CN115439429A - Weld quality real-time online evaluation method and device, storage medium and terminal - Google Patents
Weld quality real-time online evaluation method and device, storage medium and terminal Download PDFInfo
- Publication number
- CN115439429A CN115439429A CN202211030198.9A CN202211030198A CN115439429A CN 115439429 A CN115439429 A CN 115439429A CN 202211030198 A CN202211030198 A CN 202211030198A CN 115439429 A CN115439429 A CN 115439429A
- Authority
- CN
- China
- Prior art keywords
- welding
- real
- image
- evaluation
- time
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
Abstract
The invention provides a real-time online evaluation method, a real-time online evaluation system, a storage medium and a real-time online evaluation terminal for weld quality, wherein the method comprises the following steps: step 1, acquiring a weld quality real-time evaluation model established based on a preset machine learning method; and 2, performing automatic motion control on the welding process through a welding control platform, acquiring a current welding image through an imaging system, and evaluating the current welding image by adopting the real-time welding seam quality evaluation model to generate a welding seam quality evaluation result. And the weld pool image is rapidly processed and evaluated to realize real-time online evaluation of the weld quality, and a student observes the weld image and the weld quality information through a computer, remotely adjusts welding parameters and obtains an optimal welding scheme. The learning effect is improved through instant feedback; keep away from abominable welding environment, adopt remote operation, overcome the fearing emotion of beginner to the highlight, high temperature, smoke and dust, noise in the welding, improve the operational environment, reduce student's fearing psychology.
Description
Technical Field
The invention belongs to the technical field of welding, and particularly relates to a method and a device for real-time online evaluation of weld quality, a storage medium and a terminal.
Background
With the development of production, welding is widely applied to the industrial departments of aerospace, aviation, nuclear industry, shipbuilding, building, mechanical manufacturing and the like, and the welding technical level is an important index for measuring the strength of a manufacturing country. The demand of China for welders keeps a great trend all the time, the welders have been continuously selected for three periods, 100 careers with nationwide recruitment larger than the job hunting of the 'the fewest workers' are ranked and listed as the top mao each time. Internationally, the shortage of skilled welders-finding, training, and retaining welders remains a major challenge facing the industry. For example, in the United states, the index of procurement managers for month 5 is 62.1, which is higher than 43.1 in 2020.
Electric welding belongs to special operation, needs to be certified, welding skill training is an important way for improving welding level and obtaining a welder certificate, and how to cultivate welding technicians meeting the requirements of modern industry becomes a new current subject of a welding training department. The traditional welding training practice effect needs to be observed after the welding line is cooled, evaluation can be carried out only after 30min at least, evaluation is mainly carried out according to the experience of teachers, the requirement on the skills of the teachers is high, and the current professional teachers and resource amount of the welding in universities and colleges of professorship are generally insufficient. And for internal defects, the weld joints are required to be cut for metallographic structure detection, so that the evaluation cost is high, the time consumption is long, and the learning rule of students is not met.
With the coming of a new information technology revolution based on cloud computing, big data, the Internet of things and artificial intelligence, the automatic welding detection technology is developed in a breakthrough manner. The existing technology is mainly embodied in on-line monitoring of a welding process, a subsequent welding seam quality evaluation link is lacked, and some on-line detection technologies mainly reflect welding seam quality defects indirectly by acquiring spectrum and sound signals, can not realize direct end-to-end welding seam defect image identification, are not convenient and poor in intuition, and are not beneficial to learning of students.
Although the existing emerging virtual welding training technology has many advantages, training equipment is long in development period, expensive in construction cost, complex in calculation, poor in reality sense and different from real welding in a certain degree, and worn equipment can cause personnel dizziness.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides a method and a device for real-time online evaluation of weld quality, a storage medium and a terminal.
In order to achieve the technical purpose, the technical scheme adopted by the invention for solving the technical problems is as follows:
a real-time online evaluation method for weld quality comprises the following steps:
step 1, acquiring a weld quality real-time evaluation model established based on a preset machine learning method;
and 2, performing automatic motion control on the welding process through a welding control platform, acquiring a current welding image through an imaging system, and evaluating the current welding image by adopting the real-time welding seam quality evaluation model to generate a welding seam quality evaluation result.
As a preferred embodiment, the step of establishing the weld quality real-time evaluation model based on the preset machine learning method comprises the following steps:
step 101, acquiring welding sample pictures corresponding to different joint forms and different weldment thicknesses according to preset welding requirements, classifying and labeling each welding sample picture, and dividing the labeled welding sample pictures into a training set, a verification set and a test set;
102, performing image processing on each welding sample picture in the training set, the verification set and the test set for model training;
103, training a plurality of preset machine learning models of the Scikit-learn library respectively based on python language and the welding sample pictures of the training set to generate corresponding initial evaluation models;
104, verifying each initial evaluation model by adopting the verification set, generating f1-score corresponding to each initial evaluation model, screening out at least one initial evaluation model with the f1-score larger than a first preset threshold value, and taking the initial evaluation model as a primary selection model;
and 105, evaluating the welding sample pictures of the test set in real time by adopting each primary selection model, generating evaluation test time corresponding to each primary selection model, and taking the primary selection model with the evaluation test time smaller than a second preset threshold value as the real-time weld quality evaluation model.
As a preferred embodiment, each of the welding sample pictures includes a weld puddle image, an arc image and/or a weld image; the labeling results of classifying and labeling each welding sample picture comprise good, burn-through, pollution, incomplete fusion, lack of protective gas and too fast movement of a welding gun.
The invention rapidly processes and evaluates the weld pool image, realizes the real-time online evaluation of the weld quality, and enables students to observe the weld image and the weld quality information through a computer, remotely operate and adjust welding parameters, and obtain the optimal welding scheme. Can promote the learning effect through instant feedback, can keep away from abominable welding environment again, adopt remote operation, overcome the difficult emotion of fearing of beginner to highlight, high temperature, smoke and dust, noise in the welding, improve the operational environment, reduce student's the psychology of fearing.
As a preferred embodiment, the image processing on each welding sample picture in the step 102 includes the following steps:
step 1021, reading the welding sample picture;
step 1022, converting the pixels of the welding sample picture into 1280 × 1024;
step 1023, cutting pixels of the non-information area in the welding sample picture to change the pixels of the welding sample picture from 1280 x 1024 to 1280 x 700;
step 1024, performing graying processing on the cut welding sample picture to generate a grayscale image;
step 1025, converting the pixels of the gray map from 1280 x 700 to 40 x 22, and converting the data type to float32 format;
in step 1026, the converted gray-scale map is normalized, so that the pixels of the gray-scale map are converted from 40 × 22 into a one-dimensional array for subsequent model training.
As a preferred embodiment, the imaging system includes a high-speed camera, a synchronous controller, a current-voltage synchronous playback module and an imaging optical path, the high-speed camera is disposed on a mechanical arm of the welding control platform and forms a 45-degree included angle with a welding gun of the welding control platform, and the synchronous controller controls the high-speed camera and the welding gun to move synchronously, so as to acquire an area image directly in front of a welding pool and a welding arc in real time and use the area image as the current welding image.
As a preferred embodiment, the method further comprises a step 3, wherein the step 3 specifically comprises:
step 301, continuously acquiring a welding image shot by the high-speed camera at a preset frequency, recording a welding parameter value corresponding to the welding image, and performing real-time evaluation on the welding image by using the real-time evaluation model of the weld quality to generate a corresponding weld quality evaluation result;
step 302, marking the welding parameter values and the welding seam quality evaluation results to corresponding welding images;
303, merging the marked welding images according to a target editing sequence to generate a target welding video;
304, displaying the target welding video through an upper computer, and generating an optimized welding parameter value reaching a preset optimized target according to the target welding video;
the welding parameter values include an arc voltage value, a welding current value, and/or a torch movement speed.
The invention develops a welding training real-time evaluation system based on a mixed reality technology by carrying out system integration with the existing hardware platform control system. The welding camera with the high dynamic range is utilized to clearly and visually display the real-time situation of the molten pool to the student through the screen, the welding seam quality evaluation result is displayed on the screen in real time, the student can acquire the welding seam image and the welding seam quality information simultaneously, whether the welding practice result is qualified or not and the welding seam quality defect type are known in time, and the teaching effect is improved. In the welding teaching process, a student can adjust welding parameters (including voltage, current, welding gun moving speed and the like) in real time through keys, observe the change of a molten pool under different parameter conditions in real time, so as to obtain the optimal welding parameters under the working condition, and record the optimal welding parameters in a file.
As a preferred embodiment, the step 303 of merging the marked welding images according to a target editing order to generate a target welding video specifically includes the following steps:
3031, obtaining a pre-established mapping relation table, wherein the mapping relation table comprises a number, acquisition time, welding parameter values and a welding seam quality evaluation result corresponding to each welding image;
3031, acquiring a user instruction generated according to user requirements, wherein the user instruction comprises target optimization welding parameters and an image editing sequence;
3032, inquiring the mapping relation table according to the target optimized welding parameters to obtain at least one marked target welding image, and combining the at least one target welding image according to the image editing sequence to generate a target welding video.
As a preferred embodiment, a time threshold corresponding to a preset synchronization degree is generated according to the switching time of each frame of image in the target welding video, and the time threshold is used as the second preset threshold.
The second aspect of the embodiment of the invention provides a real-time online evaluation device for the weld quality, which is characterized by comprising an acquisition module and a weld quality evaluation module;
the acquisition module is used for acquiring a welding seam quality real-time evaluation model established based on a preset machine learning method;
the welding seam quality evaluation module is used for automatically controlling the motion of a welding process through a welding control platform, acquiring a current welding image through an imaging system, and evaluating the current welding image by adopting the welding seam quality real-time evaluation model to generate a welding seam quality evaluation result.
As a preferred embodiment, the obtaining module specifically includes:
the labeling and classifying unit is used for acquiring welding sample pictures corresponding to different joint forms and different weldment thicknesses according to preset welding requirements, classifying and labeling each welding sample picture, and dividing the labeled welding sample pictures into a training set, a verification set and a test set;
the image processing unit is used for carrying out image processing on each welding sample picture so as to carry out model training;
the training unit is used for training a plurality of preset machine learning models of the Scikit-learn library respectively based on python language and the welding sample pictures of the training set to generate corresponding initial evaluation models;
the first selection unit is used for verifying each initial evaluation model by adopting the verification set, generating f1-score corresponding to each initial evaluation model, screening out at least one initial evaluation model with the f1-score larger than a first preset threshold value, and taking the initial evaluation model as a primary selection model;
and the second selection unit is used for evaluating the welding sample picture of the test set in real time by adopting each primary selection model, generating evaluation test time corresponding to each primary selection model, and taking the primary selection model with the evaluation test time smaller than a second preset threshold value as the real-time weld quality evaluation model.
As a preferred embodiment, the image processing unit specifically includes:
the reading unit is used for reading the welding sample picture;
a first conversion unit, configured to convert pixels of the welding sample picture into 1280 × 1024;
the cutting unit is used for cutting pixels of the information-free area in the welding sample picture to change the pixels of the welding sample picture from 1280 x 1024 to 1280 x 700;
the gray processing unit is used for carrying out gray processing on the cut welding sample picture to generate a gray image;
the second conversion unit is used for converting the pixels of the gray scale map from 1280 x 700 to 40 x 22, and converting the data type into float32 format;
and the regularization unit is used for regularizing the converted gray-scale image so as to convert the pixels of the gray-scale image from 40 × 22 into a one-dimensional array for subsequent model training.
As a preferred embodiment, the method further includes a model building module, specifically including:
the first acquisition unit is used for continuously acquiring the welding images shot by the high-speed camera at a preset frequency, recording welding parameter values corresponding to the welding images, and performing real-time evaluation on the welding images by adopting the real-time evaluation model of the welding seam quality to generate corresponding evaluation results of the welding seam quality;
the marking unit is used for marking the welding parameter values and the welding seam quality evaluation results into corresponding welding images;
the merging unit is used for merging the marked welding images according to the target editing sequence to generate a target welding video;
the optimization unit is used for displaying the target welding video through an upper computer and generating an optimized welding parameter value reaching a preset optimized target according to the target welding video;
the optimized welding parameter values include an arc voltage value, a welding current value, and/or a welding gun movement speed.
As a preferred embodiment, the merging unit specifically further includes:
the second acquisition unit is used for acquiring a pre-established mapping relation table, and the mapping relation table comprises a number, acquisition time, welding parameter values and a welding seam quality evaluation result corresponding to each welding image;
the third acquisition unit is used for acquiring a user instruction generated according to the user requirement, wherein the user instruction comprises a target optimization welding parameter and an image editing sequence;
and the video generating unit is used for inquiring the mapping relation table according to the target optimized welding parameters to obtain at least one marked target welding image, and combining the at least one target welding image according to the image editing sequence to generate a target welding video.
A third aspect of the present invention provides a computer-readable storage medium storing a computer program, which when executed by a processor, implements the above-described weld quality real-time online evaluation method.
A fourth aspect of the embodiments of the present invention provides a real-time online evaluation terminal for weld quality, including the computer-readable storage medium and a processor, where the processor implements the steps of the weld quality real-time online evaluation method when executing a computer program on the computer-readable storage medium.
The invention has the following beneficial effects:
(1) The invention rapidly processes and evaluates the weld pool image, realizes the real-time on-line evaluation of the weld quality, and the student observes the weld image and the weld quality information through the computer, remotely adjusts the welding parameters and obtains the optimal welding scheme. Can promote the learning effect through instant feedback, can keep away from abominable welding environment again, adopt remote operation, overcome the difficult emotion of fearing of beginner to highlight, high temperature, smoke and dust, noise in the welding, improve the operational environment, reduce student's the psychology of fearing.
(2) The invention carries out system integration with the existing hardware platform control system and develops a welding training real-time evaluation system based on the mixed reality technology. The welding camera with the high dynamic range is utilized to clearly and visually display the real-time conditions of the molten pool to the student through the screen, the welding seam quality evaluation result is displayed on the screen in real time, the student can simultaneously acquire the welding seam image and the welding seam quality information, whether the welding practice result is qualified or not and the type of the welding seam quality defect are known in time, and the teaching effect is improved. In the welding teaching process, a student can adjust welding parameters (including voltage, current, welding gun moving speed and the like) in real time through keys, observe the change of a molten pool under different parameter conditions in real time, so as to obtain the optimal welding parameters under the working condition, and record the optimal welding parameters in a file.
(3) The invention has good real experience and mixed reality, can observe the welding condition in real time in real welding, and is convenient for learning and mastering.
(4) The invention has less hardware resource requirement, can meet the requirement of a common PC machine, has low cost, is suitable for training of vocational schools and factories, can be applied to an image recognition system of an automatic welding robot, and improves the intelligent level of the welding robot.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a schematic flow chart of a real-time online evaluation method for weld quality provided in example 1;
FIG. 2 is a schematic view of a flowchart of image processing performed on a welding sample picture according to embodiment 1
FIG. 3 is a schematic structural diagram of a real-time on-line evaluation apparatus for weld quality provided in example 2;
fig. 4 is a schematic structural diagram of a terminal provided in embodiment 3.
Detailed Description
In order to make the objects, technical solutions and advantageous technical effects of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and the detailed description. It should be understood that the detailed description and specific examples, while indicating the preferred embodiment of the invention, are given by way of illustration only and not by way of limitation.
Fig. 1 is a schematic flow chart of a real-time online evaluation method for weld quality provided in embodiment 1, as shown in fig. 1, including the following steps:
step 1, obtaining a welding seam quality real-time evaluation model established based on a preset machine learning method.
In a preferred embodiment, the weld quality real-time evaluation model established based on the preset machine learning method comprises the following steps:
step 101, acquiring welding sample pictures corresponding to different joint forms and different weldment thicknesses according to preset welding requirements, classifying and labeling each welding sample picture, and dividing the labeled welding sample pictures into a training set, a verification set and a test set; wherein each of the welding sample pictures comprises a weld puddle image, an arc image, and/or a weld image; the labeling results of classifying and labeling each welding sample picture comprise good, burn-through, pollution, incomplete fusion, lack of protective gas and too fast movement of a welding gun.
In a specific embodiment, tig welding and stainless steel materials are selected as welding parts, wherein tig welding is one of green welding technologies, stainless steel materials are currently used, and other welding methods can be adopted in other embodiments. Taking stainless steel as an example, welding requirements such as joint forms, weldment thicknesses and the like commonly used in stainless steel welding are investigated firstly, and a large number of stainless steel protective welding sample pictures corresponding to different joint forms, weldment thicknesses and the like are collected comprehensively according to different welding requirements, wherein the pictures can be historical welding sample pictures generated by the welding control platform, and can also be historical welding sample pictures obtained by taking pictures or shooting pictures at an actual welding site. Each welding sample picture comprises a molten pool image, an electric arc image and/or a welding seam image, and then each picture is classified and labeled, wherein the classification comprises 6 welding seam quality results of good, burn-through, pollution, incomplete fusion, protective gas deficiency and welding gun moving too fast, namely 1 good in 5 defects. And then, dividing the marked pictures into a training set, a verification set and a test set according to uniform weld quality or according to a certain proportion, such as 8.
Then, step 102 is executed, and image processing is performed on each welding sample picture of the training set, the verification set and the test set so as to prepare for model training.
In a preferred embodiment, referring to fig. 2, the image processing for each welding sample picture in step 102 includes the following steps:
step 1021, reading the welding sample picture;
step 1022, converting the pixels of the welding sample picture into 1280 × 1024;
step 1023, cutting pixels of the non-information area in the welding sample picture to change the pixels of the welding sample picture from 1280 x 1024 to 1280 x 700;
step 1024, performing graying processing on the cut welding sample picture to generate a grayscale image;
step 1025, converting the pixels of the gray map from 1280 x 700 to 40 x 22, and converting the data type to float32 format; wherein, float32 is also called single precision format, and a number is expressed by 32 bits, namely 4 bytes, and comprises 1 sign bit, 8 exponent bits and 23 decimal bits.
In step 1026, the converted gray-scale map is normalized, so that the pixels of the gray-scale map are converted from 40 × 22 into a one-dimensional array for subsequent model training.
By adopting the steps, the welding sample pictures can be converted into the one-dimensional array with uniform format, so that the training difficulty is reduced in the subsequent model training process, and the training efficiency and effect are improved.
Then, step 103 is executed, a plurality of preset machine learning models of the Scikit-leann library are trained respectively based on the python language and the welding sample picture of the training set, and corresponding initial evaluation models are generated. Wherein Scikit-leann is a free software machine learning library for Python programming language. The method has various classification, regression and clustering algorithms, including a support vector machine, a random forest, gradient promotion, a k-means and DBSCAN, and can be used together with a Python numerical science library NumPy and SciPy, thereby realizing the effect of machine training. The embodiment of the invention trains a plurality of preset machine learning models of the Scikit-learn library respectively, for example, KNN, CNN, SVM, CART and NB models respectively, so as to train and generate corresponding initial evaluation models.
And then, executing step 104, verifying each initial evaluation model by using the verification set, generating f1-score corresponding to each initial evaluation model, screening out at least one initial evaluation model with the f1-score larger than a first preset threshold value, and taking the initial evaluation model as a primary selection model.
Where F1-Score (F1 Score) is an index used in statistics to measure the accuracy of a two-class (or multi-task two-class) model. The method gives consideration to the accuracy and the recall rate of the classification model, namely the specific F1 score can be regarded as a weighted average of the accuracy and the recall rate of the model, the maximum value of the weighted average is 1, the minimum value of the weighted average is 0, and the larger the value is, the better the model is. According to the technical scheme, first screening is carried out according to f1-score, and at least one initially selected model with the f1-score value larger than a first preset threshold value is obtained. For example, in one embodiment, the accuracy and F1 score of initial evaluation models generated by training, such as KNN, CNN, SVM, CART, and NB, are shown in table 1 below:
table 1: initial KNN, CNN, SVM, CART and NB evaluation model accuracy and F1 score
| Model (model) | Rate of accuracy | f1-score |
| KNN | 0.98 | 0.97 |
| CNN | 0.92 | 0.77 |
| SVM | 0.9 | 0.74 |
| CART | 0.82 | 0.63 |
| NB | 0.28 | 0.39 |
When the first preset threshold is set to 0.75, two primary selection models, KNN-based and CNN-based, can be screened out.
And finally, executing a step 105, evaluating the welding sample picture of the test set in real time by adopting each primary selection model, generating evaluation test time corresponding to each primary selection model, and taking the primary selection model with the evaluation test time smaller than a second preset threshold value as the real-time weld quality evaluation model.
In a preferred embodiment, when each primary selection model is used to predict and evaluate a welding sample picture in a test set in real time, timing is started from the reading of the welding sample picture, the evaluation test time consumed in the whole process of reading, preprocessing and generating a prediction result of the welding sample picture is obtained, the primary selection model with the evaluation test time smaller than a second preset threshold is used as the real-time weld quality evaluation model, for example, in one embodiment, two primary selection models, namely, a KNN-based primary selection model and a CNN-based primary selection model, are included, and finally, the KNN model with the evaluation test time smaller than the second preset threshold is used as the real-time weld quality evaluation model.
In the above embodiment, if there are a plurality of primary models whose evaluation test time is less than the second preset threshold, the primary model whose evaluation test time is the smallest is selected as the real-time weld quality evaluation model.
And if the number of the initial selection models with the evaluation test time smaller than the second preset threshold is 0, returning to the step 104 for training again, adjusting the training target in the training process according to the difference between the evaluation test time and the second preset threshold, namely, screening a plurality of initial evaluation models with f1-score larger than the first preset threshold again to be used as the initial selection models, and repeating the step 105 until the welding seam quality real-time evaluation models meeting the requirements are obtained.
And then executing step 2, performing automatic motion control on the welding process through a welding control platform, acquiring a current welding image through an imaging system, and evaluating the current welding image by adopting the real-time welding seam quality evaluation model to generate a welding seam quality evaluation result.
In the above embodiment, the welding control platform can implement automatic welding experiments of some typical welding positions. Particularly, the equipment required by the welding control platform comprises a welding robot body, a control cabinet, an independent welding positioner, a single-shaft motion control mechanism and a workpiece clamping mechanism, and the welding control platform can realize automatic motion control of the welding process, avoid uncertain factors in manual welding experiments and improve the scientificity of the experiments.
As a preferred embodiment, the imaging system includes a high-speed camera, a synchronous controller, a current-voltage synchronous playback module and an imaging optical path, the high-speed camera is disposed on a mechanical arm of the welding control platform and forms a 45-degree included angle with a welding gun of the welding control platform, and the synchronous controller controls the high-speed camera and the welding gun to move synchronously, so as to acquire an area image directly in front of a welding pool and a welding arc in real time and use the area image as the current welding image.
The high-speed camera realizes high-speed imaging acquisition of a welding target area, the synchronous controller realizes synchronization with a current voltage signal, a playback function of a shot image and the current voltage signal can be realized through playback software, and an imaging optical path in the welding process provides imaging optical condition support for imaging of the target area, and the imaging optical condition support comprises a camera lens, a backlight light source, a lens, an optical filter, a dimmer and the like, so that an observed object (molten drops, electric arcs, a molten pool and the like) can be imaged clearly.
In the specific embodiment, as the strong arc generated in the welding process brings huge challenges to the real-time detection and evaluation of the quality of the welding seam, the image acquisition and image processing process becomes more difficult, in the embodiment, a high-speed camera of Xiris XVC-1000 is adopted, and the camera is suitable for TIG, MIG/MAG, plasma, laser and laser arc composite, electron beam welding and other scenes. The camera has a dynamic range of 140db and is capable of absorbing sufficient light to increase the brightness of the area around the arc while avoiding overexposure of the arc. In actual operation, the Xiris XVC-1000 high-speed camera is arranged on a mechanical arm of the welding control platform and forms a 45-degree included angle with a welding gun of the welding control platform, and the Xiris XVC-1000 high-speed camera and the welding gun are controlled to synchronously move through the synchronous controller, so that area images in front of a welding pool and a welding arc are collected in real time and serve as the current welding image, and the influence of the clamping position and angle of the high-speed camera on the shot image is reduced.
In order to better improve the welding training effect, the method for real-time online evaluation of the weld quality in the preferred embodiment further includes step 3, where step 3 specifically includes:
step 301, continuously acquiring a welding image shot by the high-speed camera at a preset frequency, recording a welding parameter value corresponding to the welding image, and performing real-time evaluation on the welding image by using the real-time evaluation model of the weld quality to generate a corresponding weld quality evaluation result;
step 302, marking the welding parameter values and the welding seam quality evaluation results to corresponding welding images;
303, merging the marked welding images according to a target editing sequence to generate a target welding video;
304, displaying the target welding video through an upper computer, and generating an optimized welding parameter value reaching a preset optimized target according to the target welding video;
the welding parameter values include an arc voltage value, a welding current value, and/or a torch movement speed.
In the preferred embodiment, the high-speed camera with a high dynamic range is used for clearly and intuitively displaying the real-time conditions of the molten pool to the student through the screen, the welding parameters and the welding seam quality evaluation results are displayed on the screen in real time, the student can simultaneously acquire the welding seam images, the welding parameters and the welding seam quality information, know whether the welding practice results are qualified or not and the types of the welding seam quality defects in time, and improve the teaching effect. Meanwhile, in the welding teaching or training process, welding parameters (including voltage, current, welding gun moving speed and the like) can be adjusted through real-time key pressing by a student, or the welding parameters are automatically adjusted according to a preset optimization scheme, and then the welding control platform adjusts corresponding welding actions in real time according to the adjusted welding parameters, so that the student can conveniently observe the change of a molten pool under different parameter conditions in real time, the optimal welding parameters under the working condition can be obtained, and the optimal welding parameters can be recorded in a file.
In one particular embodiment, the welding parameter values generally include an arc voltage value, a welding current value, and/or a torch travel speed. The welding parameters of the TIG welding can also comprise the diameter of a tungsten rod, the arc length, the flow of the shielding gas, the aperture and the height of a nozzle, the inclination angle of a filler wire and the like, and different welding processes can be trained and optimized by pertinently selecting different welding parameters.
Taking welding current, arc voltage and welding speed as examples, all three will have direct influence on the evaluation result of the welding seam quality.
When the welding current is increased (other conditions are unchanged), the penetration and the extra height of the welding seam are increased, and the fusion width is unchanged (or slightly increased) for the following reasons: after the current is increased, the electric arc force and the heat input on the workpiece are increased, the heat source position moves downwards, the melting depth is increased, and the melting depth and the current are close to a direct proportion relation. The amount of melted wire increases approximately proportionally, and the height of the wire increases because the melt width is approximately constant. The arc column diameter increases, but the depth of arc penetration into the workpiece increases, and the range of arc spot movement is limited, so that the weld width is nearly constant.
After the arc voltage is increased, the arc power is increased, the heat input of the workpiece is increased, the arc length is elongated, the distribution radius is increased, and therefore the fusion depth is slightly reduced and the fusion width is increased; the extra height is reduced because the weld width is increased and the amount of weld wire melting is slightly reduced.
And the linear energy is reduced when the welding speed is increased, and the penetration, the fusion width and the residual height are reduced. This is because the deposition amount of the wire metal per unit length of the weld bead is inversely proportional to the welding speed, and the weld width is inversely proportional to the evolution of the welding speed.
Therefore, welding current, arc voltage and welding speed in the welding parameters influence the penetration, the fusion width, the welding seam and the like, and the optimal welding parameters under the working condition are obtained by adjusting the welding parameters in real time according to the changes of a molten pool, the welding seam and the like under different parameter conditions, so that guidance is provided for obtaining the optimal welding scheme, and the training effect on the welding level of students is improved.
In another preferred embodiment, the method can also adopt an interactive interface of a single desktop graphical interface without depending on the internet, and simultaneously has a review function of displaying at least one marked target welding image and at least one marked target welding video, and provides video playing, welding seam quality evaluation and welding parameter display in a specific operation process, so that the operation is convenient.
As a preferred embodiment, the step 303 of merging the marked welding images according to a target editing order to generate a target welding video specifically includes the following steps:
3031, obtaining a pre-established mapping relation table, wherein the mapping relation table comprises a number, acquisition time, welding parameter values and a welding seam quality evaluation result corresponding to each welding image;
3031, acquiring a user instruction generated according to user requirements, wherein the user instruction comprises target optimization welding parameters and an image editing sequence;
3032, inquiring the mapping relation table according to the target optimized welding parameters to obtain at least one marked target welding image, and combining the at least one target welding image according to the image editing sequence to generate a target welding video.
The above preferred embodiment may not display all welding images, but select a target welding image representing the change of the target welding parameter from the target welding parameters focused by the user according to the user's requirements, for example, and then sequentially edit and combine the target welding images according to the parameter size change sequence, time sequence, and the like, thereby generating a target welding video.
As a preferred embodiment, a time threshold corresponding to a preset synchronization degree is generated according to the switching time of each frame of image in the target welding video, and the time threshold is used as the second preset threshold. Specifically, in a preferred embodiment, if the switching time of each frame of image is T 1 I.e. per interval T 1 The next target weld image will be switched to. To satisfy the display effect of synchronism, the display effect is required to be T 1 Obtaining the weld quality evaluation result of the target welding image within time, so that the time threshold value T = A (T) 1 -T 0 ) Said A isThe value range is 0.8-1.2 for adjusting the parameters, and the value of the adjustment parameter A is taken according to the preset synchronization degree, wherein the higher the preset synchronization degree is, the smaller the value of A is. T is 1 Presetting switching time for each frame of image in a target welding video; t is 0 The time is marked, namely, the corresponding welding parameter value is read, and the sum of the time required by marking the welding parameter value and the welding seam quality evaluation result to the corresponding welding image can be obtained by calculation according to historical data, so that the synchronism of video playing and model prediction results is ensured.
As a preferred embodiment, in a classification model algorithm of KNN, CNN, SVM, CART and NB, the KNN model is adopted in this embodiment, common weld quality defects can be evaluated in real time, the weld pool condition of the weld can be observed in real time through a computer screen, the intuitiveness is good, the visualization degree is high, no time delay exists, problems in welding can be found in time, welding parameters can be adjusted in time, students are instructed to perform the operation, the accuracy is higher and can reach 98%, and 2% of misjudgment conditions are that good welds are mistakenly judged as defective welds, and the product qualification rate flowing into a next procedure can be ensured to be 100%; the requirement on hardware resources is low, and a common PC can meet the requirement; meanwhile, the characteristic right-lifting is not needed manually, the algorithm is simple, the system response is fast, the average time for identifying a single image is 33ms, and the real-time performance is good.
The results of the comparison of the in-weld and post-weld test methods in the above examples are shown in table 2:
TABLE 2 comparison of the two detection methods
| Duration of detection | Product percent of pass | Check theMeasure personnel requirements | |
| In-weld detection | Real time | 100% | Automatic judgment of No, system |
| Post weld inspection | 30min to 7 days | About 98 percent | Require a high level of knowledge and experience |
FIG. 3 is a weld quality real-time on-line evaluation apparatus provided in embodiment 2, which includes an acquisition module 100 and a weld quality evaluation module 200;
the acquisition module 100 is used for acquiring a welding seam quality real-time evaluation model established based on a preset machine learning method;
the welding seam quality evaluation module 200 is used for performing automatic motion control on a welding process through a welding control platform, acquiring a current welding image through an imaging system, and evaluating the current welding image by adopting the welding seam quality real-time evaluation model to generate a welding seam quality evaluation result.
In a preferred embodiment, the obtaining module specifically includes:
label classification unit 1001: the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring welding sample pictures corresponding to different joint forms and different weldment thicknesses according to preset welding requirements, classifying and labeling each welding sample picture, and dividing the labeled welding sample pictures into a training set, a verification set and a test set;
the image processing unit 1002: the image processing device is used for carrying out image processing on each welding sample picture so as to carry out model training;
training unit 1003: the method comprises the steps that a plurality of preset machine learning models of a Sciket-learn library are trained respectively based on a python language and a welding sample picture of a training set to generate corresponding initial evaluation models;
the first selection unit 1004 is used for verifying each initial evaluation model by adopting the verification set, generating f1-score corresponding to each initial evaluation model, screening out at least one initial evaluation model with the f1-score larger than a first preset threshold value, and taking the initial evaluation model as a primary selection model;
the second selection unit 1005: and the evaluation test time corresponding to each primary selection model is generated, and the primary selection model with the evaluation test time smaller than a second preset threshold value is used as the real-time weld quality evaluation model.
In a preferred embodiment, the image processing unit 1002 specifically includes:
reading unit 10021: the image reading unit is used for reading the welding sample image;
first conversion unit 10022: for converting pixels of the welding sample picture to 1280 by 1024;
cutting unit 10023: the welding sample picture is used for cutting pixels of the non-information areas in the welding sample picture to change the pixels of the welding sample picture from 1280 x 1024 to 1280 x 700;
the gradation processing unit 10024: the system is used for carrying out graying processing on the cut welding sample picture to generate a grayscale image;
second conversion unit 10025: the pixel conversion module is used for converting the pixels of the gray map from 1280 x 700 to 40 x 22, and converting the data type to float32 format;
regularization unit 10026: the method is used for carrying out regularization processing on the converted gray-scale image so that pixels of the gray-scale image are converted from 40 × 22 into a one-dimensional array for subsequent model training.
In a preferred embodiment, the method further includes a model building module 300, where the model building module 300 specifically includes:
a first obtaining unit 3001, configured to continuously obtain a welding image captured by the high-speed camera at a preset frequency, record a welding parameter value corresponding to the welding image, and perform real-time evaluation on the welding image by using the real-time weld quality evaluation model to generate a corresponding weld quality evaluation result;
a marking unit 3002, configured to mark the welding parameter value and the weld quality evaluation result into a corresponding welding image;
a merging unit 3003, configured to merge the marked welding images according to a target editing order to generate a target welding video;
the optimization unit 3004 is configured to display the target welding video through an upper computer, and generate an optimized welding parameter value that reaches a preset optimization target according to the target welding video;
the optimized welding parameter values include an arc voltage value, a welding current value, and/or a welding torch movement speed.
In a preferred embodiment, the merging unit 3003 further includes:
a second obtaining unit 30031, configured to obtain a pre-established mapping relationship table, where the mapping relationship table includes a number, acquisition time, a welding parameter value, and a weld quality evaluation result corresponding to each welding image;
a third obtaining unit 30032, configured to obtain a user instruction generated according to a user requirement, where the user instruction includes a target optimized welding parameter and an image editing sequence;
a video generating unit 30033, configured to query the mapping relationship table according to the target optimized welding parameter to obtain at least one marked target welding image, and combine the at least one target welding image according to the image editing sequence to generate a target welding video.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by functions and internal logic of the process, and should not limit the implementation process of the embodiments of the present invention in any way.
The embodiment of the invention also provides a computer-readable storage medium, which comprises the computer-readable storage medium and a processor, wherein the processor realizes the steps of the weld quality real-time online evaluation method when executing the computer program on the computer-readable storage medium.
Fig. 4 is a schematic diagram of a real-time online evaluation terminal for weld quality provided in embodiment 3 of the present invention, and as shown in fig. 4, the real-time online evaluation terminal 8 for weld quality of this embodiment includes: a processor 80, a readable storage medium 81 and a computer program 82 stored in said readable storage medium 81 and executable on said processor 80. The processor 80, when executing the computer program 82, implements the steps in the various method embodiments described above, such as steps 1-2 shown in fig. 1. Alternatively, the processor 80, when executing the computer program 82, implements the functions of the modules in the above-described device embodiments, such as the functions of the modules 100 to 200 shown in fig. 3.
Illustratively, the computer program 82 may be partitioned into one or more modules that are stored in the readable storage medium 81 and executed by the processor 80 to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions for describing the execution process of the computer program 82 in the real-time online evaluation terminal 8 of the weld quality.
The real-time online evaluation terminal 8 for the weld quality can include, but is not limited to, a processor 80 and a readable storage medium 81. Those skilled in the art will appreciate that fig. 3 is merely an example of the real-time online evaluation terminal 8 based on the weld quality, and does not constitute a limitation of the real-time online evaluation terminal 8 of the weld quality, and may include more or less components than those shown, or combine some components, or different components, for example, the real-time online evaluation terminal of the weld quality may further include a power management module, an arithmetic processing module, an input-output device, a network access device, a bus, and the like.
The Processor 80 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The readable storage medium 81 may be an internal storage unit of the real-time online evaluation terminal 8 for the weld quality, such as a hard disk or a memory of the real-time online evaluation terminal 8 for the weld quality. The readable storage medium 81 may also be an external storage device of the real-time online evaluation terminal 8 for the quality of the weld seam, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like, which is equipped on the real-time online evaluation terminal 8 for the quality of the weld seam. Further, the readable storage medium 81 may also include both an internal storage unit and an external storage device of the real-time online evaluation terminal 8 of the weld quality. The readable storage medium 81 is used to store the computer program and other programs and data required for the real-time online evaluation terminal of the weld quality. The readable storage medium 81 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the description of each embodiment has its own emphasis, and reference may be made to the related description of other embodiments for parts that are not described or recited in any embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described embodiments of the apparatus/terminal device are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The invention is not limited solely to that described in the specification and embodiments, and additional advantages and modifications will readily occur to those skilled in the art, so that the invention is not limited to the specific details, representative apparatus, and illustrative examples shown and described herein, without departing from the spirit and scope of the general concept as defined by the appended claims and their equivalents.
Claims (10)
1. A real-time online evaluation method for weld quality is characterized by comprising the following steps:
step 1, acquiring a weld quality real-time evaluation model established based on a preset machine learning method;
and 2, performing automatic motion control on the welding process through a welding control platform, acquiring a current welding image through an imaging system, and evaluating the current welding image by adopting the real-time welding seam quality evaluation model to generate a welding seam quality evaluation result.
2. The real-time online evaluation method for the weld quality according to claim 1, wherein the establishment of the real-time evaluation model for the weld quality based on the preset machine learning method comprises the following steps:
step 101, acquiring welding sample pictures corresponding to different joint forms and different weldment thicknesses according to preset welding requirements, classifying and labeling each welding sample picture, and dividing the labeled welding sample pictures into a training set, a verification set and a test set;
102, performing image processing on each welding sample picture in the training set, the verification set and the test set for model training;
103, training a plurality of preset machine learning models of the Sciket-learn library respectively based on the python language and the welding sample pictures of the training set to generate corresponding initial evaluation models;
104, verifying each initial evaluation model by adopting the verification set, generating f1-score corresponding to each initial evaluation model, screening out at least one initial evaluation model with the f1-score larger than a first preset threshold value, and taking the initial evaluation model as a primary selection model;
and 105, evaluating the welding sample pictures of the test set in real time by adopting each primary selection model, generating evaluation test time corresponding to each primary selection model, and taking the primary selection model with the evaluation test time smaller than a second preset threshold value as the real-time weld quality evaluation model.
3. The method for the real-time online evaluation of the weld quality according to claim 2, wherein each welding sample picture comprises a weld pool image, an arc image and/or a weld bead image; the labeling results of classifying and labeling each welding sample picture comprise good, burn-through, pollution, incomplete fusion, lack of protective gas and too fast movement of a welding gun.
4. The method for real-time online evaluation of the weld quality according to claim 2, wherein the image processing of each welding sample picture in the step 102 comprises the following steps:
step 1021, reading the welding sample picture;
step 1022, converting the pixels of the welding sample picture into 1280 × 1024;
step 1023, cutting pixels of the non-information area in the welding sample picture to change the pixels of the welding sample picture from 1280 x 1024 to 1280 x 700;
step 1024, performing graying processing on the cut welding sample picture to generate a grayscale image;
step 1025, converting the pixels of the gray map from 1280 x 700 to 40 x 22, and converting the data type to float32 format;
and 1026, performing regularization processing on the converted gray-scale map, so that pixels of the gray-scale map are converted into a one-dimensional array for subsequent model training from 40 × 22.
5. The method for real-time online evaluation of the weld quality according to any one of claims 2 to 4, wherein the imaging system comprises a high-speed camera, a synchronous controller, a current-voltage synchronous playback module and an imaging optical path, the high-speed camera is arranged on a mechanical arm of the welding control platform and forms an included angle of 45 degrees with a welding gun of the welding control platform, and the high-speed camera and the welding gun are controlled by the synchronous controller to move synchronously so as to acquire images of a region right in front of a welding pool and a welding arc in real time and serve as the current welding image.
6. The weld quality real-time online evaluation method according to claim 5, further comprising a step 3, wherein the step 3 specifically comprises:
step 301, continuously acquiring a welding image shot by the high-speed camera at a preset frequency, recording a welding parameter value corresponding to the welding image, and performing real-time evaluation on the welding image by using the real-time welding seam quality evaluation model to generate a corresponding welding seam quality evaluation result;
step 302, marking the welding parameter values and the welding seam quality evaluation results to corresponding welding images;
303, merging the marked welding images according to a target editing sequence to generate a target welding video;
304, displaying the target welding video through an upper computer, and generating an optimized welding parameter value reaching a preset optimized target according to the target welding video;
the welding parameter values include an arc voltage value, a welding current value, and/or a torch movement speed.
7. The method for real-time online evaluation of the weld quality according to claim 6, wherein the step 303 of merging the marked welding images according to a target editing sequence to generate a target welding video specifically comprises the following steps:
3031, obtaining a pre-established mapping relation table, wherein the mapping relation table comprises a number, acquisition time, welding parameter values and a welding seam quality evaluation result corresponding to each welding image;
3031, acquiring a user instruction generated according to user requirements, wherein the user instruction comprises target optimization welding parameters and an image editing sequence;
3032, inquiring the mapping relation table according to the target optimized welding parameters to obtain at least one marked target welding image, and combining the at least one target welding image according to the image editing sequence to generate a target welding video.
8. A real-time online evaluation device for the weld quality is characterized by comprising an acquisition module and a weld quality evaluation module;
the acquisition module is used for acquiring a welding seam quality real-time evaluation model established based on a preset machine learning method;
the welding seam quality evaluation module is used for automatically controlling the motion of a welding process through the welding control platform, acquiring a current welding image through the imaging system, and evaluating the current welding image by adopting the welding seam quality real-time evaluation model to generate a welding seam quality evaluation result.
9. A computer-readable storage medium storing a computer program which, when executed by a processor, implements the weld quality real-time on-line assessment method according to any one of claims 1 to 7 above.
10. A real-time online evaluation terminal for weld quality comprises the computer-readable storage medium and a processor, wherein the processor realizes the steps of the real-time online evaluation method for weld quality according to any one of the above claims 1-7 when executing a computer program on the computer-readable storage medium.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211030198.9A CN115439429A (en) | 2022-08-26 | 2022-08-26 | Weld quality real-time online evaluation method and device, storage medium and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211030198.9A CN115439429A (en) | 2022-08-26 | 2022-08-26 | Weld quality real-time online evaluation method and device, storage medium and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115439429A true CN115439429A (en) | 2022-12-06 |
Family
ID=84245424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211030198.9A Pending CN115439429A (en) | 2022-08-26 | 2022-08-26 | Weld quality real-time online evaluation method and device, storage medium and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115439429A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115988866A (en) * | 2023-03-21 | 2023-04-18 | 深圳市利和兴股份有限公司 | NFC LAMI processing control method and system based on machine vision |
| CN116452588A (en) * | 2023-06-15 | 2023-07-18 | 苏州松德激光科技有限公司 | Welding quality assessment method and system |
| CN116871632A (en) * | 2023-09-08 | 2023-10-13 | 南京理工大学 | Robot backing welding forming defect monitoring method based on arc voltage and molten pool vision |
| CN117020361A (en) * | 2023-08-30 | 2023-11-10 | 鹤山市福维德智能设备有限公司 | Automatic detection device and detection method for mechanical manufacture |
| CN117436769A (en) * | 2023-12-20 | 2024-01-23 | 山东方垠智能制造有限公司 | Structural part welding quality monitoring method, system, storage medium and equipment |
| CN117745718A (en) * | 2024-02-19 | 2024-03-22 | 工业云制造(四川)创新中心有限公司 | Information interaction method based on cloud manufacturing |
-
2022
- 2022-08-26 CN CN202211030198.9A patent/CN115439429A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115988866A (en) * | 2023-03-21 | 2023-04-18 | 深圳市利和兴股份有限公司 | NFC LAMI processing control method and system based on machine vision |
| CN116452588A (en) * | 2023-06-15 | 2023-07-18 | 苏州松德激光科技有限公司 | Welding quality assessment method and system |
| CN116452588B (en) * | 2023-06-15 | 2023-10-31 | 苏州松德激光科技有限公司 | Welding quality assessment method and system |
| CN117020361A (en) * | 2023-08-30 | 2023-11-10 | 鹤山市福维德智能设备有限公司 | Automatic detection device and detection method for mechanical manufacture |
| CN117020361B (en) * | 2023-08-30 | 2024-02-09 | 鹤山市福维德智能设备有限公司 | Automatic detection device and detection method for mechanical manufacture |
| CN116871632A (en) * | 2023-09-08 | 2023-10-13 | 南京理工大学 | Robot backing welding forming defect monitoring method based on arc voltage and molten pool vision |
| CN116871632B (en) * | 2023-09-08 | 2023-11-21 | 南京理工大学 | Robot backing welding forming defect monitoring method based on arc voltage and molten pool vision |
| CN117436769A (en) * | 2023-12-20 | 2024-01-23 | 山东方垠智能制造有限公司 | Structural part welding quality monitoring method, system, storage medium and equipment |
| CN117745718A (en) * | 2024-02-19 | 2024-03-22 | 工业云制造(四川)创新中心有限公司 | Information interaction method based on cloud manufacturing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115439429A (en) | Weld quality real-time online evaluation method and device, storage medium and terminal | |
| US20210331266A1 (en) | System and method of exporting or using welding sequencer data for external systems | |
| JP6975762B2 (en) | Systems and methods that provide enhanced education and training in virtual reality environments | |
| CN104221069B (en) | Virtual welding system | |
| CN102165505B (en) | Real-time molten bath is used to feed back the system and method providing arc welding training under the reality environment of real-time simulation | |
| US10994357B2 (en) | System and method for creating or modifying a welding sequence | |
| US9928755B2 (en) | Virtual reality GTAW and pipe welding simulator and setup | |
| CN105788409A (en) | Importing and analyzing external data using a virtual reality welding system | |
| JP2016511443A (en) | Virtual reality GTAW and pipe welding simulator and setup | |
| CN105531750A (en) | Learning management system for a real-time simulated virtual reality welding training environment | |
| WO2014140749A1 (en) | Systems and method for creating or modifying a welding sequence | |
| AU2021259109A1 (en) | Method, device and computer program for determining the performance of a welding method via digital processing of an image of the welded workpiece | |
| CN212070747U (en) | Intelligent virtual welding training device | |
| CN111872589B (en) | Welding test control method, system, computer storage medium and equipment | |
| CN110718109A (en) | VR virtual welding simulation practical training system and practical training method | |
| CN111168211A (en) | Intelligent virtual welding training device and method | |
| CN113909636A (en) | Control method and control device of welding equipment, processor and welding system | |
| KR102572304B1 (en) | Method and apparatus for providing real-time welding inspection and defect prevention service based on artificial intelligence algorithm | |
| CN112400199B (en) | System and method for simulating robot operation | |
| Trubetskaya et al. | An implementation model for digitisation of visual management to develop a smart manufacturing process | |
| Bolotov et al. | Intelligent welding control system | |
| Sonar et al. | Indian railways on fast track with welding industry 4.0: Application of internet of things and artificial intelligence | |
| Veningston et al. | VR and AR in Welding Technologies | |
| Jiao et al. | Image Processing and Feature Extraction for Hull Structure GMAW Based on Weld Pool Visual Sensing | |
| Okumoto et al. | E-Training System of Welding Work |
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 |