GAN ENABLED WELDING

Shailendra Moyal; Sarbajit K. Rakshit; Partho Ghosh

Patent

Atlas literature

Patent

US 12,254,619 B2

GAN ENABLED WELDING

Shailendra Moyal, Sarbajit K. Rakshit, Partho Ghosh

International Business Machines Corporation, Armonk, NY (US)·Mar. 18, 2025·US

Drawings

Patent drawings and their descriptions. Click a drawing to enlarge it.

FIG. 1

FIG. 1 illustrates a ﬂowchart of an example method for 50 performing a welding procedure, in accordance with embodiments of the present disclosure.

FIG. 2

FIG. 2 depicts a block diagram illustrating an example computing environment in which illustrative embodiments of the present disclosure can be implemented, in …

FIG. 3

FIG. 3 depicts a high-level block diagram illustrating an example computer system that can be used in implementing one or more of the methods, tools, modules, …

FIG. 4

FIG. 4 depicts a cloud computing environment, in accor- dance with embodiments of the present disclosure.

FIG. 5

FIG. 5 depicts abstraction model layers, in accordance with embodiments of the present disclosure. 65 While the embodiments described herein are amenable to …

Claims

Claims define the patent's legal scope. Independent claims stand alone; dependent claims (nested) narrow them. Click a claim to expand its dependents.

3 independent · 17 dependent

1
Independent
A method, comprising: determining conditions of a welding procedure using a virtual image of a structure on which the welding procedure is to be performed, wherein the virtual image is a photograph of the structure; generating speciﬁcations for performing the welding pro-cedure based on the conditions; generating a virtual reference shape based on the condi-tions and the speciﬁcations, the virtual reference shape generated by modifying the virtual image and the virtual reference shape is a desired resulting shape; comparing a resulting shape of the welding procedure with the virtual reference shape; determining a quality of the welding procedure based on the comparison; and storing data pertaining to the welding procedure.
1. 2
  Dependent← claim 1
  The method of claim 1, wherein: generating the virtual reference shape includes utilizing a generative adversarial network.
2. 3
  Dependent← claim 1
  The method of claim 1, further comprising: performing the welding procedure based on the speciﬁ-cations and the virtual reference shape.
3. 4
  Dependent← claim 1
  The method of claim 1, further comprising: determining further conditions of a further welding pro-cedure; and generating further speciﬁcations for performing the fur-ther welding procedure based on the further conditions, wherein generating the further speciﬁcations includes utilizing the stored data pertaining to the quality of the welding procedure.
4. 7
  Dependent← claim 1
  The method of claim 1, wherein: determining the quality of the welding procedure includes generating a score representing the quality of the weld-ing procedure; and storing the data includes storing the score.
8
Independent
A system comprising: one or more processors; and one or more computer-readable storage media collectively storing program instructions which, when executed by the one or more processors, are conﬁgured to cause the one or more processors to perform a method compris-ing: determining conditions of a welding procedure using a virtual image of a structure on which the welding procedure is to be performed, wherein the virtual image is a photograph of the structure; generating speciﬁcations for performing the welding procedure based on the conditions; generating a virtual reference shape based on the con-ditions and the speciﬁcations, the virtual reference shape generated by modifying the virtual image and the virtual reference shape is a desired resulting shape; comparing a resulting shape of the welding procedure with the virtual reference shape; determining a quality of the welding procedure based on the comparison; and storing data pertaining to the welding procedure.
1. 9
  Dependent← claim 8
  The system of claim 8, wherein: generating the virtual reference shape includes utilizing a generative adversarial network.
2. 10
  Dependent← claim 8
  The system of claim 8, wherein the method further comprises: performing the welding procedure based on the speciﬁ-cations and the virtual reference shape.
3. 11
  Dependent← claim 8
  The system of claim 8, wherein the method further comprises: determining further conditions of a further welding pro-cedure; and generating further speciﬁcations for performing the fur-ther welding procedure based on the further conditions, wherein generating the further speciﬁcations includes utilizing the stored data pertaining to the quality of the welding procedure.
4. 14
  Dependent← claim 8
  The system of claim 8, wherein: determining the quality of the welding procedure includes generating a score representing the quality of the weld-ing procedure; and storing the data includes storing the score.
15
Independent
A computer program product comprising one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising instruc-tions conﬁgured to cause one or more processors to perform a method comprising: determining conditions of a welding procedure using a virtual image of a structure on which the welding procedure is to be performed, wherein the virtual image is a photograph of the structure; generating speciﬁcations for performing the welding pro-cedure based on the conditions; generating a virtual reference shape based on the condi-tions and the speciﬁcations, the virtual reference shape generated by modifying the virtual image and the virtual reference shape is a desired resulting shape; comparing a resulting shape of the welding procedure with the virtual reference shape; determining a quality of the welding procedure based on the comparison; and storing data pertaining to the welding procedure.
1. 16
  Dependent← claim 15
  The computer program product of claim 15, wherein: generating the virtual reference shape includes utilizing a generative adversarial network.
2. 17
  Dependent← claim 15
  The computer program product of claim 15, wherein the method further comprises: performing the welding procedure based on the speciﬁ-cations and the virtual reference shape.
3. 18
  Dependent← claim 15
  The computer program product of claim 15, wherein the method further comprises: determining further conditions of a further welding pro-cedure; and generating further speciﬁcations for performing the fur-ther welding procedure based on the further conditions, wherein generating the further speciﬁcations includes utilizing the stored data pertaining to the quality of the welding procedure.
4. 20
  Dependent← claim 15
  The computer program product of claim 15, wherein: the method further comprises determining the quality of the welding procedure includes generating a score representing the quality of the welding procedure; and storing the data includes storing the score. ∗ ∗ ∗ ∗ ∗

Characterization

Measurements and analyses referenced in the patent, with their drawing references.

durability test

Durability

durability and life of dynamically welded struc-tures is often determined, in part, by the welds. Accordingly, improving the quality of welding procedures can have long-reaching impacts on the welded structures. FIG. 1 depicts a ﬂowchart illustrating an exampl

Cited prior art

Patents and literature cited by this patent (applicant and examiner references).

Cited patents · 5

US 2017/0200395 A12017/0200395 A1 * 7/2017 Albrecht............... G06F 3/0488examiner
US 2021/0312832 A12021/0312832 A1 * 10/2021 Salsich.................. G09B 19/24examiner
US 2022/0035961 A12022/0035961 A1 * 2/2022 Ziabari.................. G06N 3/045examiner
US 2023/0056400 A12023/0056400 A1 * 2/2023 Chakraborty.......... B23K 9/125examiner

Why these are connected

Related documents with shared materials, methods, properties, or citations.

GAN ENABLED WELDING | Matter42 Literature

Cited non-patent literature · 10

Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Lei Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin”, 2020, Wiley, vol. 2020, Article ID 3758730, 13 pages (Year: 2020).
A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery. Brownlee, J., “A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery,” https://machinelearningmastery.com/what-are-generative-adversarial- networks-gans/, Jul. 19, 2019, 15 pgs.
Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion. Fan et al., “Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion,” https://www.hindawi.com/journals/scanning/2020/3758730/, Sensors 2021, 21, 13 pgs.
Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence. Gantala et al., “Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence,” https://publications.iitm.ac.in/publication/automated-defect-recognition- for-welds-using-simulation-assisted, published online Feb. 25, 2021, 24 pgs.
A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition. Gao et al., “A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition,” DOI 10.1109/TII.2020.3008703, IEEE Transactions on Industrial Informatics, https://www.researchgate.net/publication/342912153_A_Generative_Adversarial_Network_Based_Deep_ Learning_Method_for_Low-Quality_Defect_Image_Reconstruction_ and_Recognition, 10 pgs.10.1109/TII.2020.3008703
A Laser-Based Vision System for Weld Quality Inspection. Huang et al., “A Laser-Based Vision System for Weld Quality Inspection,” Sensors 2011, 11, 506-521; doi: 10.3390/s110100506, https://www.mdpi.com/1424-8220/11/1/506.10.3390/s110100506
Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin,” https://www.hindawi.com/journals/scanning/2020/3758730/, Hindawi, Scan- ning, vol. 2020, Article ID 3758730, 13 pgs.
The NIST Deﬁnition of Cloud Computing. Mell et al., “The NIST Deﬁnition of Cloud Computing,” Recom- mendations of the National Institute of Standards and Technology, U.S. Department of Commerce, Special Publication 800-145, Sep. 2011, 7 pgs.
Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography. Naddaf et al., “Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography,” https://www.researchgate. net/publication/340234164_Next-Generation_of_Weld_Quality_ Assessment_Using_Deep_Learning_and_Digital_Radiography, Mar. 2020, 5 pgs.
Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection. Zhang et al., “Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection,” https://openaccess.thecvf.com/content/WACV2021/papers/Zhang_Defect-GAN_High-Fidelity_Defect_ Synthesis_for_Automated_Defect_Inspection_WACV_2021_paper. pdf, pp. 2524-2534, Jan. 2021.

Cited non-patent literature · 10

Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Lei Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin”, 2020, Wiley, vol. 2020, Article ID 3758730, 13 pages (Year: 2020).
A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery. Brownlee, J., “A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery,” https://machinelearningmastery.com/what-are-generative-adversarial- networks-gans/, Jul. 19, 2019, 15 pgs.
Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion. Fan et al., “Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion,” https://www.hindawi.com/journals/scanning/2020/3758730/, Sensors 2021, 21, 13 pgs.
Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence. Gantala et al., “Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence,” https://publications.iitm.ac.in/publication/automated-defect-recognition- for-welds-using-simulation-assisted, published online Feb. 25, 2021, 24 pgs.
A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition. Gao et al., “A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition,” DOI 10.1109/TII.2020.3008703, IEEE Transactions on Industrial Informatics, https://www.researchgate.net/publication/342912153_A_Generative_Adversarial_Network_Based_Deep_ Learning_Method_for_Low-Quality_Defect_Image_Reconstruction_ and_Recognition, 10 pgs.10.1109/TII.2020.3008703
A Laser-Based Vision System for Weld Quality Inspection. Huang et al., “A Laser-Based Vision System for Weld Quality Inspection,” Sensors 2011, 11, 506-521; doi: 10.3390/s110100506, https://www.mdpi.com/1424-8220/11/1/506.10.3390/s110100506
Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin,” https://www.hindawi.com/journals/scanning/2020/3758730/, Hindawi, Scan- ning, vol. 2020, Article ID 3758730, 13 pgs.
The NIST Deﬁnition of Cloud Computing. Mell et al., “The NIST Deﬁnition of Cloud Computing,” Recom- mendations of the National Institute of Standards and Technology, U.S. Department of Commerce, Special Publication 800-145, Sep. 2011, 7 pgs.
Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography. Naddaf et al., “Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography,” https://www.researchgate. net/publication/340234164_Next-Generation_of_Weld_Quality_ Assessment_Using_Deep_Learning_and_Digital_Radiography, Mar. 2020, 5 pgs.
Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection. Zhang et al., “Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection,” https://openaccess.thecvf.com/content/WACV2021/papers/Zhang_Defect-GAN_High-Fidelity_Defect_ Synthesis_for_Automated_Defect_Inspection_WACV_2021_paper. pdf, pp. 2524-2534, Jan. 2021.

Cited non-patent literature · 10

Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Lei Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin”, 2020, Wiley, vol. 2020, Article ID 3758730, 13 pages (Year: 2020).
A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery. Brownlee, J., “A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery,” https://machinelearningmastery.com/what-are-generative-adversarial- networks-gans/, Jul. 19, 2019, 15 pgs.
Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion. Fan et al., “Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion,” https://www.hindawi.com/journals/scanning/2020/3758730/, Sensors 2021, 21, 13 pgs.
Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence. Gantala et al., “Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence,” https://publications.iitm.ac.in/publication/automated-defect-recognition- for-welds-using-simulation-assisted, published online Feb. 25, 2021, 24 pgs.
A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition. Gao et al., “A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition,” DOI 10.1109/TII.2020.3008703, IEEE Transactions on Industrial Informatics, https://www.researchgate.net/publication/342912153_A_Generative_Adversarial_Network_Based_Deep_ Learning_Method_for_Low-Quality_Defect_Image_Reconstruction_ and_Recognition, 10 pgs.10.1109/TII.2020.3008703
A Laser-Based Vision System for Weld Quality Inspection. Huang et al., “A Laser-Based Vision System for Weld Quality Inspection,” Sensors 2011, 11, 506-521; doi: 10.3390/s110100506, https://www.mdpi.com/1424-8220/11/1/506.10.3390/s110100506
Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin,” https://www.hindawi.com/journals/scanning/2020/3758730/, Hindawi, Scan- ning, vol. 2020, Article ID 3758730, 13 pgs.
The NIST Deﬁnition of Cloud Computing. Mell et al., “The NIST Deﬁnition of Cloud Computing,” Recom- mendations of the National Institute of Standards and Technology, U.S. Department of Commerce, Special Publication 800-145, Sep. 2011, 7 pgs.
Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography. Naddaf et al., “Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography,” https://www.researchgate. net/publication/340234164_Next-Generation_of_Weld_Quality_ Assessment_Using_Deep_Learning_and_Digital_Radiography, Mar. 2020, 5 pgs.
Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection. Zhang et al., “Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection,” https://openaccess.thecvf.com/content/WACV2021/papers/Zhang_Defect-GAN_High-Fidelity_Defect_ Synthesis_for_Automated_Defect_Inspection_WACV_2021_paper. pdf, pp. 2524-2534, Jan. 2021.

Cited non-patent literature · 10

Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Lei Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin”, 2020, Wiley, vol. 2020, Article ID 3758730, 13 pages (Year: 2020).
A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery. Brownlee, J., “A Gentle Introduction to Generative Adversarial Networks (GANs)—Machine Learning Mastery,” https://machinelearningmastery.com/what-are-generative-adversarial- networks-gans/, Jul. 19, 2019, 15 pgs.
Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion. Fan et al., “Real-Time High-Performance Laser Welding Defect Detection by Combining ACGAN-Based Data Enhancement and Multi-Model Fusion,” https://www.hindawi.com/journals/scanning/2020/3758730/, Sensors 2021, 21, 13 pgs.
Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence. Gantala et al., “Automated Defect Recognition for Welds Using Simulation Assisted TFM Imaging with Artiﬁcial Intelligence,” https://publications.iitm.ac.in/publication/automated-defect-recognition- for-welds-using-simulation-assisted, published online Feb. 25, 2021, 24 pgs.
A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition. Gao et al., “A Generative Adversarial Network-based Deep Learn- ing Method for Low-quality Defect Image Reconstruction and Recognition,” DOI 10.1109/TII.2020.3008703, IEEE Transactions on Industrial Informatics, https://www.researchgate.net/publication/342912153_A_Generative_Adversarial_Network_Based_Deep_ Learning_Method_for_Low-Quality_Defect_Image_Reconstruction_ and_Recognition, 10 pgs.10.1109/TII.2020.3008703
A Laser-Based Vision System for Weld Quality Inspection. Huang et al., “A Laser-Based Vision System for Weld Quality Inspection,” Sensors 2011, 11, 506-521; doi: 10.3390/s110100506, https://www.mdpi.com/1424-8220/11/1/506.10.3390/s110100506
Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin. Li et al., “Quality Prediction and Control of Assembly and Welding Process for Ship Group Product Based on Digital Twin,” https://www.hindawi.com/journals/scanning/2020/3758730/, Hindawi, Scan- ning, vol. 2020, Article ID 3758730, 13 pgs.
The NIST Deﬁnition of Cloud Computing. Mell et al., “The NIST Deﬁnition of Cloud Computing,” Recom- mendations of the National Institute of Standards and Technology, U.S. Department of Commerce, Special Publication 800-145, Sep. 2011, 7 pgs.
Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography. Naddaf et al., “Next-Generation of Weld Quality Assessment Using Deep Learning and Digital Radiography,” https://www.researchgate. net/publication/340234164_Next-Generation_of_Weld_Quality_ Assessment_Using_Deep_Learning_and_Digital_Radiography, Mar. 2020, 5 pgs.
Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection. Zhang et al., “Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection,” https://openaccess.thecvf.com/content/WACV2021/papers/Zhang_Defect-GAN_High-Fidelity_Defect_ Synthesis_for_Automated_Defect_Inspection_WACV_2021_paper. pdf, pp. 2524-2534, Jan. 2021.