Neural network based image segmentation for spatter extraction during laser-based powder bed fusion processing

Tan, Zhenbiao; Fang, Qihang; Li, Hui; Liu, Sheng; Zhu, Wenkang; Ding, Yang

doi:10.1016/j.optlastec.2020.106347

Cited by 47 publications

(17 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the little difference in brightness between the powder spatter and power bed within the view field of the high-speed camera, only the droplet spatters were detected, but not the nonmolten powder particles. Tan et al [40] used a computational technique to analyze the obtained images, segmenting each block to extract the spatter. In the same year, Yin et al [41] introduced an external light source (a CAVILUX ® pulsed high-power diode laser light source) and a high-speed camera (Phantom V2012) to detect the spatter and obtain clearer images.…”

Section: Visible-light High-speed Detectormentioning

confidence: 99%

“…Algorithms for 2D image processing are less complex than those for 3D image processing. Tan et al [40] captured spatter images using Kalman filter tracking, segmented the images with grayscale and edge information, and obtained spatter information using fully convolutional networks and Mask R-CNN. Yin et al [61] projected the 3D spatter trajectory into a 2D plane with image processing, used a filtering technique to improve the sharpness of the spatter image, and tracked the spatter motion information frame by frame using ImageJ.…”

Section: Spatter 2d Image Processing Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

A Review of Spatter in Laser Powder Bed Fusion Additive Manufacturing: In Situ Detection, Generation, Effects, and Countermeasures

Yin

et al. 2022

Micromachines

View full text Add to dashboard Cite

Spatter is an inherent, unpreventable, and undesired phenomenon in laser powder bed fusion (L-PBF) additive manufacturing. Spatter behavior has an intrinsic correlation with the forming quality in L-PBF because it leads to metallurgical defects and the degradation of mechanical properties. This impact becomes more severe in the fabrication of large-sized parts during the multi-laser L-PBF process. Therefore, investigations of spatter generation and countermeasures have become more urgent. Although much research has provided insights into the melt pool, microstructure, and mechanical property, reviews of spatter in L-PBF are still limited. This work reviews the literature on the in situ detection, generation, effects, and countermeasures of spatter in L-PBF. It is expected to pave the way towards a novel generation of highly efficient and intelligent L-PBF systems.

show abstract

Section: Visible-light High-speed Detectormentioning

confidence: 99%

Section: Spatter 2d Image Processing Algorithmmentioning

confidence: 99%

A Review of Spatter in Laser Powder Bed Fusion Additive Manufacturing: In Situ Detection, Generation, Effects, and Countermeasures

Yin

et al. 2022

Micromachines

View full text Add to dashboard Cite

show abstract

“…Properly trained ANNs can model the correlations between the given input and output data and accordingly predict the responses based on unseen input values. ANNs have been employed in the SLM process in various ways; some of the applications are the design for AM, including topology and material design [52], in-situ process monitoring, including melt pool or powder bed monitoring using optical or acoustic techniques [53][54][55][56], and process-property correlation [57,58]. Recently, the process-property correlation application of the ANN models has been extended to the optimization of this process for different materials.…”

Section: Introductionmentioning

confidence: 99%

Experimental, Computational, and Data-Driven Study of the Effects of Selective Laser Melting (SLM) Process Parameters on Single-Layer Raster Scanning Surface Characteristics

Fotovvati

Rauniyar²,

Arnold³

et al. 2022

Preprint

View full text Add to dashboard Cite

In this study, the influence of selective laser melting (SLM) parameters on the surface characteristics of “single-layer” raster scanning deposits, with Ti6Al4V powder, is thoroughly investigated through experimental, computational, and data-driven approaches. In layer-wise manufacturing processes, such as SLM, the evaluation and understanding of the surface characteristics of individual layers of materials are important; however, there is limited work in the literature on the “single-layer” surface characteristics in SLM. Single-layer samples are designed on a smooth surface on top of semi-cylinders and fabricated using different levels of SLM processing parameters, i.e., laser power, scan speed, and hatch spacing, in different numbers of scan lines, i.e., 5, 10, 20, and 40. The fabricated sample surface data are acquired using non-contact optical interferometry. The computational approach utilizes the discrete element method (DEM) for the powder bed distribution, as well as computational fluid dynamics (CFD) for simulating the laser-powder interaction. One of the main contributions of this study is to investigate and provide predictive models for the correlations between the surface roughness of raster scans and SLM process parameters through experiments and computational work. In addition, a method for comparing the surface characteristics of the thermo-fluid simulation and experimental results is presented. A technique is employed to extract the surface roughness from the simulated results. The discussion on the experiment-simulation comparison challenges provides a deep insight into the criteria selection for experiment-simulation comparison in the SLM process. Moreover, an artificial neural network was trained using the back-propagation method and K-fold cross-validation, as well as regularization techniques to avoid overfitting. The laser power proved to be the most significant parameter, having a contribution of 48.3%, in determining the surface roughness with a unique trend affected by the spattering of powder particles. The simulation results showed noticeable deviations from the experiments in cases with extreme parameters, e.g., too high or too low values of power, speed, or hatch spacing. The machine learning algorithm achieves reasonable predictability of the single-layer surface roughness, showing a root mean square error and a coefficient of determination of 0.9589 µm and 98.78%, respectively, for a separate testing dataset.

show abstract

“…Although it is not possible to completely eliminate spatter, it is necessary to regulate the behavior of spatter. Currently, researchers have conducted some experimental studies on SLM spatter, and the main idea is to record the generation and motion of spatter with the help of real-time monitoring means such as high-speed cameras [7][8][9][10], X-ray imaging [11], and acoustic signal acquisition [12] and to study the effects of process parameters such as scanning speed [13], laser power [13], laser beam number [14], and the protective gas type and flow rate [15,16] on spatter behavior, and then to propose SLM spatter regulation strategies such as adjusting the forming cavity flow field [17] and adding nanoparticles [18]. Experimental studies play an important role in the in-depth understanding of SLM spatter, but because SLM spatter is in a dramatically changing high-temperature and high-speed environment, the cost of experimental studies is high, and it is difficult to achieve quantitative analysis, so numerical simulation becomes a necessary aid to research.…”

Section: Introductionmentioning

confidence: 99%

CFD–DPM Simulation Study of the Effect of Powder Layer Thickness on the SLM Spatter Behavior

Cao

Zhang

Meng

2022

Metals

View full text Add to dashboard Cite

Selective Laser Melting (SLM) has significant advantages in manufacturing complex structural components and refining the alloy microstructure; however, spatter, as a phenomenon that accompanies the entire SLM forming process, is prone to problems such as inclusions, porosity, and low powder recovery quality. In this paper, a Computational Fluid Dynamics–Discrete Particle Method (CFD–DPM) simulation flow for predicting the SLM spatter behavior is established based on the open-source code OpenFOAM. Among them, the single-phase flow Navier–Stokes equation is used in the Eulerian framework to equivalently describe the effect of metal vapor and protective gas on the flow field of the forming cavity, and the DPM method is used in the Lagrangian framework to describe the metal particle motion, and the factors affecting the particle motion include particle–particle collision, particle–wall collision, fluid drag force, gravity, buoyancy force, and additional mass force. In addition, the equivalent volume force and fluid drag force are used to characterize the fluid–particle coupling interaction. For the spatter behavior and powder bed denudation phenomenon, the calculation results show that the spatter height and the drop location show a clear correlation, and the powder bed denudation phenomenon is caused by the high-speed gas flow, causing the surrounding gas to gather in the forming area, which in turn drives the motion of the powder bed particles. For the effect of powder layer thickness on spatter and powder bed denudation, the calculation results show that the effect of powder layer thickness on the number of spatters is large (when the thickness was increased from 50 μm to 100 μm, the number of spatters increased by 157%), but the effect on spatter height and drop location distribution is small. When the powder layer thickness is small, the width of the denudation zone is significantly larger, but when the powder layer reaches a certain thickness, the width of the denudation zone does not show significant changes. It should be noted that the presented model has not been directly validated by experiments so far due to the difficulty of tracking the large-scale motion of SLM spatter in real time by current experimental means.

show abstract

Neural network based image segmentation for spatter extraction during laser-based powder bed fusion processing

Cited by 47 publications

References 29 publications

A Review of Spatter in Laser Powder Bed Fusion Additive Manufacturing: In Situ Detection, Generation, Effects, and Countermeasures

A Review of Spatter in Laser Powder Bed Fusion Additive Manufacturing: In Situ Detection, Generation, Effects, and Countermeasures

Experimental, Computational, and Data-Driven Study of the Effects of Selective Laser Melting (SLM) Process Parameters on Single-Layer Raster Scanning Surface Characteristics

CFD–DPM Simulation Study of the Effect of Powder Layer Thickness on the SLM Spatter Behavior

Contact Info

Product

Resources

About