Deep transfer learning of additive manufacturing mechanisms across materials in metal-based laser powder bed fusion process

Pandiyan, Vigneashwara; Drissi-Daoudi, Rita; Shevchik, Sergey; Masinelli, Giulio; Le-Quang, Tri; Logé, Roland E.; Wasmer, Kilian

doi:10.1016/j.jmatprotec.2022.117531

Cited by 66 publications

(14 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The windows of respective lengths were computed across all three processing regimes without overlaps. The length of the windows or the size of the input to train the hybrid DL network were decided based on the authors' previous works [50,70,91]. The choice of windows length, i.e., data points to train the CNN-LSTM network, is not related to the sampling frequency but rather to the resolution of the monitoring strategy.…”

Section: Dataset and Methodologymentioning

confidence: 99%

Deep Learning-Based Monitoring of Laser Powder Bed Fusion Process on Variable Time-Scales Using Heterogeneous Sensing and Operando X-Ray Radiography Guidance

pandiyan¹,

Masinelli²,

Claire³

et al. 2022

SSRN Journal

Self Cite

View full text Add to dashboard Cite

Section: Dataset and Methodologymentioning

confidence: 99%

Deep Learning-Based Monitoring of Laser Powder Bed Fusion Process on Variable Time-Scales Using Heterogeneous Sensing and Operando X-Ray Radiography Guidance

pandiyan¹,

Masinelli²,

Claire³

et al. 2022

SSRN Journal

Self Cite

View full text Add to dashboard Cite

“…The main goal is to reduce the amount of data needed for the next material as well as the computational efforts. In this section, we will just summarize one possible approach proposed by Pandiyan et al [35], which is also known as deep transfer learning. Actually, transfer learning can be defined as an AI method where a model developed for a specific task is reused as the starting point for a new model on a second task.…”

Section: Transfer Learning Of Process Regimes Across Materials During...mentioning

confidence: 99%

“…Actually, transfer learning can be defined as an AI method where a model developed for a specific task is reused as the starting point for a new model on a second task. In Pandiyan et al [35], the task can be referred to as two different materials; stainless steel (316L) and bronze (CuSn8). Also the major four process regimes were selected: balling, LoF pores, conduction mode and keyhole pores.…”

Section: Transfer Learning Of Process Regimes Across Materials During...mentioning

confidence: 99%

When am (additive manufacturing) meets ae (acoustic emission) and AI (artificial intelligence)

Wasmer¹,

Drissi-daoudi²,

Masinelli³

et al. 2023

eJNDT

View full text Add to dashboard Cite

Acoustic Emission (AE) is an effective method to monitor and control the quality in different technical processes and phenomena, including tribology and fracture mechanics. However, in highly dynamic processes such as Laser Additive Manufacturing (LAM) of metal, the processing of AE signals is very burdensome. At the same time, artificial intelligence (AI) has been considered as a new and powerful tool to overcome the complexity of the large data processing with a reasonable computational time. In this contribution, we have summarized our prior works. To start with, we demonstrated that the combination of AE with state-of-the-art signal processing including AI makes it possible to differentiate several process conditions, workpiece quality and materials. Then, we present some limitations when changing process parameters and materials. Finally, we also introduce alternative AI methods to reduce the amount of data needed to train the AI algorithms as well as transfer the knowledge from one material to another one. This will give the reader an overview of the advances of monitoring the LAM process combining AE with AI techniques to make a significant step forward in in situ and real-time process monitoring and quality control. In particular,

show abstract

“…Processing the raw data acquired from sensory modules is a crucial step in generating awareness and extracting knowledge from the undergoing processes. Researchers proposed various methods, from conventional thresholding and image processing techniques (Baumann & Roller, 2016;Borish et al, 2020;Harbig et al, 2022) to the latest DL based algorithms Pandiyan et al, 2022). Traditional methods require manual or semi-manual filters and feature generation and selection, which in small-scale problems would be feasible.…”

Section: Introductionmentioning

confidence: 99%

In-Situ Quality Assessment and Control in Additive Manufacturing Processes Using Laser Surface Profilometer and Deep Learning Techniques

Akhavan

Lyu

Manoochehri

2022

Preprint

View full text Add to dashboard Cite

Additive Manufacturing (AM) has become one of the most popular manufacturing techniques in various fields. Their layer-by-layer printing process allows an easier fabrication of complex geometries. However, the quality and accuracy of fabricated artifacts in these techniques have low repeatability. In the era of Industry 4.0 by using the emerging sensory and data processing capabilities such as Laser Surface Profilometer (LSP) and Deep Learning (DL), it is possible to improve the repeatability and quality of AM processes. This work presents an in-situ quality assessment and improvement using LSP for data acquisition and DL for data processing and decision making. The utilized LSP module generates a point cloud dataset containing information about the top surface geometry and quality. Once the point cloud data is preprocessed, an improved deep Hybrid Convolutional Auto-Encoder decoder (HCAE) model is used to perform the artifact's quality measurement and statistical representation. The HCAE model's statistical representation is comprised of 9*9 segments, each including four channels with the segment's probability to contain one of four labels, 1) Under-printed region, 2) Normally printed region, 3) Over-printed region, 4) Empty region. This data structure plays a significant role in determining the commands needed to optimize the fabrication process. The implemented HCAE model's accuracy and repeatability were measured by a multi-label multi-output metric developed in this study. The assessments made by HCAE are then used to perform an in-situ process adjustment by manipulating the future layer's fabrication through the G-code modification. By adjusting the machine's print speed and feedrate, the control algorithm exploits the next layer deposition, segment by segment. The algorithm is then tested with two manually defected settings, one part with severe under extrusion and one with over extrusion parameters. Both test artifacts' quality advanced significantly and converged to an acceptable state by four iterations.

show abstract

Deep transfer learning of additive manufacturing mechanisms across materials in metal-based laser powder bed fusion process

Cited by 66 publications

References 41 publications

Deep Learning-Based Monitoring of Laser Powder Bed Fusion Process on Variable Time-Scales Using Heterogeneous Sensing and Operando X-Ray Radiography Guidance

Deep Learning-Based Monitoring of Laser Powder Bed Fusion Process on Variable Time-Scales Using Heterogeneous Sensing and Operando X-Ray Radiography Guidance

When am (additive manufacturing) meets ae (acoustic emission) and AI (artificial intelligence)

In-Situ Quality Assessment and Control in Additive Manufacturing Processes Using Laser Surface Profilometer and Deep Learning Techniques

Contact Info

Product

Resources

About