CN116275742A - Welding method based on digital modeling and flexible weld joint recognition - Google Patents
Welding method based on digital modeling and flexible weld joint recognition Download PDFInfo
- Publication number
- CN116275742A CN116275742A CN202310377717.7A CN202310377717A CN116275742A CN 116275742 A CN116275742 A CN 116275742A CN 202310377717 A CN202310377717 A CN 202310377717A CN 116275742 A CN116275742 A CN 116275742A
- Authority
- CN
- China
- Prior art keywords
- welding
- model
- data
- layer
- parameters
- 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.)
- Withdrawn
Links
- 238000003466 welding Methods 0.000 title claims abstract description 248
- 238000000034 method Methods 0.000 title claims abstract description 103
- 238000005516 engineering process Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 43
- 238000000605 extraction Methods 0.000 claims description 21
- 238000013178 mathematical model Methods 0.000 claims description 18
- 210000002569 neuron Anatomy 0.000 claims description 18
- 238000007781 pre-processing Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 15
- 238000004519 manufacturing process Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 14
- 238000011176 pooling Methods 0.000 claims description 12
- 238000013507 mapping Methods 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000009499 grossing Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 7
- 230000007547 defect Effects 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 230000002068 genetic effect Effects 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 6
- 238000000513 principal component analysis Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 6
- 238000005493 welding type Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010224 classification analysis Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 230000000877 morphologic effect Effects 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000000611 regression analysis Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- 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
- B23K31/00—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups
- B23K31/02—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups relating to soldering or welding
-
- 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
- B23K37/00—Auxiliary devices or processes, not specially adapted to a procedure covered by only one of the preceding main groups
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Mechanical Engineering (AREA)
- Software Systems (AREA)
- Computer Graphics (AREA)
- Quality & Reliability (AREA)
- Optics & Photonics (AREA)
- Numerical Control (AREA)
Abstract
The invention belongs to the technical field of welding, in particular to a welding method based on digital modeling and flexible weld recognition, which comprises the following steps: firstly, carrying out three-dimensional modeling on a welded workpiece by a digital modeling technology, wherein the three-dimensional modeling comprises parameters such as weld joint position, shape and the like; then, recognizing the position and the shape of the welding seam by using a flexible welding seam recognition technology, and comparing and correcting the position and the shape with a digital modeling result; finally, according to the corrected weld position and shape, welding parameters including welding current, welding speed and the like are determined for welding.
Description
Technical Field
The invention relates to the technical field of welding, in particular to a welding method based on digital modeling and flexible weld joint identification.
Background
Welding is a process in which metallic or non-metallic materials are joined to one another by heat or pressure or both. In industrial production, welding is widely applied to the fields of mechanical manufacture, aerospace, automobile manufacture, ship construction, petrochemical industry and the like. In order to improve welding quality and efficiency, various automated or intelligent welding apparatuses and methods have been developed, such as robot welding, laser welding, arc welding, etc.
In the related art, the existing automatic or intelligent welding equipment and method still have some problems and defects, and the preset welding path is inconsistent with the actually required welding path due to the possible change or error of the shape and the size of a welded piece, so that the quality and the position accuracy of a welding line are affected.
The above information disclosed in this background section is only for the understanding of the background of the inventive concept and, therefore, it may contain information that does not form the prior art.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a welding method based on digital modeling and flexible weld recognition, which solves the problem that the preset welding path is inconsistent with the actually required welding path due to the possible change or error of the shape and the size of a welded part in the prior art, thereby influencing the quality and the position accuracy of the weld.
The technical scheme adopted for solving the technical problems is as follows: a welding method based on digital modeling and flexible weld joint recognition comprises the following steps,
a. determining welding type and welding material: before a three-dimensional mathematical model is established, the type of welding and the welding materials need to be determined;
b. collecting relevant data: collecting relevant data including three-dimensional graphics of the welding piece, coordinate axis position and attitude information of the welding robot, welding parameters and the like, wherein the data are used for constructing a three-dimensional mathematical model;
c. determining a modeling method: determining a method for establishing a three-dimensional mathematical model, and modeling based on a mathematical formula;
d. establishing a three-dimensional mathematical model: according to the collected data and the determined modeling method, a welded three-dimensional mathematical model is established, wherein the model comprises the shape, size, position, posture and other information of a welding part;
e. adjusting model parameters: according to the actual welding condition and the need, the model is adjusted, including the adjustment of parameters such as welding speed, welding angle, welding temperature and the like;
f. and (3) verifying the accuracy of the model: verifying the accuracy of the model by comparing the difference between the actual welding result and the model prediction result;
g. and (3) image acquisition: using a camera or a laser scanner to acquire images of the welded workpiece and acquire image information of the welding seam;
h. image preprocessing: preprocessing the acquired welding image, including denoising, graying, binarization, filtering and other operations, so as to improve the image quality and extract the characteristic information of the welding seam;
i. and (3) weld joint identification: carrying out automatic identification on the welding seam through image processing and feature extraction, wherein the automatic identification comprises the steps of determining parameters such as the position, the shape, the size and the like of the welding seam;
j. smoothing the identified weld edge and center line by curve fitting technique to eliminate noise and error and obtain corrected weld parameters
k. Digital modeling: comparing and correcting the identified weld joint position and shape with an original design model by utilizing a digital modeling technology so as to ensure the accuracy and stability of welding parameters;
and l, correcting welding parameters: according to the corrected weld position and shape, welding parameters including welding current, welding speed, welding angle and the like are adjusted to ensure the quality and stability of the weld;
m, welding operation: the hand-held arc welding equipment, the robot welding equipment and the like are utilized to carry out welding operation, so that accurate control and automatic operation of welding parameters are realized;
n, quality detection: and detecting the quality of the welded seam, including the aspects of appearance quality, dimensional accuracy, internal defects of the welded seam and the like, so as to ensure that the welding quality meets the requirements.
As an optimized technical scheme, the system used by the welding method comprises the following parts:
image acquisition equipment: the method comprises the steps of acquiring an image of a welded workpiece;
an image processing apparatus: the method comprises the steps of preprocessing an acquired image and extracting features;
weld recognition model apparatus: the method comprises the steps of classifying and identifying the preprocessed images, and determining the position and shape of a welding line;
correction algorithm equipment: the method comprises the steps of comparing and correcting the identified weld joint position and shape with a digital modeling result, and finally determining welding parameters for welding;
welding equipment: for performing a welding operation;
and (3) a control system: the device is used for carrying out linkage control on the equipment and realizing accurate adjustment and automatic control of welding parameters.
As an optimized technical scheme, the method also comprises the steps of establishing a mechanical learning model for flexible weld joint identification correction,
(1) Data acquisition and pretreatment: collecting data of a welding process, including sensor signals, video images, sounds and the like, and preprocessing, including denoising, filtering and normalization;
(2) Feature extraction and selection: extracting representative features from the preprocessed data, including morphological features, gray features and texture features, and selecting a proper feature set;
(3) Data set partitioning: dividing a data set into a training set, a verification set and a test set according to a certain proportion;
(4) Model selection and training: selecting a proper mechanical learning model, adopting a CNN model, and training the model by using a training set;
(5) Model optimization: adjusting parameters of the trained model, including selecting different optimizers, learning rates, batch sizes and the like, so as to improve the accuracy and robustness of the model;
(6) Model evaluation: evaluating the model by using the verification set, wherein the operation comprises calculating indexes such as accuracy, recall rate, F1 value and the like so as to determine the performance of the model;
(7) Model application: applying the trained model to data in a test set, and evaluating generalization capability and actual application effect of the model;
(8) Model update: updating and optimizing the model according to the actual application effect so as to improve the performance and the application value of the model;
(9) Model deployment: deploying the trained model into an actual application environment, such as a welding production line or an automated welding device;
(10) Real-time monitoring and correction: in the practical application process, the welding process is monitored and corrected in real time, for example, the position deviation of a welding line, the welding speed, the welding current and the like are detected, and the model is corrected according to the monitoring result, so that the self-adaptability and the robustness of the model are improved.
As an optimized technical scheme, the CNN model can be used for characteristic extraction and classification recognition of welding seams, and the main formulas comprise a convolution layer, a pooling layer, a full connection layer and the like, and the specific formulas are as follows:
wherein,,feature map, K, representing ith row and jth column of layer l l Represents the number of layer I convolution kernels, h l And w l Representing the height and width of the convolution kernel of layer I, < >>Weights representing the ith row and the ith column of the kth convolution kernel of the first layer,values representing the kth feature map of the ith+u-1 row, jth+v-1 column, b of layer 1 l A bias term representing a first layer;
wherein,,representing pooling results of ith row and jth column of the first layer, and s represents a pooling step length;
wherein y is k Represents the output of the layer L, the layer k neuron, J represents the number of layer L-1 neurons,representing the connection weights between the jth neuron of the L-1 layer and the kth neuron of the L-1 layer,>representing the output of the jth neuron of layer L-1, bk represents the bias term of the kth neuron of layer L, and f (·) represents the activation function.
As an optimized technical scheme, when the three-dimensional mathematical model is established, non-uniform rational B-spline interpolation (NUBIC) is adopted to perform curve fitting and path planning, the steps are as follows,
and (3) data acquisition: collecting curve data in the welding process, wherein the curve data comprises information such as coordinates and curvature of a curve;
pretreatment: preprocessing the acquired curve data, including data denoising, smoothing and sampling, so as to reduce noise and improve data quality;
NUBIC interpolation: interpolating the preprocessed curve data by using a non-uniform rational B-spline interpolation method to obtain a continuous and smooth curve representation;
and (3) parameter determination: determining parameters in NUBIC interpolation by using a genetic algorithm, including control point coordinates, weights, the times of B-splines and the like;
path planning: carrying out path planning according to the interpolation curve and the determined parameters to obtain a smooth welding path which meets the welding requirements;
the NUBIC interpolation formula is as follows:
c (u) represents the coordinate of a point on the interpolation curve, P i Representing the coordinates of the control point, w i Representing the weight of the control point, R i,k (u) represents non-uniformity BA spline basis function, n represents the number of control points, k represents the number of times of B-spline, and u represents interpolation parameters;
the path planning formula is as follows:
where θ (u) represents the directional angle of the welding path at a point, and x '(u) and y' (u) represent the lateral and longitudinal derivatives of the interpolation curve at that point, respectively.
As an optimized technical scheme, the genetic algorithm mainly comprises the following steps:
(1) Initializing: randomly generating parameters such as coordinates, weights, B spline times and the like of a group of individuals, namely a group of control points;
(2) And (3) adaptability evaluation: each individual is brought into a NUBIC interpolation formula to calculate interpolation errors, and the interpolation errors are used as fitness values of the individuals;
(3) Selecting: selecting a part of excellent individuals as parents according to the fitness value;
(4) Crossing: performing cross operation on the parent individuals to generate a group of new offspring individuals;
(5) Variation: performing mutation operation on offspring individuals to generate slightly different individuals;
(6) And (5) repeatedly executing the steps (2) to (5) until the stopping condition is met.
As an optimized technical scheme, the operation steps of feature extraction and selection are as follows,
1. and (3) data acquisition: collecting video images, sensor signals and other data related to the welding process;
2. data preprocessing: denoising, normalizing, smoothing and the like are carried out on the data so as to reduce noise interference;
3. feature extraction: extracting useful features from the preprocessed data;
3.1. video feature extraction: feature information in the image may be extracted using image processing techniques such as color histograms, texture features, shape features, etc.;
3.2. sensor signal feature extraction: the characteristic information in the sensor signal may be extracted using signal processing techniques such as fourier transforms, wavelet transforms, time-frequency analysis, etc.;
4. feature selection: selecting a most representative feature from the extracted features;
4.1. and (3) filtering type feature selection: sorting the features according to indexes such as correlation and importance among the features, and selecting the most representative features;
4.2. and (3) parcel type feature selection: selecting an optimal feature subset by continuously adjusting feature combinations;
4.3. and (3) embedded feature selection: embedding a feature selection process into a model training process, and selecting the most representative features;
5. feature dimension reduction: reducing the feature space of high dimension to low dimension for better classification or regression analysis;
5.1. principal Component Analysis (PCA): mapping the original features into a low-dimensional feature space, and reserving the most representative features;
5.2. linear Discriminant Analysis (LDA): mapping the original features into a low-dimensional feature space while maximizing inter-class distances and minimizing intra-class distances;
5.3. nonlinear dimension reduction (t-SNE): mapping the high-dimensional features into the low-dimensional feature space, and preserving the relative distance between samples.
As an optimized technical solution, the operation of the welding controller comprises the following steps:
sensor signal acquisition: acquiring signals such as current, voltage, temperature, displacement and the like in the welding process by using a sensor to obtain real-time welding parameters;
control signal calculation: according to the collected sensor signals, a control algorithm is used for calculating corresponding control signals such as welding current, welding speed and the like;
control signal output: outputting the calculated control signal to welding equipment to control parameters in the welding process;
feedback control: according to the welding process signals acquired in real time, the control algorithm is corrected, and the stability and the accuracy of the welding process are ensured;
the PID algorithm is adopted, the formula is as follows,
wherein u (t) represents a control signal, e (t) represents a current error, and KP, KI and KD respectively represent proportional, integral and differential coefficients.
The invention has the beneficial effects that:
compared with the prior art, the invention has the following advantages,
(1) Improving welding quality and efficiency: the digital modeling technology can help a welder predict the problems in the welding process in advance, avoid the defects of incomplete welding of a welding line, deformation of the welding line and the like, improve the welding quality, and simultaneously, the flexible welding line recognition technology and the path planning can automate the welding process, adjust the welding path according to the information fed back in real time and improve the welding efficiency;
(2) The human intervention is reduced: the automatic control function of the welding controller can reduce the human intervention of a welder, reduce human errors and improve the welding precision and stability;
(3) The cost is reduced: the intelligent control of the welding controller can accurately control welding parameters, reduce rejection rate and production cost, and simultaneously, the digital modeling technology can reduce secondary processing and reworking through the problem of advanced prediction, so that the production cost is further reduced;
(4) Realizing intelligent production: by adopting modern technologies such as digital modeling, artificial intelligence and the like, the automatic control, monitoring and optimization of the welding process can be realized, thereby realizing the intelligent production of the welding process and improving the production efficiency and quality.
Drawings
FIG. 1 is a flow chart diagram of a welding method based on digital modeling and flexible weld recognition.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and detailed description, wherein it is to be understood that, on the premise of no conflict, the following embodiments or technical features may be arbitrarily combined to form new embodiments.
It should be noted that, the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," and the like refer to an azimuth or a positional relationship based on that shown in the drawings, or that the inventive product is commonly put in place when used, merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific azimuth, be configured and operated in a specific azimuth, and thus should not be construed as limiting the present invention.
In the description of the present invention, it should also be noted that, unless explicitly specified and limited otherwise, the terms "disposed," "mounted," "connected," and "connected" 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 will be understood in specific cases by those of ordinary skill in the art.
Example 1:
a welding method based on digital modeling and flexible weld recognition in this embodiment includes the steps of,
a. determining welding type and welding material: before a three-dimensional mathematical model is established, the type of welding and the welding materials need to be determined;
b. collecting relevant data: collecting relevant data including three-dimensional graphics of the welding piece, coordinate axis position and attitude information of the welding robot, welding parameters and the like, wherein the data are used for constructing a three-dimensional mathematical model;
c. determining a modeling method: determining a method for establishing a three-dimensional mathematical model, and modeling based on a mathematical formula, wherein the method can be selected based on the shape, size, position, posture and other information of a welding piece;
d. establishing a three-dimensional mathematical model: according to the collected data and the determined modeling method, a welded three-dimensional mathematical model is established, wherein the model comprises the shape, size, position, posture and other information of a welding part;
e. adjusting model parameters: according to actual welding conditions and needs, the model is adjusted, including adjustment of parameters such as welding speed, welding angle, welding temperature and the like, specifically, reasonable parameter values can be set according to actual needs by setting corresponding parameter ranges and target values such as welding speed, temperature, angle and the like in a welding controller, then real-time feedback information such as welding quality, welding temperature and the like is acquired through a sensor in the welding process, the information is input into the controller, and welding parameters are automatically adjusted according to the set target values and the feedback information;
f. and (3) verifying the accuracy of the model: verifying the accuracy of the model by comparing the difference between the actual welding result and the model prediction result;
g. and (3) image acquisition: using a camera or a laser scanner to acquire images of the welded workpiece and acquire image information of the welding seam;
h. image preprocessing: preprocessing the acquired welding image, including denoising, graying, binarization, filtering and other operations, so as to improve the image quality and extract the characteristic information of the welding seam;
i. and (3) weld joint identification: carrying out automatic identification on the welding seam through image processing and feature extraction, wherein the automatic identification comprises the steps of determining parameters such as the position, the shape, the size and the like of the welding seam;
j. smoothing the identified weld edge and center line by curve fitting technique to eliminate noise and error and obtain corrected weld parameters
k. Digital modeling: comparing and correcting the identified weld joint position and shape with an original design model by utilizing a digital modeling technology so as to ensure the accuracy and stability of welding parameters;
and l, correcting welding parameters: according to the corrected weld position and shape, welding parameters including welding current, welding speed, welding angle and the like are adjusted to ensure the quality and stability of the weld;
m, welding operation: the hand-held arc welding equipment, the robot welding equipment and the like are utilized to carry out welding operation, so that accurate control and automatic operation of welding parameters are realized;
n, quality detection: and detecting the quality of the welded seam, including the aspects of appearance quality, dimensional accuracy, internal defects of the welded seam and the like, so as to ensure that the welding quality meets the requirements.
As an optimized technical scheme, the system used by the welding method comprises the following parts:
image acquisition equipment: the method comprises the steps of acquiring an image of a welded workpiece;
an image processing apparatus: the method comprises the steps of preprocessing an acquired image and extracting features;
weld recognition model apparatus: the method comprises the steps of classifying and identifying the preprocessed images, and determining the position and shape of a welding line;
correction algorithm equipment: the method comprises the steps of comparing and correcting the identified weld joint position and shape with a digital modeling result, and finally determining welding parameters for welding;
welding equipment: for performing a welding operation;
and (3) a control system: the device is used for carrying out linkage control on the equipment so as to realize accurate adjustment and automatic control of welding parameters;
a mechanical learning model for flexible weld joint identification correction is established, the steps are as follows,
(1) Data acquisition and pretreatment: collecting data of a welding process, including sensor signals, video images, sounds and the like, and preprocessing, including denoising, filtering and normalization;
(2) Feature extraction and selection: extracting representative features from the preprocessed data, including morphological features, gray features and texture features, and selecting a proper feature set;
(3) Data set partitioning: dividing a data set into a training set, a verification set and a test set according to a certain proportion;
(4) Model selection and training: selecting a proper mechanical learning model, adopting a CNN model, and training the model by using a training set;
(5) Model optimization: adjusting parameters of the trained model, including selecting different optimizers, learning rates, batch sizes and the like, so as to improve the accuracy and robustness of the model;
(6) Model evaluation: evaluating the model by using the verification set, wherein the operation comprises calculating indexes such as accuracy, recall rate, F1 value and the like so as to determine the performance of the model;
(7) Model application: applying the trained model to data in a test set, and evaluating generalization capability and actual application effect of the model;
(8) Model update: updating and optimizing the model according to the actual application effect so as to improve the performance and the application value of the model;
(9) Model deployment: deploying the trained model into an actual application environment, such as a welding production line or an automated welding device;
(10) Real-time monitoring and correction: in the practical application process, the welding process is monitored and corrected in real time, for example, the position deviation of a welding line, the welding speed, the welding current and the like are detected, and the model is corrected according to the monitoring result, so that the self-adaptability and the robustness of the model are improved.
As an optimized technical scheme, the CNN model can be used for characteristic extraction and classification recognition of welding seams, and the main formulas comprise a convolution layer, a pooling layer, a full-connection layer and the like, and the specific formulas are as follows:
wherein,,feature map, K, representing ith row and jth column of layer l l Represents the number of layer I convolution kernels, h l And w l Representing the height and width of the layer I convolution kernel,/>Weights representing the ith row and the ith column of the kth convolution kernel of the first layer,values representing the kth feature map of the ith+u-1 row, jth+v-1 column, b of layer 1 l A bias term representing a first layer;
wherein,,representing pooling results of ith row and jth column of the first layer, and s represents a pooling step length;
wherein y is k Represents the output of the layer L, the layer k neuron, J represents the number of layer L-1 neurons,representing the connection weights between the jth neuron of the L-1 layer and the kth neuron of the L-1 layer,>representing the output of the jth neuron of layer L-1, bk represents the bias term of the kth neuron of layer L, and f (·) represents the activation function.
As an optimized technical scheme, when a three-dimensional mathematical model is established, non-uniform rational B-spline interpolation (NUBIC) is adopted to perform curve fitting and path planning, the steps are as follows,
and (3) data acquisition: collecting curve data in the welding process, wherein the curve data comprises information such as coordinates and curvature of a curve;
pretreatment: preprocessing the acquired curve data, including data denoising, smoothing and sampling to reduce noise and improve data quality, and Haar wavelet transform, HOG algorithm and the like can be adopted for extracting characteristic information of welding seams, such as length, width, curvature and the like;
NUBIC interpolation: interpolating the preprocessed curve data by using a non-uniform rational B-spline interpolation method to obtain a continuous and smooth curve representation;
and (3) parameter determination: determining parameters in NUBIC interpolation by using a genetic algorithm, including control point coordinates, weights, the times of B-splines and the like;
path planning: carrying out path planning according to the interpolation curve and the determined parameters to obtain a smooth welding path which meets the welding requirements;
the NUBIC interpolation formula is as follows:
c (u) represents the coordinate of a point on the interpolation curve, P i Representing the coordinates of the control point, w i Representing the weight of the control point, R i,k (u) represents a non-uniform B-spline basis function, n represents the number of control points, k represents the number of times of B-spline, and u represents an interpolation parameter;
the path planning formula is as follows:
where θ (u) represents the directional angle of the welding path at a point, and x '(u) and y' (u) represent the lateral and longitudinal derivatives of the interpolation curve at that point, respectively.
In the embodiment, the digital modeling technology can help a welder predict the problem in the welding process in advance, avoid the defects of incomplete welding of a welding line, deformation of the welding line and the like, improve the welding quality, and simultaneously, the flexible welding line recognition technology and path planning can automate the welding process, adjust the welding path according to the information fed back in real time and improve the welding efficiency;
the automatic control function of the welding controller can reduce the human intervention of a welder, reduce human errors and improve the welding precision and stability;
the intelligent control of the welding controller can accurately control welding parameters, reduce rejection rate and production cost, and simultaneously, the digital modeling technology can reduce secondary processing and reworking through the problem of advanced prediction, so that the production cost is further reduced; by adopting modern technologies such as digital modeling, artificial intelligence and the like, the automatic control, monitoring and optimization of the welding process can be realized, thereby realizing the intelligent production of the welding process and improving the production efficiency and quality.
Example 2:
in this embodiment, the genetic algorithm mainly includes the following steps:
(1) Initializing: randomly generating parameters such as coordinates, weights, B spline times and the like of a group of individuals, namely a group of control points;
(2) And (3) adaptability evaluation: each individual is brought into a NUBIC interpolation formula to calculate interpolation errors, and the interpolation errors are used as fitness values of the individuals;
(3) Selecting: selecting a part of excellent individuals as parents according to the fitness value;
(4) Crossing: performing cross operation on the parent individuals to generate a group of new offspring individuals;
(5) Variation: performing mutation operation on offspring individuals to generate slightly different individuals;
(6) And (5) repeatedly executing the steps (2) to (5) until the stopping condition is met.
Specifically, the operation steps of feature extraction and selection are as follows,
1. and (3) data acquisition: collecting video images, sensor signals and other data related to the welding process;
2. data preprocessing: denoising, normalizing, smoothing and the like are carried out on the data so as to reduce noise interference;
3. feature extraction: extracting useful features from the preprocessed data;
3.1. video feature extraction: feature information in the image may be extracted using image processing techniques such as color histograms, texture features, shape features, etc.;
3.2. sensor signal feature extraction: the characteristic information in the sensor signal may be extracted using signal processing techniques such as fourier transforms, wavelet transforms, time-frequency analysis, etc.;
4. feature selection: selecting a most representative feature from the extracted features;
4.1. and (3) filtering type feature selection: sorting the features according to indexes such as correlation and importance among the features, and selecting the most representative features;
4.2. and (3) parcel type feature selection: selecting an optimal feature subset by continuously adjusting feature combinations;
4.3. and (3) embedded feature selection: embedding a feature selection process into a model training process, and selecting the most representative features;
5. feature dimension reduction: reducing the feature space of high dimension to low dimension for better classification or regression analysis;
5.1. principal Component Analysis (PCA): mapping the original features into a low-dimensional feature space, and reserving the most representative features;
5.2. linear Discriminant Analysis (LDA): mapping the original features into a low-dimensional feature space while maximizing inter-class distances and minimizing intra-class distances;
5.3. nonlinear dimension reduction (t-SNE): mapping the high-dimensional features into the low-dimensional feature space, preserving the relative distance between samples, t-SNE (t-Distributed Stochastic Neighbor Embedding) is a nonlinear dimension reduction algorithm used for mapping the high-dimensional features into the low-dimensional space for visualization and analysis. It is an improvement and extension based on the Stochastic Neighbor Embedding (SNE) algorithm. Compared with other dimension reduction methods, the t-SNE is excellent in terms of preserving the local structure and relative distance of data, the t-SNE represents the original data by creating a set of data points in a low-dimensional space, then using a Gaussian distribution to represent the similarity between the data points, and minimizing the Kullback-Leibler divergence (KL divergence) between the data points in the low-dimensional space by optimizing an objective function, thereby preserving the local structure of the original data to the maximum.
The operation of the welding controller includes the steps of:
sensor signal acquisition: acquiring signals such as current, voltage, temperature, displacement and the like in the welding process by using a sensor to obtain real-time welding parameters;
control signal calculation: according to the collected sensor signals, a control algorithm is used for calculating corresponding control signals such as welding current, welding speed and the like;
control signal output: outputting the calculated control signal to welding equipment to control parameters in the welding process;
feedback control: according to the welding process signals acquired in real time, the control algorithm is corrected, and the stability and the accuracy of the welding process are ensured;
the PID algorithm is adopted, and is a classical control algorithm, and the name of the PID algorithm is from three control parameters: proportional (pro), integral (Integral), and differential (Derivative). The main purpose of the PID algorithm is to achieve the optimal control effect between the actual output and the expected output of the controlled system by adjusting the output signal, in the PID algorithm, the proportional term is adjusted according to the error between the actual output and the expected output, the integral term is adjusted according to the accumulation condition of the error, and the differential term is adjusted according to the change rate of the error. The weights of these three terms may be determined by trial and error or other adjustment methods, which are formulated as,
wherein u (t) represents a control signal, e (t) represents a current error, and KP, KI and KD respectively represent proportional, integral and differential coefficients.
The above embodiments are only preferred embodiments of the present invention, and should not be construed as limiting the scope of the present invention, and various combinations, modifications and equivalents may be made by those skilled in the art without departing from the spirit and scope of the invention, and the scope of the claims should be construed as including the appended claims.
Claims (8)
1. A welding method based on digital modeling and flexible weld joint identification is characterized by comprising the following steps: comprises the steps of,
a. determining welding type and welding material: before a three-dimensional mathematical model is established, the type of welding and the welding materials need to be determined;
b. collecting relevant data: collecting relevant data including three-dimensional graphics of the welding piece, coordinate axis position and attitude information of the welding robot, welding parameters and the like, wherein the data are used for constructing a three-dimensional mathematical model;
c. determining a modeling method: determining a method for establishing a three-dimensional mathematical model, and modeling based on a mathematical formula;
d. establishing a three-dimensional mathematical model: according to the collected data and the determined modeling method, a welded three-dimensional mathematical model is established, wherein the model comprises the shape, size, position, posture and other information of a welding part;
e. adjusting model parameters: according to the actual welding condition and the need, the model is adjusted, including the adjustment of parameters such as welding speed, welding angle, welding temperature and the like;
f. and (3) verifying the accuracy of the model: verifying the accuracy of the model by comparing the difference between the actual welding result and the model prediction result;
g. and (3) image acquisition: using a camera or a laser scanner to acquire images of the welded workpiece and acquire image information of the welding seam;
h. image preprocessing: preprocessing the acquired welding image, including denoising, graying, binarization, filtering and other operations, so as to improve the image quality and extract the characteristic information of the welding seam;
i. and (3) weld joint identification: carrying out automatic identification on the welding seam through image processing and feature extraction, wherein the automatic identification comprises the steps of determining parameters such as the position, the shape, the size and the like of the welding seam;
j. smoothing the identified weld edge and center line by curve fitting technique to eliminate noise and error and obtain corrected weld parameters
k. Digital modeling: comparing and correcting the identified weld joint position and shape with an original design model by utilizing a digital modeling technology so as to ensure the accuracy and stability of welding parameters;
and l, correcting welding parameters: according to the corrected weld position and shape, welding parameters including welding current, welding speed, welding angle and the like are adjusted to ensure the quality and stability of the weld;
m, welding operation: the hand-held arc welding equipment, the robot welding equipment and the like are utilized to carry out welding operation, so that accurate control and automatic operation of welding parameters are realized;
n, quality detection: and detecting the quality of the welded seam, including the aspects of appearance quality, dimensional accuracy, internal defects of the welded seam and the like, so as to ensure that the welding quality meets the requirements.
2. The welding method based on digital modeling and flexible weld recognition as defined in claim 1, wherein: the system used by the welding method comprises the following parts:
image acquisition equipment: the method comprises the steps of acquiring an image of a welded workpiece;
an image processing apparatus: the method comprises the steps of preprocessing an acquired image and extracting features;
weld recognition model apparatus: the method comprises the steps of classifying and identifying the preprocessed images, and determining the position and shape of a welding line;
correction algorithm equipment: the method comprises the steps of comparing and correcting the identified weld joint position and shape with a digital modeling result, and finally determining welding parameters for welding;
welding equipment: for performing a welding operation;
and (3) a control system: the device is used for carrying out linkage control on the equipment and realizing accurate adjustment and automatic control of welding parameters.
3. The welding method based on digital modeling and flexible weld recognition as defined in claim 1, wherein: the method also comprises the steps of establishing a mechanical learning model for flexible weld joint identification correction,
(1) Data acquisition and pretreatment: collecting data of a welding process, including sensor signals, video images, sounds and the like, and preprocessing, including denoising, filtering and normalization;
(2) Feature extraction and selection: extracting representative features from the preprocessed data, including morphological features, gray features and texture features, and selecting a proper feature set;
(3) Data set partitioning: dividing a data set into a training set, a verification set and a test set according to a certain proportion;
(4) Model selection and training: selecting a proper mechanical learning model, adopting a CNN model, and training the model by using a training set;
(5) Model optimization: adjusting parameters of the trained model, including selecting different optimizers, learning rates, batch sizes and the like, so as to improve the accuracy and robustness of the model;
(6) Model evaluation: evaluating the model by using the verification set, wherein the operation comprises calculating indexes such as accuracy, recall rate, F1 value and the like so as to determine the performance of the model;
(7) Model application: applying the trained model to data in a test set, and evaluating generalization capability and actual application effect of the model;
(8) Model update: updating and optimizing the model according to the actual application effect so as to improve the performance and the application value of the model;
(9) Model deployment: deploying the trained model into an actual application environment, such as a welding production line or an automated welding device;
(10) Real-time monitoring and correction: in the practical application process, the welding process is monitored and corrected in real time, for example, the position deviation of a welding line, the welding speed, the welding current and the like are detected, and the model is corrected according to the monitoring result, so that the self-adaptability and the robustness of the model are improved.
4. A welding method based on digital modeling and flexible weld recognition as defined in claim 3, wherein: the CNN model can be used for characteristic extraction and classification recognition of welding seams, and the main formulas comprise a convolution layer, a pooling layer, a full connection layer and the like, and the specific formulas are as follows:
wherein,,feature map, K, representing ith row and jth column of layer l l Represents the number of layer I convolution kernels, h l And w l Representing the height and width of the convolution kernel of layer I, < >>Weights representing the ith row and the ith column of the kth convolution kernel of the first layer,values representing the kth feature map of the ith+u-1 row, jth+v-1 column, b of layer 1 l A bias term representing a first layer;
wherein,,representing pooling results of ith row and jth column of the first layer, and s represents a pooling step length;
wherein y is k Represents the output of the layer L, the layer k neuron, J represents the number of layer L-1 neurons,representing the connection weights between the jth neuron of the L-1 layer and the kth neuron of the L-1 layer,>representing the output of the jth neuron of layer L-1, bk represents the bias term of the kth neuron of layer L, and f (·) represents the activation function.
5. The welding method based on digital modeling and flexible weld recognition as defined in claim 1, wherein: when the three-dimensional mathematical model is established, non-uniform rational B-spline interpolation (NUBIC) is adopted to perform curve fitting and path planning, the steps are as follows,
and (3) data acquisition: collecting curve data in the welding process, wherein the curve data comprises information such as coordinates and curvature of a curve;
pretreatment: preprocessing the acquired curve data, including data denoising, smoothing and sampling, so as to reduce noise and improve data quality;
NUBIC interpolation: interpolating the preprocessed curve data by using a non-uniform rational B-spline interpolation method to obtain a continuous and smooth curve representation;
and (3) parameter determination: determining parameters in NUBIC interpolation by using a genetic algorithm, including control point coordinates, weights, the times of B-splines and the like;
path planning: carrying out path planning according to the interpolation curve and the determined parameters to obtain a smooth welding path which meets the welding requirements;
the NUBIC interpolation formula is as follows:
c (u) represents the coordinate of a point on the interpolation curve, P i Representing the coordinates of the control point, w i Representing the weight of the control point, R i,k (u) represents a non-uniform B-spline basis function, n represents the number of control points, k represents the number of times of B-spline, and u represents an interpolation parameter;
the path planning formula is as follows:
6. The welding method based on digital modeling and flexible weld recognition according to claim 5, wherein the welding method comprises the following steps: the genetic algorithm mainly comprises the following steps:
(1) Initializing: randomly generating parameters such as coordinates, weights, B spline times and the like of a group of individuals, namely a group of control points;
(2) And (3) adaptability evaluation: each individual is brought into a NUBIC interpolation formula to calculate interpolation errors, and the interpolation errors are used as fitness values of the individuals;
(3) Selecting: selecting a part of excellent individuals as parents according to the fitness value;
(4) Crossing: performing cross operation on the parent individuals to generate a group of new offspring individuals;
(5) Variation: performing mutation operation on offspring individuals to generate slightly different individuals;
(6) And (5) repeatedly executing the steps (2) to (5) until the stopping condition is met.
7. A welding method based on digital modeling and flexible weld recognition as defined in claim 3, wherein: the specific operation steps of the feature extraction and selection are as follows,
1. and (3) data acquisition: collecting video images, sensor signals and other data related to the welding process;
2. data preprocessing: denoising, normalizing, smoothing and the like are carried out on the data so as to reduce noise interference;
3. feature extraction: extracting useful features from the preprocessed data;
3.1. video feature extraction: feature information in the image may be extracted using image processing techniques such as color histograms, texture features, shape features, etc.;
3.2. sensor signal feature extraction: the characteristic information in the sensor signal may be extracted using signal processing techniques such as fourier transforms, wavelet transforms, time-frequency analysis, etc.;
4. feature selection: selecting a most representative feature from the extracted features;
4.1. and (3) filtering type feature selection: sorting the features according to indexes such as correlation and importance among the features, and selecting the most representative features;
4.2. and (3) parcel type feature selection: selecting an optimal feature subset by continuously adjusting feature combinations;
4.3. and (3) embedded feature selection: embedding a feature selection process into a model training process, and selecting the most representative features;
5. feature dimension reduction: reducing the feature space of high dimension to low dimension for better classification or regression analysis;
5.1. principal Component Analysis (PCA): mapping the original features into a low-dimensional feature space, and reserving the most representative features;
5.2. linear Discriminant Analysis (LDA): mapping the original features into a low-dimensional feature space while maximizing inter-class distances and minimizing intra-class distances;
5.3. nonlinear dimension reduction (t-SNE): mapping the high-dimensional features into the low-dimensional feature space, and preserving the relative distance between samples.
8. The welding method based on digital modeling and flexible weld recognition as defined in claim 1, wherein: the operation of the welding controller includes the steps of:
sensor signal acquisition: acquiring signals such as current, voltage, temperature, displacement and the like in the welding process by using a sensor to obtain real-time welding parameters;
control signal calculation: according to the collected sensor signals, a control algorithm is used for calculating corresponding control signals such as welding current, welding speed and the like;
control signal output: outputting the calculated control signal to welding equipment to control parameters in the welding process;
feedback control: according to the welding process signals acquired in real time, the control algorithm is corrected, and the stability and the accuracy of the welding process are ensured;
the PID algorithm is adopted, the formula is as follows,
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310377717.7A CN116275742A (en) | 2023-04-11 | 2023-04-11 | Welding method based on digital modeling and flexible weld joint recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310377717.7A CN116275742A (en) | 2023-04-11 | 2023-04-11 | Welding method based on digital modeling and flexible weld joint recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116275742A true CN116275742A (en) | 2023-06-23 |
Family
ID=86830649
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310377717.7A Withdrawn CN116275742A (en) | 2023-04-11 | 2023-04-11 | Welding method based on digital modeling and flexible weld joint recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116275742A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116967649A (en) * | 2023-09-24 | 2023-10-31 | 新研氢能源科技有限公司 | Intelligent welding method and system for metal bipolar plate |
CN118023791A (en) * | 2024-04-11 | 2024-05-14 | 常州市闳晖科技发展股份有限公司 | Welding method and system for precise shell |
-
2023
- 2023-04-11 CN CN202310377717.7A patent/CN116275742A/en not_active Withdrawn
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116967649A (en) * | 2023-09-24 | 2023-10-31 | 新研氢能源科技有限公司 | Intelligent welding method and system for metal bipolar plate |
CN116967649B (en) * | 2023-09-24 | 2023-12-26 | 新研氢能源科技有限公司 | Intelligent welding method and system for metal bipolar plate |
CN118023791A (en) * | 2024-04-11 | 2024-05-14 | 常州市闳晖科技发展股份有限公司 | Welding method and system for precise shell |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116275742A (en) | Welding method based on digital modeling and flexible weld joint recognition | |
CN109308693B (en) | Single-binocular vision system for target detection and pose measurement constructed by one PTZ camera | |
CN108537808B (en) | A kind of gluing online test method based on robot teaching point information | |
USRE47557E1 (en) | Method and computer system for characterizing a sheet metal part | |
CN107230203B (en) | Casting defect identification method based on human eye visual attention mechanism | |
JP4681426B2 (en) | Apparatus and method for analyzing relation between operation and quality in manufacturing process, computer program, and computer-readable recording medium | |
CN103425988B (en) | Real-time positioning and matching method with arc geometric primitives | |
CN105740899A (en) | Machine vision image characteristic point detection and matching combination optimization method | |
CN111815555A (en) | Metal additive manufacturing image detection method and device combining anti-neural network with local binary | |
CN113920060A (en) | Autonomous operation method and device for welding robot, electronic device, and storage medium | |
CN114706358B (en) | Method and system for processing welding process information of straight welded pipe | |
CN110751604A (en) | Machine vision-based steel pipe weld defect online detection method | |
CN109344768A (en) | Pointer breaker recognition methods based on crusing robot | |
Yang et al. | Detection of weld groove edge based on multilayer convolution neural network | |
JP4653547B2 (en) | Apparatus and method for analyzing relation between operation and quality in manufacturing process, computer program, and computer-readable recording medium | |
Ma et al. | An efficient and robust complex weld seam feature point extraction method for seam tracking and posture adjustment | |
CN115239760A (en) | Target tracking method, system, equipment and storage medium | |
CN114926633A (en) | High-precision pig weight estimation method based on deep learning | |
CN105224941A (en) | Process identification and localization method | |
CN117817111A (en) | Method and system for intelligently identifying and matching process parameters in laser welding | |
Zirngibl et al. | Approach for the automated analysis of geometrical clinch joint characteristics | |
CN113210936B (en) | Welding method and system for straight welded pipe and readable medium | |
CN114742975B (en) | Vehicle-mounted image rail curve modeling method | |
CN117943256B (en) | Self-adaptive gluing process control method and system | |
CN115239657B (en) | Industrial part increment identification method based on deep learning target segmentation |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230623 |
|
WW01 | Invention patent application withdrawn after publication |