation parameters for welding phenomenon analysis are determined according to the extracted waveform characteristics, so that the waveform characteristics, the welding quality evaluation parameters and the analysis result of the welding image are synthesized, the welding phenomenon in the arc welding process is comprehensively monitored, the accuracy of the monitoring result can be effectively improved, and a favorable basis is provided for the adjustment of the arc welding process.
Based on the foregoing, as an alternative embodiment, the waveform characteristics include at least one of:
pulse peak currentIPAPulse base currentIBAPeak pulse voltageIPVPulse base voltageIBVPulse frequencyFPulseCycle average currentAvg_AAnd pulse period average voltageAvg_V。
Wherein the peak pulse currentIPAAll peak current values within a pulse period may be identified, extracted by selecting the maximum value thereof or using a peak detection algorithm.
Pulse base currentIBAAll base current values within a pulse period can be calculated, which can be extracted, typically by averaging or using a filter.
Pulse peak voltageIPVAll peak voltage values within a pulse period can be identified, and then extracted by selecting the maximum value thereof or using a peak detection algorithm.
Pulse base voltageIBVIt can be extracted by calculating all base voltages over the pulse period, typically by averaging or using a filter.
Pulse frequencyFThe number of pulses per unit time can be obtained by calculating the reciprocal of the pulse period.
Pulse period average currentAvg_AThe determination may be made by averaging the current over the pulse period.
Pulse period average voltageAvg_VThe average value can be determined by averaging the voltage flow over the pulse period.
In the arc welding quality monitoring method provided by the invention, the acquisition frequency of the welding waveform is generally required to be very high (100K acquisition frequency), so that the acquired welding waveform can play a very good role in early warning, and therefore, important waveform characteristics of the welding waveform are fed back at a high speed in real time, so that the welding phenomenon can be accurately pre-judged in advance.
The invention can obtain the corresponding waveform characteristics by collecting and recording the waveform data of the welding voltage and the welding current in the arc welding process and combining the extraction method. These waveform parameters reflect the characteristics of the pulse during arc welding and the impact on the weld. Analysis of these waveform characteristics may reflect, to some extent, assessing weld quality and optimizing weld parameters.
It should be noted that, in practical applications, the method for extracting waveform features may vary depending on experimental equipment and specific requirements.
As an alternative embodiment, the present invention provides a method of extracting a plurality of waveform features of the welding waveform described above, including, but not limited to:
A real-time welding waveform, including a welding voltage waveform and a welding current waveform, is acquired during a current sampling period.
Determining each waveform key point of the real-time welding waveform; the waveform key points mainly comprise rising edge starting time, rising edge ending time, falling edge starting time and falling edge ending time of each pulse in the welding waveform.
Based on the rising edge start time of each pulse, the pulse period and pulse frequency of the real-time welding waveform are determined, such as calculating the average current in the pulse period from the welding current waveform, and calculating the average voltage from the welding voltage waveform as the pulse period average current and pulse period average voltage.
Then, each pulse peak plateau is determined based on the rising edge start time and the rising edge end time of each pulse to determine a pulse peak current and a pulse peak voltage based on the pulse peak plateau.
Finally, each pulse basis value plateau is determined based on the falling edge start time and the falling edge end time of the pulse to determine the pulse basis value current and the pulse basis value voltage based on the pulse basis value plateau.
Generally, the most important task of extracting waveform features from a welding waveform is to extract key time points, i.e. whether the key time points of the welding waveform are accurately extracted is an important link for accurately extracting the waveform features of the welding waveform.
In the actual operation process, the welding current waveform is more stable than the welding voltage waveform and the welding current waveform and the welding voltage waveform synchronously change, so that the key moment point can be extracted based on the welding current waveform and then applied to the welding voltage waveform.
Further, some critical points of time that need to be extracted are determined mainly by identifying the rising and falling edges of the pulse. The rising edge and the falling edge of each pulse in the welding waveform can be identified through a signal processing algorithm or an edge detection method.
The pulse rising edge and the pulse falling edge in the welding waveform are identified through a signal processing algorithm or an edge detection method, and the following specific implementation modes can be adopted:
1) Edge detection algorithms, commonly used edge detection algorithms include differential algorithms and threshold comparison algorithms. Wherein the differential algorithm comprises: and carrying out differential calculation on two continuous welding waveform data, then judging by setting a threshold value, and indicating that an edge appears when the differential value exceeds the threshold value. A threshold comparison algorithm comprising: the welding waveform data is compared to a predefined threshold and when the waveform value rises or falls past the threshold, an edge is detected. The two algorithms can be selected and adjusted according to the characteristics of the welding waveform and the actual requirements so as to identify the rising edge and the falling edge of the pulse.
2) A smoothing filter method comprising: and filtering the welding waveform data by using a smoothing filter, and determining a key moment point by searching an extreme point or an inflection point of the filtered waveform. The smoothing filter can adopt methods such as moving average filtering, gaussian filtering and the like to carry out smoothing processing on waveform data, and the overall trend of the waveform is extracted. And then searching an extreme point or an inflection point as a rising edge and a falling edge of the pulse according to the change characteristics of the smoothed waveform.
3) A fourier transform method, comprising: the signal is transformed from the time domain to the frequency domain by fourier transforming the welding waveform, and then the main frequency components of the welding waveform are found from the spectral features to determine the rising and falling edges of the individual pulses. For any pulse, obvious harmonic components are shown in the frequency spectrum, and the rising edge and the falling edge of the pulse can be identified according to the occurrence frequency and the amplitude variation of the harmonic.
4) Other methods can be used, including differential operation, waveform profile extraction, etc., to identify rising and falling edges of the waveform, and can be specifically selected according to practical situations.
The above methods and ideas are common, and the actual characteristics of the welding waveform, the influence of noise, and the complexity and performance requirements of the algorithm are considered in actual implementation. For complex signal processing tasks, it may be necessary to combine multiple methods and algorithms, or employ more advanced algorithms such as machine learning for edge detection and feature extraction.
As an alternative embodiment, the present invention also provides a method for simply extracting waveform key points, including but not limited to:
1) Extracting the starting moment of the rising edge of each pulse in the welding waveform mainly based on factors such as the slope, the amplitude, the continuous rising sample number and the like of each pulse;
2) Aiming at the rising edge end time of each pulse in the welding waveform, namely the peak value platform start time, the rising edge end time is mainly extracted through factors such as the slope, the amplitude and the like of each pulse;
3) Aiming at the beginning moment of the falling edge of each pulse in the welding waveform, namely the ending moment of the peak value platform, the method mainly extracts the factors such as the amplitude value of each pulse, the continuous falling sample number and the like;
4) The end time of the falling edge of each pulse in the welding waveform is mainly extracted by factors such as the slope, the amplitude and the like of each pulse.
After the key time points are determined, corresponding waveform characteristics can be determined according to the key time points.
Specifically, for the calculation of the pulse period and the pulse frequency, the pulse period may be determined by performing a time difference calculation for the start time of the rising edge of the adjacent pulse. And then obtaining the pulse frequency according to the reciprocal of the pulse period, namely the number of pulses in unit time.
The time range of each pulse can be defined according to the starting moment of the rising edge of each pulse, and the average value of the current and the average value of the voltage in the time range are calculated to obtain the average current and the average voltage of each pulse in the period and respectively serve as the average current and the average voltage of the pulse period.
The corresponding waveform interval can be determined according to the start time and the end time of the rising edge of each pulse. In this interval, the maximum value of the current and the maximum value of the voltage are found as the pulse peak current and the pulse peak voltage.
The corresponding waveform intervals can be determined according to the start time and the end time of the falling edge of each pulse, and each waveform interval can be regarded as a pulse basic value platform. In this waveform section, the minimum value of the current and the minimum value of the voltage are found as the pulse base current and the pulse base voltage.
Through the steps, the determination of each key time point of the welding waveform can be realized, but the specific implementation mode can be selected and adjusted according to the specific software and hardware platform and algorithm requirements.
Based on the foregoing embodiment, as an alternative embodiment, in a case where it is determined that waveform distortion occurs in the real-time welding waveform acquired in the current sampling period, the following operations are performed:
Replacing the real-time welding waveform by using the historical welding waveform acquired in the previous sampling period of the current sampling period;
the weld quality assessment parameter is determined based on the waveform characteristics of the historical welding waveform.
Specifically, in the case where it is determined that the distortion occurs in the real-time welding waveform in the current sampling period, the following operations may be performed:
firstly, waveform data of each sampling period in the welding process needs to be stored and recorded in advance, and a historical welding waveform of the previous sampling period is obtained from the stored historical data.
Then, the real-time welding waveform distorted in the current sampling period is replaced by the historical welding waveform of the previous sampling period. This eliminates the effect of distortion on subsequent waveform signature analysis and weld quality assessment.
Finally, based on waveform characteristics of the historical welding waveform, determining welding quality evaluation parameters: using methods similar to those previously discussed, the desired waveform characteristics are extracted or calculated from the characteristics of the historical welding waveform.
In particular, real-time monitoring and analysis of the real-time welding waveform during the sampling period is required. When waveform distortion occurs, replacing the real-time waveform of the current sampling period with the historical waveform of the previous sampling period, and calculating corresponding welding quality evaluation parameters. This process may be implemented in a microcontroller or embedded system by real-time sampling, data processing and decision control. At the same time, appropriate judgment conditions and distortion detection algorithms need to be set to trigger the substitution operation and ensure accurate assessment of welding quality.
As an alternative embodiment, the method for determining whether the waveform distortion occurs in the real-time welding waveform may be selected according to the specific application and welding requirements, and one or more of the following methods may be used:
1) The current real-time welding waveform is compared with the previously recorded standard welding waveform by a sample comparison method. If there is obvious difference between the two, it can be judged that the distortion occurs in the real-time welding waveform. This method requires that some standard welding waveforms be pre-recorded, which can be obtained experimentally, for reference.
2) By threshold comparison, a specific threshold is set first, and when the amplitude or other waveform characteristics of the real-time welding waveform exceed or fall below the specific threshold, the real-time welding waveform is considered distorted.
3) And (3) adopting a historical data analysis method, carrying out data analysis by utilizing historical welding waveform data, and establishing a model or a threshold range for judging whether the real-time welding waveform deviates from a normal range. For example, by statistically analyzing characteristic parameters such as the mean value and the spectrum distribution of the welding waveform, when the characteristics of the real-time welding waveform are significantly changed compared with the historical data, it can be determined that the real-time welding waveform is distorted.
4) And analyzing the real-time welding waveform by adopting an abnormality detection algorithm, namely a preset abnormality detection algorithm such as Gaussian mixture model, outlier detection and the like, and judging that the welding waveform is distorted if abnormal data points or abnormal data distribution occur.
5) By analyzing the input-output relationship, it is possible to determine whether the shape of the real-time welding waveform is inconsistent with the expectation, and thus whether waveform distortion occurs, by analyzing the input-output relationship, such as the relationship between current and voltage, in the welding system.
In actual operation, the method can be selected according to specific welding systems and requirements, and whether waveform distortion occurs in the real-time welding waveform can be judged by combining experience and knowledge in actual application.
According to the arc welding quality monitoring method provided by the invention, the abnormal condition of the current period can be eliminated and the reliability of data can be improved by replacing the distorted real-time welding waveform in the current sampling period and analyzing and evaluating the distorted real-time welding waveform in the previous sampling period by utilizing the historical waveform of the previous sampling period. In addition, the distorted waveform may mean that problems occur in the welding process, such as material changes, gun drift, etc. By replacing in real time and evaluating based on the historical waveform, feedback adjustment can be found and performed in time so as to ensure the stability and consistency of welding quality. When the historical waveform is adopted for evaluation, more comprehensive and accurate waveform characteristics and welding quality evaluation parameters can be obtained based on longer data histories. This allows for a more efficient analysis of potential problems in the welding process and provides an accurate reference for further optimization. And, since the historical waveform data is to be saved as compared to the current waveform data, the likelihood of backup and trend analysis of the welding process is increased. Through statistics and analysis of historical data, more comprehensive welding quality information can be obtained, and deeper data mining and optimization can be performed.
In summary, by replacing the distorted real-time welding waveform in the current period with the historical waveform of the previous sampling period, the reliability, real-time feedback and adjustment of the data can be improved, the evaluation accuracy can be improved, and more opportunities for data backup and analysis can be provided to realize a more stable and high-quality welding process.
Based on the foregoing embodiments, as an alternative embodiment, the determining the pulse peak current and the pulse peak voltage based on the pulse peak platform includes:
determining the pulse peak current and the pulse peak voltage based on waveforms near a start point, an end point, and a middle point of the pulse peak plateau;
alternatively, the pulse peak current is determined based on a current average of all waveforms of the pulse peak plateau, and the pulse peak voltage is determined based on a voltage average of all waveforms of the pulse peak plateau;
alternatively, the pulse peak current is determined based on a current median of all waveforms of the pulse peak plateau, and the pulse peak voltage is determined based on a voltage median of all waveforms of the pulse peak plateau.
The present invention provides various methods of determining a pulse peak current and the pulse peak voltage.
In one aspect, determining the pulse peak current and the pulse peak voltage based on waveforms near a start point, an end point, and a middle point of the pulse peak plateau may employ the steps of:
first, the positions of the start point and the end point of the pulse peak platform can be determined according to the characteristics of specific pulses. The start point of the pulse peak plateau is typically the start position of the rising edge of the pulse and the end point of the pulse peak plateau is typically the end position of the falling edge of the pulse. The starting point and the ending point may be found using the edge detection algorithm or other feature extraction methods provided in the above embodiments, which are not described herein.
Further, by the start point and the end point, the position of the pulse peak platform is finally determined, and any point on the pulse peak platform can be taken as an intermediate point.
Then, waveform data in the vicinity of the position of the pulse peak stage is extracted from the position. A fixed window size may be set or dynamically selected according to specific needs.
Finally, analyzing the extracted waveform data to determine a pulse peak current and a pulse peak voltage, including: finding the maximum value of the current value in the welding current waveform near the pulse peak value platform, and determining the pulse peak value current; and finding the maximum value of the voltage value in the welding voltage waveform near the pulse peak value platform, and determining the pulse peak value voltage.
It should be noted that the specific operation is to be adjusted according to the actual situation and signal characteristics. The sampling rate, noise level, sharpness of rising and falling edges of the welding waveform, etc. may all have an impact on the result. Therefore, in practical application, proper algorithm and parameters can be selected according to specific situations, and proper calibration and verification can be performed to ensure the accuracy and reliability of the result.
On the other hand, when determining the pulse peak current based on the current average value of all waveforms of the pulse peak stage and determining the pulse peak voltage based on the voltage average value of all waveforms of the pulse peak stage, the following steps may be performed:
first, for each pulse waveform, a start point and an end point of a pulse peak plateau are determined. The start point of the pulse peak plateau may be the start position of the rising edge of the pulse and the end point of the pulse peak plateau may be the end position of the falling edge of the pulse. An edge detection algorithm or other feature extraction method may be employed to determine the start and end points.
And extracting waveform data of the whole pulse peak value platform range according to the determined starting point and ending point of the pulse peak value platform.
And sampling the waveform data of each pulse peak value platform, and calculating the average value of the obtained sampling values to obtain the current average value of the pulse peak value platform.
The above steps are repeated, but the object of this calculation is the average of the voltages in the welding voltage waveform. The voltage average of the pulse peak plateau can be obtained.
The calculation of the pulse peak current and the pulse peak voltage may also be performed in another manner, including:
1) For each pulse of each welding waveform, a start point and an end point of a pulse peak plateau are determined. The start point may be the start position of the rising edge of the pulse, the end point may be the end position of the falling edge of the pulse, and an edge detection algorithm or other feature extraction method may be used to determine the start and end points of the pulse peak plateau.
2) And extracting waveform data of the whole pulse peak value platform range according to the determined starting point and ending point of the pulse peak value platform.
3) The waveform data of each pulse is sampled, and the sampled currents are arranged in ascending order.
4) The median of the aligned current values is found. This will give the current median of the pulse peak plateau, which is used as an estimate of the pulse peak current.
Repeating steps 3) and 4), but the object of this calculation is the voltage median of the welding voltage waveform voltage. This will give the voltage median of the pulse peak plateau, which is used as an estimate of the pulse peak voltage.
It should be noted that in order to accurately calculate the median, it is necessary to ensure that the waveform data is sampled sufficiently to cover the entire pulse peak plateau. Furthermore, in order to reduce the effect of noise, filtering or other signal processing operations may be performed prior to calculating the median.
The invention provides a plurality of methods for determining the pulse peak current and the pulse peak voltage, which can be selected according to specific situations and system requirements in practical application. By considering the waveform data of the entire pulse peak plateau, these methods can provide more comprehensive information to estimate pulse peak current and voltage rather than relying on just a single sampling point. By adopting a plurality of waveforms or data samples, the methods can reduce the influence of noise to a certain extent and improve the estimation accuracy and stability of peak current and voltage. By extracting and analyzing waveform data of the whole pulse peak value platform, visual visualization can be carried out on characteristics of pulse signals, understanding and explaining of the properties and the characteristics of the signals are facilitated, and the method can be combined with other signal processing technologies, algorithms and models, so that accuracy and reliability of estimating pulse peak value current and pulse peak value voltage are further enhanced.
Based on the foregoing, as an alternative embodiment, the welding quality assessment parameter may include a pulse arc length voltage; the determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform characteristics includes:
determining a dry extension resistance based on the pulse peak voltage, the pulse base voltage, the pulse peak current, and the pulse base current;
determining a dry extension voltage based on the pulse peak current and the dry extension resistance;
the pulse arc length voltage is determined based on the pulse peak voltage and the dry stretch voltage.
Dry extension resistorR Dry The resistance value of the extension dry extension is mainly used for representing the resistance value of the extension dry extension during welding, the extension dry extension resistance is larger to show that the extension dry extension is longer, the extension dry extension is shorter, and the specific calculation formula is as follows:
R dry =（IPV-IBV）/(IPA-IBA)；
Wherein,IPVin the event of a pulse peak voltage,IBVfor the voltage of the base value of the pulse,IPAin the event of a pulse peak current,IBAis a pulse base current.
And dry elongation resistanceR Dry Characterization was substantially consistent, dry elongation voltageUextThe change of the dry elongation is characterized, the increase of the value indicates the longer dry elongation, the decrease of the value indicates the shorter dry elongation, and the specific calculation formula is as follows:
Uext=((IPV-IBV）/(IPA-IBA))*IPA；
pulse arc length voltage UaThe method is mainly used for representing the change of the arc length, and when the value is larger, the arc length is longer, and when the value is smaller, the arc length is shorter. Too long an arc length results in poor welding formation, while too short an arc length results in a large spatter during welding.
According to the invention, the pulse arc length voltage is comprehensively calculated as the welding quality evaluation parameter by analyzing the waveform characteristics, so that the method is used for realizing the evaluation of welding phenomena by combining part of waveform characteristics and the primary recognition result of welding images, and compared with the method which directly analyzes the waveform characteristics or compared with other welding parameters, the method can directly evaluate the stability and consistency of the welding quality by analyzing the change of the pulse arc length voltage because the pulse arc length voltage is one of important parameters in the arc welding process. Different welding defects and adverse phenomena (such as welding wire splashing, poor weld joint shape, air holes and the like) can cause abnormal changes of the pulse arc length voltage, so that the welding defects can be detected by monitoring the pulse arc length voltage, and the welding process can be controlled and optimized in real time by monitoring and analyzing the pulse arc length voltage.
In the actual detection process, when welding is performed using different welding power sources, there are different arc feedback control modes. Generally, two current arc feedback control methods include: pulse frequency modulation feedback (PFM feedback for short) and pulse amplitude adjustment feedback (PFM-PAM hybrid feedback for short) are combined with pulse frequency modulation. The PFM-PAM mixed feedback refers to that PFM and PAM are applied to different feedback links so as to realize accurate control of output voltage or current. By PFM-PAM mixed feedback, higher control accuracy and response speed can be provided.
FIG. 5 is a schematic diagram showing waveform characteristics under PFM feedback according to the present invention, and in combination with FIG. 5, when the PFM feedback control scheme is used, the pulse frequency is the apparent variation in waveform characteristics during weldingFWhile the peak current of the pulseIPAAnd pulse base currentIBAAll are unchanged. In addition, with pulse frequencyFVariation of pulse period average currentAvg_AAnd pulse period average voltageAvg_VIs changed with follow-up, and the pulse frequency in waveform characteristics is mainly monitored when welding phenomenon analysis is carried out for the PFM feedback control mode under the power supplyFAverage current of pulse periodAvg_AAnd pulse period average voltageAvg_VAnd welding quality assessment parameter pulse arc length voltageUaEtc.
Here, CH1 and CH2 in fig. 5 refer to a welding voltage waveform and a welding current waveform, respectively, with the difference that the welding voltage waveform CH1 and the welding current waveform CH2 in the lower half are amplification processing of the welding voltage waveform CH1 and the welding current waveform CH2 in the upper half. For example, the welding voltage waveform CH1 and the welding current waveform CH2 in the upper half are acquired in units of seconds, and the welding voltage waveform CH1 and the welding current waveform CH2 in the lower half are amplified in units of milliseconds.
FIG. 6 is a schematic diagram of waveform characteristics under the PFM-PAM mixed feedback provided by the present invention, and in combination with the control mode of FIG. 6, the control mode of the PFM-PAM mixed feedback is shown, since the obvious variation in waveform characteristics includes pulse frequencyFPeak pulse currentIPAPulse base currentIBAPulse peak voltageIPVPulse base voltageIBV. While pulse arc length voltage is estimated due to welding qualityUaIs the main variable, the pulse peak currentIPAPulse base currentIBAPulse peak voltageIPVPulse base voltageIBVAll followUaIs changed by a change in (a). Therefore, it is more desirable to monitor the variation of the reference pulse arc length voltage at the weld quality.
Here, CH1 and CH2 in fig. 6 also refer to a welding voltage waveform and a welding current waveform, respectively, and are also different in that the welding voltage waveform and the welding current waveform in the lower half are amplification processing of the welding voltage waveform CH1 and the welding current waveform CH2 in the upper half.
Fig. 7 is a schematic diagram of a change in welding phenomenon corresponding to a change in welding waveform provided by the present invention, as shown in fig. 7, on the one hand, it can be seen that the welding waveform changes with fluctuation in length of dry extension, and the arc length is essentially adjusted, so that the control object is the welding waveform, and the object of the substantial control is the arc length. On the other hand, the welding waveform changes with fluctuation of the wire feed length (i.e., the height of the welding gun), and the arc length is essentially adjusted, so that the control object is the welding waveform, and the actual control object is the arc length.
In fig. 7, the fluctuation of the wire feed length (i.e., the height change of the welding gun) is represented by a bold curve, while the correlation between the fluctuation of the wire feed length (i.e., the height change of the welding gun) and the welding voltage waveform and the welding current waveform is represented by a bold arrow.
In summary, the change of the welding waveform is basically as shown in fig. 7, regardless of the active change of the dry elongation or the arc length change along with the wire feeding (the phenomena of wire clamping, wire feeding failure, and passive lengthening of the arc length occur when the wire feeding resistance is different).
When the welding gun is fixed, the change of the welding waveform is mostly caused by unsmooth wire feeding, so if the on-site welding waveform fluctuates severely in a short time, the wire feeding is considered to be unsmooth; if the arc length of the arc is almost unchanged in the whole welding process, but the welding waveform changes in a large wave shape in the whole welding process, the arc length is considered to be caused by out-of-round barrel.
Based on the principle of the phenomenon, the invention aims to closely relate the welding phenomenon with different waveform characteristics and actual values of welding quality evaluation parameters by researching the relevance of the welding quality evaluation parameters to welding phenomenon quality inspection, and provides a method capable of realizing accurate welding quality monitoring.
Fig. 8 is a schematic flow chart of the welding phenomenon determination logic provided in the present invention, as shown in fig. 8, according to the welding quality evaluation parameter and the waveform feature, and in combination with the primary recognition result of the welding image during arc welding, determining the welding phenomenon during arc welding mainly includes:
determining a pulse average voltage variation based on an average value of the average voltage of the continuous plurality of pulse periods;
determining a pulse average current variation based on an average value of the average current of a plurality of continuous pulse periods;
when the preliminary identification result is determined to be the penetration, the pulse average voltage variation is greater than or equal to a first preset threshold (hereinafter referred to as t 1 ) And the pulse average current variation is greater than or equal to a second preset threshold (hereinafter referred to as t 2 ) In the case of (2), the welding phenomenon at the time of arc welding is determined to be penetration.
Specifically, the above calculation formula for determining the pulse average voltage variation based on the average value of the average voltage of the continuous plural pulse periods can be expressed as:
；
wherein,refers tonAverage voltage of each pulse period +.>Refers ton-average voltage of 1 pulse period>Represent the firstnAverage voltage variation value of each pulse period; / >The reference value representing the average voltage of each pulse period can be obtained experimentally, then +.>The mean voltage variation of the pulses is shown as a percentage of the data.
Similarly, a calculation formula for determining the pulse average current variation based on the average value of the average current of a plurality of continuous pulse periods can be expressed as:
；/>
wherein,refers tonAverage current for each pulse period, +.>Refers ton-average current of 1 pulse period>Represent the firstnAverage current variation value for each pulse period; />The reference value representing the average current of each pulse period can be obtained experimentally, then +.>The mean current variation of the pulse is shown as a percentage of the system data.
Further, if the first preset threshold is about 20% and the second preset threshold is about 20%, then in determining、And under the condition that the welding phenomenon determined by the preliminary identification result of the welding image is the welding penetration, the welding phenomenon currently performed in arc welding can be finally determined to be the welding penetration by combining the pulse average voltage variation and the pulse average current variation.
In order to avoid the influence of random fluctuation on the final detection result during detection, the 5 continuous pulse period waveform data can be collected and analyzed, and if at least two of the welding phenomena correspondingly identified in the pulse period are welding through, the welding phenomenon currently in arc welding can be finally determined to be the welding through.
Further, if it is determined that any one of the above conditions is not satisfied, it cannot be finally determined that the welding phenomenon at the time of arc welding is penetration.
In addition, the condition in "judging whether or not the foregoing conditions are all satisfied" shown in fig. 8 means that it is noted in the immediately preceding step: the average voltage variation is greater than or equal to a first preset threshold t 1 The average current variation is greater than or equal to a second preset threshold t 2 And the preliminary recognition result through the image is a weld-through.
FIG. 9 is a second flow chart of the welding phenomenon determination logic according to the present invention, as shown in FIG. 9, in determining the welding phenomenonThe preliminary identification result is necking, and the average voltage variation of the pulseGreater than or equal to said first preset threshold value and said pulse average current variation +.>Greater than or equal to a third preset threshold (hereinafter referred to as t 3 ) In the case of (2), the welding phenomenon at the time of arc welding can be finally determined to be necking.
It should be noted that the third preset threshold at this time is smaller than the second preset threshold mentioned in the above embodiment.
Assuming that the third preset threshold is about 10%, the first preset threshold and the second preset threshold remain unchanged, both being about 20%. Then at the determination of 、/>And under the condition that the welding phenomenon determined by the preliminary identification result of the welding image is necking, the welding phenomenon currently performed in arc welding can be finally determined to be necking by combining the pulse average voltage variation and the pulse average current variation.
Similarly, at this time, the waveform data of 5 continuous pulse periods may be collected and analyzed, and if at least two of the pulse periods correspond to the identified welding phenomenon as necking, the welding phenomenon currently performed in arc welding may be finally determined as necking.
The condition in "judging whether the preceding conditions are all satisfied" shown in fig. 9 means that it is noted in the immediately preceding step: "the average voltage variation is equal to or greater than the first preset threshold t1, the average current variation is equal to or greater than the second preset threshold t2, and the preliminary recognition result of the image is the weld through", and the following steps are noted: the average voltage variation is larger than or equal to a first preset threshold t1, the average current variation is larger than or equal to a third preset threshold t3, and the preliminary recognition result of the image is necking.
Fig. 10 is a third flow chart of the welding phenomenon determination logic provided in the present invention, and the determination of whether the previous conditions are satisfied in fig. 10 refers to the determination of whether the conditions in the immediately preceding step are satisfied. If the condition that the relevant condition finally determined as necking is not met is determined, determining the pulse arc length voltage variation;
And determining pulse arc length voltage change ratios of the pulse periods according to the pulse arc length voltage change amounts, wherein the pulse arc length voltage change ratios are used for representing the change amplitude of the pulse arc length voltage in the continuous pulse periods.
Upon determining that the preliminary identification result is normal, the pulse average voltage variation is equal to or greater than the fourth preset threshold (hereinafter referred to as t 4 ) But less than or equal to the first preset threshold, the pulse average current variation is less than or equal to the third preset threshold but greater than or equal to a fifth preset threshold (hereinafter referred to as t) 5 ) And the pulse arc length voltage variation ratio is greater than or equal to the sixth preset threshold (hereinafter referred to as t 6 ) But less than or equal to the seventh preset threshold (hereinafter referred to as t 7 ) In the case of (2), the welding phenomenon at the time of arc welding is determined to be arc jumping instability.
Wherein the pulse arc length voltage change ratioPulse arc length voltage based on a continuous number of pulse periods +.>The calculation formula is determined as follows:
；
wherein,is the firstnPulse arc length voltage of one pulse period, +.>The reference value representing the pulse arc length voltage of each pulse period can be obtained experimentally.
After determining that the fourth preset threshold is about 10%, the fifth preset threshold is about 5%. If the identification result of the welding image is normal, the pulse arc length voltage change ratio is determined to be about 5% in the case where the sixth preset threshold is about 10% in the seventh preset thresholdIs in the range of +.>Pulse average voltage variation->Is in the range of +.>And the range of the pulse average current variation is +.>The welding phenomenon at the time of arc welding can be determined to be arc jumping instability.
Similarly, at this time, the waveform data of 5 continuous pulse periods can be collected and analyzed, and if at least two of the pulse periods correspond to and identify that the welding phenomenon is arc jumping instability, the welding phenomenon currently performed in arc welding can be finally determined to be arc jumping instability.
Fig. 11 is a schematic flow chart of the welding phenomenon determination logic provided in the present invention, and the determination of whether the foregoing conditions are satisfied in fig. 11 refers to the determination of whether the conditions in the immediately preceding step are satisfied. On the basis of the above embodiment, in the case where it is determined that the condition related to the final determination of the arc jump instability is not currently satisfied based on the current welding quality evaluation parameter, the waveform characteristic, and the preliminary recognition result of the welding image at the time of arc welding, if it is determined that the preliminary recognition result is normal, the pulse average voltage variation amount Is larger than or equal to the first preset threshold value of 20%, and the pulse average current variation is +.>If the welding speed is greater than or equal to 10% of the third preset threshold, it can be judged that the welding phenomenon in the current arc welding is not only arc jumping instability, but also the essential reason is unstable wire feeding.
Similarly, at this time, the waveform data of 5 continuous pulse periods can be collected and analyzed, and if at least two of the pulse periods correspond to and identify that the welding phenomenon is unstable in wire feeding, the welding phenomenon currently performed in arc welding can be finally determined to be unstable in wire feeding.
Fig. 12 is a schematic flow chart of the welding phenomenon determination logic provided in the present invention, and the determination of whether the foregoing conditions are satisfied in fig. 12 refers to the determination of whether the conditions in the immediately preceding step are satisfied. In the case that it is determined that the relevant condition, which is finally determined as unstable wire feeding, is not satisfied at present based on the current welding quality evaluation parameter, waveform characteristics, and preliminary recognition result of the welding image at the time of arc welding, it is necessary to further determine the average voltage variation according to the pulsePulse average current variation->Pulse arc length voltage variation ratio +.>Further judging the correlation calculated value of the (C).
Setting pulse arc length voltage change ratioThe related eighth threshold value is 25%, if the primary identification result is normal, the pulse average voltage variation is +.>Pulse average current variation->And pulse arc length voltage variation ratio +.>If the welding phenomenon is greater than or equal to the eighth threshold value by 25%, the welding phenomenon at the current arc welding can be determined to be out of round.
When it is determined that the welding phenomenon at the time of the current arc welding is a cylinder non-circular, it is preferable to comprehensively analyze a plurality of continuous pulse period waveform data. For example, if the determination results of the waveform data of 5 consecutive pulse periods are all the cylinder out-of-round, it is determined that the welding phenomenon finally determined is the cylinder out-of-round.
On the basis of the embodiment, if it cannot be determined that the welding phenomenon is any one of the states of penetration, necking, unstable arc jumping, unstable wire feeding, out-of-round barrel and the like, the steps are iteratively executed, and the next round of detection is performed again.
The method for monitoring the welding quality of the arc welding continuously detects the welding quality of the arc welding, if the welding phenomenon is detected to be any one of the states of welding through, necking, unstable arc jumping, unstable wire feeding, non-round barrel and the like in the one-round detection process, the machine is required to be stopped for adjustment, and if the welding phenomenon is not detected, the current welding is normal, the iteration is continued for monitoring.
According to the arc welding quality monitoring method provided by the invention, through deep analysis of welding phenomena, a welding image is used as one judging signal source, welding quality evaluation parameters and waveform characteristics are used as another judging signal source, different priority levels are respectively set, different welding phenomena are defined, the priority judging signal sources of different welding phenomena are defined, the judging signal source with high priority is used as a main judging basis, and the other judging signal source is used as an auxiliary judging basis. Meanwhile, the defects that waveform characteristics are not fully extracted and the relation between welding phenomenon and waveform characteristics and welding quality evaluation parameters is not clear in the prior art are overcome, and the accuracy of electric arc welding quality monitoring is effectively improved.
As an alternative embodiment, the preliminary recognition result of the welding image at the time of the arc welding is determined based on the steps of:
acquiring a welding image during the arc welding;
and inputting the welding image into a pre-trained welding quality detection model to acquire the preliminary identification result output by the welding quality detection model.
Specifically, a weld image is first acquired, including acquiring an image of a weld during welding using a suitable image acquisition device (e.g., a camera). This may be accomplished by placing the camera near the weld area or by using a monitoring system during the welding process.
The acquired welding images may be pre-processed to extract features useful for subsequent processing. The step of preprocessing may include image denoising, image enhancement, contrast adjustment, or edge detection, etc.
And inputting the preprocessed welding image into a pre-trained welding quality detection model for processing. This model may be a machine learning or deep learning based algorithm such as convolutional neural network model (Convolutional Neural Network, CNN), recurrent neural network model (Recurrent Neural Network, RNN), long Short-Term Memory network model (LSTM), generate countermeasure network model (Generative Adversarial Network, GAN), transformer network model (transducer) or deep neural network model (Deep Neural Network, DNN), etc. The convolutional neural network model has been trained on a large amount of supply data to enable accurate analysis and identification of weld quality.
And analyzing the input welding image according to the pre-trained convolutional neural network model and outputting a preliminary recognition result. The results may include an assessment of weld quality, such as weld penetration, necking, arc run-out instability, wire feed instability, or barrel out of round, etc.
Fig. 13 is a schematic structural view of an arc welding quality monitoring device provided by the present invention, and as shown in fig. 13, the present invention provides an arc welding quality monitoring device, which mainly includes: a welding information acquisition unit 131, a welding information analysis unit 132, a welding information processing unit 133, and a welding phenomenon determination unit 134.
The welding information acquisition unit 131 is mainly used for acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms;
the welding information analysis unit 132 is mainly used for extracting a plurality of waveform characteristics of the welding waveform;
the welding information processing unit 133 is mainly configured to determine at least one welding quality evaluation parameter based on at least a part of the plurality of waveform characteristics;
the welding phenomenon determination unit 134 is mainly configured to determine a welding phenomenon at the time of the arc welding based on the welding quality evaluation parameter and the waveform feature in combination with a preliminary recognition result of a welding image at the time of the arc welding.
It should be noted that, when the arc welding quality monitoring device provided in the embodiment of the present invention specifically operates, the arc welding quality monitoring method described in any one of the above embodiments may be executed, which is not described in detail in this embodiment.
As an optional embodiment, the arc welding quality monitoring device provided by the invention can also comprise a man-machine display unit, and the acquired welding voltage waveform, welding current waveform, welding image, preliminary recognition result related to the welding image and the like can be displayed on a display interface of the man-machine display unit in real time, so that real-time judgment and real-time update can be realized. Correspondingly, the welding phenomenon detected in real time can be marked in real time.
In addition, the device for monitoring the welding quality of the arc welding provided by the invention can also comprise a storage unit for counting and storing the real-time information such as the welding voltage waveform, the welding current waveform, the welding image, the preliminary identification result, the related welding phenomenon and the like acquired in real time for tracing.
According to the arc welding quality monitoring device provided by the invention, the characteristics of the welding waveform in the welding process are comprehensively extracted, and the welding quality evaluation parameters for welding phenomenon analysis are determined according to the extracted waveform characteristics, so that the waveform characteristics, the welding quality evaluation parameters and the analysis results of welding images are synthesized, the welding phenomenon in the arc welding process is comprehensively monitored, the accuracy of the monitoring results can be effectively improved, and an advantageous basis is provided for the adjustment of the arc welding process.
Fig. 14 is a schematic structural diagram of an electronic device according to the present invention, and as shown in fig. 14, the electronic device may include: processor 141, communication interface (Communications Interface) 142, memory 143 and communication bus 144, wherein processor 141, communication interface 142, memory 143 complete communication with each other through communication bus 144. Processor 141 may invoke logic instructions in memory 143 to perform an arc welding quality monitoring method comprising: acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms; extracting a plurality of waveform features of the welding waveform; determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform features; and determining the welding phenomenon during arc welding according to the welding quality evaluation parameters and the waveform characteristics and combining the primary recognition result of the welding image during arc welding.
Further, the logic instructions in the memory 143 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the arc welding quality monitoring method provided by the above embodiments, the method comprising: acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms; extracting a plurality of waveform features of the welding waveform; determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform features; and determining the welding phenomenon during arc welding according to the welding quality evaluation parameters and the waveform characteristics and combining the primary recognition result of the welding image during arc welding.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the arc welding quality monitoring method provided by the above embodiments, the method comprising: acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms; extracting a plurality of waveform features of the welding waveform; determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform features; and determining the welding phenomenon during arc welding according to the welding quality evaluation parameters and the waveform characteristics and combining the primary recognition result of the welding image during arc welding.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (16)
1. A method of monitoring the quality of an arc welding process, comprising:
acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms;
extracting a plurality of waveform features of the welding waveform;
determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform features;
and determining the welding phenomenon during arc welding according to the welding quality evaluation parameters and the waveform characteristics and combining the primary recognition result of the welding image during arc welding.
2. The arc welding quality monitoring method of claim 1, wherein the waveform characteristics comprise at least one of:
Pulse peak current, pulse base current, pulse peak voltage, pulse base voltage, pulse frequency, pulse period average current, and pulse period average voltage.
3. The arc welding quality monitoring method of claim 2, wherein the extracting the plurality of waveform features of the welding waveform comprises:
acquiring a real-time welding waveform in a current sampling period;
determining each waveform key point of the real-time welding waveform; the waveform key points comprise rising edge starting time, rising edge ending time, falling edge starting time and falling edge ending time of each pulse in the welding waveform;
determining a pulse period and a pulse frequency of the real-time welding waveform based on the rising edge starting time of each pulse so as to calculate average current and average voltage in the pulse period as the average current and average voltage of the pulse period;
determining each pulse peak plateau based on a rising edge start time and a rising edge end time of each pulse to determine the pulse peak current and the pulse peak voltage based on the pulse peak plateau;
each pulse base plateau is determined based on a falling edge start time and a falling edge end time of a pulse to determine the pulse base current and the pulse base voltage based on the pulse base plateau.
4. The arc welding quality monitoring method according to claim 3, wherein in the event that waveform distortion of the real-time welding waveform acquired during the current sampling period is determined, the following operations are performed:
replacing the real-time welding waveform by using the historical welding waveform acquired in the previous sampling period of the current sampling period;
the weld quality assessment parameter is determined based on the waveform characteristics of the historical welding waveform.
5. The arc welding quality monitoring method of claim 3 wherein the determining the pulse peak current and the pulse peak voltage based on the pulse peak plateau comprises:
determining the pulse peak current and the pulse peak voltage based on waveforms near a start point, an end point, and a middle point of the pulse peak plateau;
alternatively, the pulse peak current is determined based on a current average of all waveforms of the pulse peak plateau, and the pulse peak voltage is determined based on a voltage average of all waveforms of the pulse peak plateau;
alternatively, the pulse peak current is determined based on a current median of all waveforms of the pulse peak plateau, and the pulse peak voltage is determined based on a voltage median of all waveforms of the pulse peak plateau.
6. The arc welding quality monitoring method of any of claims 3-5, wherein the determining the pulse base current and the pulse base voltage based on the pulse base plateau comprises:
determining the pulse base current and the pulse base voltage based on waveforms near a start point and an end point of the pulse base plateau;
or, based on the maximum probability value of all sampling points of the pulse base value platform, the current and the voltage of the sampling point corresponding to the maximum probability value are used as the pulse base value current and the pulse base value voltage.
7. The arc welding quality monitoring method of claim 2 wherein the welding quality assessment parameter comprises a pulsed arc length voltage;
the determining at least one weld quality assessment parameter based on at least a portion of the plurality of waveform characteristics includes:
determining a dry extension resistance based on the pulse peak voltage, the pulse base voltage, the pulse peak current, and the pulse base current;
determining a dry extension voltage based on the pulse peak current and the dry extension resistance;
the pulse arc length voltage is determined based on the pulse peak voltage and the dry stretch voltage.
8. The arc welding quality monitoring method according to claim 7, wherein the determining the welding phenomenon at the time of the arc welding based on the welding quality evaluation parameter and the waveform characteristic in combination with the preliminary recognition result of the welding image at the time of the arc welding includes:
determining a pulse average voltage variation based on an average value of the average voltage of the continuous plurality of pulse periods;
determining a pulse average current variation based on an average value of the average current of a plurality of continuous pulse periods;
and determining that the welding phenomenon is welding penetration when the primary identification result is determined to be welding penetration, wherein the pulse average voltage variation is larger than or equal to a first preset threshold value, and the pulse average current variation is larger than or equal to a second preset threshold value.
9. The arc welding quality monitoring method of claim 8, further comprising:
determining that the welding phenomenon is necking when the electric arc welding is performed under the condition that the preliminary identification result is necking, the pulse average voltage variation is larger than or equal to the first preset threshold value, and the pulse average current variation is larger than or equal to a third preset threshold value;
The third preset threshold is smaller than the second preset threshold.
10. The arc welding quality monitoring method of claim 9 further comprising determining a pulse arc length voltage variation, the pulse arc length voltage variation determined based on a pulse arc length voltage of a continuous plurality of pulse cycles;
determining a pulse arc length voltage variation ratio of the plurality of pulse periods according to the pulse arc length voltage variation, wherein the pulse arc length voltage variation ratio is used for representing the variation amplitude of the pulse arc length voltage in the continuous plurality of pulse periods;
and determining that the welding phenomenon is arc jumping instability when the pulse arc length voltage change ratio is larger than or equal to a sixth preset threshold value but smaller than or equal to a seventh preset threshold value under the condition that the preliminary identification result is determined to be normal, the pulse average voltage change is larger than or equal to a fourth preset threshold value but smaller than or equal to the first preset threshold value, the pulse average current change is smaller than or equal to the third preset threshold value but larger than or equal to a fifth preset threshold value, and the pulse arc length voltage change ratio is larger than or equal to the sixth preset threshold value but smaller than or equal to the seventh preset threshold value.
11. The arc welding quality monitoring method of claim 10, further comprising:
And determining that the welding phenomenon is unstable when the arc welding is performed under the condition that the preliminary identification result is normal, the pulse average voltage variation is larger than or equal to the first preset threshold value, and the pulse average current variation is larger than or equal to a third preset threshold value.
12. The arc welding quality monitoring method of claim 10, further comprising:
and determining that the welding phenomenon is a cylinder out of round when the preliminary identification result is normal, the pulse average voltage variation is larger than or equal to the first preset threshold, the pulse average current variation is larger than or equal to a third preset threshold and the pulse arc length voltage variation ratio is larger than or equal to an eighth threshold.
13. The arc welding quality monitoring method according to claim 1, wherein the preliminary recognition result of a welding image at the time of the arc welding is determined based on the steps of:
acquiring a welding image during the arc welding;
and inputting the welding image into a pre-trained welding quality detection model to acquire the preliminary identification result output by the welding quality detection model.
14. An arc welding quality monitoring device, comprising:
the welding information acquisition unit is used for acquiring welding waveforms during arc welding, wherein the welding waveforms comprise welding voltage waveforms and welding current waveforms;
a welding information analysis unit for extracting a plurality of waveform features of the welding waveform;
a welding information processing unit for determining at least one welding quality assessment parameter based on at least a portion of the plurality of waveform features;
and the welding phenomenon determining unit is used for determining the welding phenomenon during the arc welding according to the welding quality evaluation parameter and the waveform characteristic and combining the primary recognition result of the welding image during the arc welding.
15. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the arc welding quality monitoring method of any one of claims 1 to 13.
16. A non-transitory computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of the arc welding quality monitoring method according to any of claims 1 to 13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311381336.2A CN117102630B (en) | 2023-10-24 | 2023-10-24 | Arc welding quality monitoring method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311381336.2A CN117102630B (en) | 2023-10-24 | 2023-10-24 | Arc welding quality monitoring method, device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117102630A true CN117102630A (en) | 2023-11-24 |
| CN117102630B CN117102630B (en) | 2024-02-27 |
Family
ID=88798705
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311381336.2A Active CN117102630B (en) | 2023-10-24 | 2023-10-24 | Arc welding quality monitoring method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117102630B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102596476A (en) * | 2009-11-13 | 2012-07-18 | 林肯环球股份有限公司 | Method and apparatus for monitoring weld quality |
| US20150041449A1 (en) * | 2012-06-18 | 2015-02-12 | Panasonic Intellectual Property Management Co., Ltd. | Arc-welding method and arc-welding apparatus |
| CN109741297A (en) * | 2018-11-28 | 2019-05-10 | 歌尔股份有限公司 | Product quality detection method and device |
| CN113920060A (en) * | 2021-09-09 | 2022-01-11 | 中国科学院自动化研究所 | Autonomous operation method and device for welding robot, electronic device, and storage medium |
| CN114559177A (en) * | 2022-03-07 | 2022-05-31 | 北京洞微科技发展有限公司 | Welding evaluation method and device based on image data analysis and storage medium |
-
2023
- 2023-10-24 CN CN202311381336.2A patent/CN117102630B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102596476A (en) * | 2009-11-13 | 2012-07-18 | 林肯环球股份有限公司 | Method and apparatus for monitoring weld quality |
| US20150041449A1 (en) * | 2012-06-18 | 2015-02-12 | Panasonic Intellectual Property Management Co., Ltd. | Arc-welding method and arc-welding apparatus |
| CN109741297A (en) * | 2018-11-28 | 2019-05-10 | 歌尔股份有限公司 | Product quality detection method and device |
| CN113920060A (en) * | 2021-09-09 | 2022-01-11 | 中国科学院自动化研究所 | Autonomous operation method and device for welding robot, electronic device, and storage medium |
| CN114559177A (en) * | 2022-03-07 | 2022-05-31 | 北京洞微科技发展有限公司 | Welding evaluation method and device based on image data analysis and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117102630B (en) | 2024-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11548088B2 (en) | Systems and methods for welding torch weaving | |
| JP5632009B2 (en) | Method and apparatus for monitoring welding quality | |
| US6441342B1 (en) | Monitor for electric arc welder | |
| DE69818275T2 (en) | QUALITY MEASUREMENT OF WELDING | |
| CN111354366B (en) | Abnormal sound detection method and abnormal sound detection device | |
| CN103357987B (en) | Automatic stability detecting method for process of CO2 electric arc welding short circuit transition welding | |
| CN105209993B (en) | System and method for creating or changing sequence of welds | |
| DE102016107398B4 (en) | QUALITY CONTROL SYSTEM FOR SPOT WELDING | |
| DE202013012002U1 (en) | Systems and devices for monitoring the quality of welding | |
| DE102018202821A1 (en) | ARC WELDING SYSTEM AND ASSESSMENT DEVICE FOR ARC FLASH | |
| JPH10235490A (en) | Method for evaluating weld state of electric welding machine and device therefor | |
| CN116423005B (en) | Tin soldering process optimization method and system for improving welding precision | |
| CN115277464A (en) | Cloud network change flow anomaly detection method based on multi-dimensional time series analysis | |
| JP2020191050A (en) | Abnormality detector and abnormality detection method | |
| CN117102630B (en) | Arc welding quality monitoring method, device, electronic equipment and storage medium | |
| CN112200000A (en) | Welding stability recognition model training method and welding stability recognition method | |
| CN111331211B (en) | On-line penetration detection method and system for electric spark small hole machining | |
| WO2020099135A1 (en) | Method and assistance system for generating training data for a learning-based configuration of a technical system | |
| EP4022408A1 (en) | Method and device for analyzing a sequential process | |
| CN115106615B (en) | Welding deviation real-time detection method and system based on intelligent working condition identification | |
| CN116060817A (en) | Method and device for determining welding quality of composite current collector and electronic equipment | |
| US8426771B2 (en) | Method of monitoring machine condition | |
| CN113487149B (en) | Welding spot abnormity identification system and method based on Catboost K-fold cross verification | |
| JP5525144B2 (en) | Waveform determination apparatus and time axis position adjustment method in the waveform determination apparatus | |
| KR101584421B1 (en) | Monitoring system for arc welding |
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 |