Quality management of laser cladding processes for additive manufacturing by new methods of visualisation and evaluation of thermographic data

Wargulski, Dan R.; Nowak, Torsten; Thiele, Magnus; Dobbelstein, Henrik; Dillkötter, David; Stoppok, Johann; Schacht, R.; Ras, Mohamad Abo

doi:10.1080/17686733.2019.1592392

Cited by 3 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…is the probability of defects. The segmentation error is calculated using equation (7), and the parameters are updated by the backpropagation algorithm to create a closed loop, and the optimal parameters of the model are finally obtained after several iterations of corrections to make the model converge in the correct direction.…”

Section: Network Loss Function Designmentioning

confidence: 99%

“…SWIR cameras are more advantageous in the process of monitoring LMD with the advantages of accurate detection, small size, and cheap price. Wargulski et al [7] proposed a new laser metal deposition method by combining the melt pool temperature and powder nozzle position, which can accurately measure the temperature without the need for additional sensors. Lane et al [8] employed an infrared thermal imaging camera to monitor the melt pool temperature during the cladding process and processed the collected monitoring data using a time-domain analysis method to achieve process feedback to the melt pool area.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Structure of a semantic segmentation-based defect detection network for laser cladding infrared images

Deng

Gao

Mao

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

When choosing the most suitable infrared thermal imaging detection scheme for online inspection during laser cladding processing, this paper designs the RespathU-net semantic segmentation defect detection network for cladding coating defects in infrared images. The network is based on the U-net network framework, and is optimized and improved by redesigning the coding network structure, expanding the network perceptual field, and connecting the paths of residuals, which improves the segmentation effect on the defective areas of the melt coating by improving the problems that the original network cannot realize the end-to-end output and the poor segmentation effect on the complex objects. Through the Kolektor SDD dataset and the infrared dataset constructed in this paper, the generalization performance test and defect detection experiment of the RespathU-net network were carried out, and the designed network was compared with Fully Convolutional Networks(FCN), SegNet, U-net and DeepLab_V3 in terms of average exchange ratio, similarity coefficient and running time, and the results showed that the proposed RespathU-net achieved good multi-size feature recognition. The effect is much better than other semantic segmentation networks, and we verify the actual defect detection accuracy of the designed network as 87.01% through experiments.

show abstract

Section: Network Loss Function Designmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Structure of a semantic segmentation-based defect detection network for laser cladding infrared images

Deng

Gao

Mao

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…In addition to the surveillance of process parameters, different in-process NDT techniques can be implemented. Thermography has been proven to be a valuable tool for melt pool control, especially in metal AM processes [1,2,6,7]. It allows the determination of the melt pool temperature, its dimensions and dynamics as well as cooling rates and spatial temperature gradients.…”

Section: Introductionmentioning

confidence: 99%

In-situ monitoring of a laser metal deposition (LMD) process: comparison of MWIR, SWIR and high-speed NIR thermography

Altenburg

Straße

Gumenyuk

et al. 2020

Quantitative InfraRed Thermography Journal

View full text Add to dashboard Cite

Additive manufacturing offers a range of novel applications. However, the manufacturing process is complex and the production of almost defect-free parts with high reliability and durability is still a challenge. Thermography is a valuable tool for process surveillance, especially in metal additive manufacturing processes. The high process temperatures allow one to use cameras usually operating in the visible spectral range. Here, we compare the results of measurements during the manufacturing process of a commercial laser metal deposition setup using a mid wavelength infrared camera with those from a short wavelength infrared camera and those from a visual spectrum high-speed camera with band pass filter in the near infrared range.

show abstract

A multi-sensor based online monitoring system for laser hot-wire surface cladding process

Yao,

Sheng,

Kong

et al. 2024

Optics & Laser Technology

View full text Add to dashboard Cite

Quality management of laser cladding processes for additive manufacturing by new methods of visualisation and evaluation of thermographic data

Cited by 3 publications

References 10 publications

Structure of a semantic segmentation-based defect detection network for laser cladding infrared images

Structure of a semantic segmentation-based defect detection network for laser cladding infrared images

In-situ monitoring of a laser metal deposition (LMD) process: comparison of MWIR, SWIR and high-speed NIR thermography

A multi-sensor based online monitoring system for laser hot-wire surface cladding process

Contact Info

Product

Resources

About