A Convolutional Neural Network(CNN) Classification To Identify The Presence of Pores in Powder Bed Fusion Images

Abstract This study focuses on the detection of seeded porosity during metal additive manufacturing by employing convolutional neural networks (CNN). The aim of the study is to demonstrate the application of Machine Learning (ML) in in-process monitoring. Laser Powder Bed Fusion (LPBF) is a selective laser melting technique used to build complex 3D parts. The current monitoring system in LPBF is inadequate to produce safety-critical parts due to the lack of automated processing of collected data. To assess the efficacy of applying ML to defect detection in LPBF by in-process images, a range of synthetic defects have been designed into cylindrical artefacts to mimic porosity occurring in different locations, shapes, and sizes. Empirical analysis has revealed insights into the importance of accurate labelling strategies required for data-driven solutions. Two labelling strategies based on the computer aided design (CAD) file and X-ray computed tomography (XCT) scan data was formulated. A novel CNN was trained from scratch and optimised by selecting the best values of an extensive range of hyper-parameters by employing Hyperband tuner. The accuracy of the model was 90% when trained using a CAD-assisted labelling, and 97% when using XCT-assisted labelling. The model successfully spotted pores as small as 0.2mm. Experiments revealed that balancing the data set improved the model's precision from 89% to 97% and recall from 85% to 97% when compared to training on an imbalanced data set. We strongly believed that the proposed model would significantly reduce post-processing cost and provide a better base model network for transfer learning of future ML models aimed at LPBF micro-defects detection.

Download Full-text

A study on surface morphology and tension in laser powder bed fusion of Ti-6Al-4V

The International Journal of Advanced Manufacturing Technology ◽

10.1007/s00170-020-06221-w ◽

2020 ◽

Vol 111 (9-10) ◽

pp. 2891-2909

Author(s):

Mahyar Khorasani ◽

AmirHossein Ghasemi ◽

Umar Shafique Awan ◽

Elahe Hadavi ◽

Martin Leary ◽

...

Keyword(s):

Surface Tension ◽

Process Parameters ◽

Surface Profile ◽

Test Case ◽

Powder Bed Fusion ◽

Laser Powder Bed Fusion ◽

Data Set ◽

Powder Bed ◽

Case Morphology ◽

Melting Pool

Abstract When reporting surface quality, the roughest surface is a reference for the measurements. In LPBF due to recoil pressure and scan movement, asymmetric surface is shaped, and surface roughness has different values in different measurement orientations. In this research, the influence of the laser powder bed fusion (LPBF) process parameters on surface tension and roughness of Ti-6AI-4 V parts in three orientations are investigated. To improve the mechanical properties, heat treatment was carried out and added to the designed matrix to generate a comprehensive data set. Taguchi design of experiment was employed to print 25 samples with five process parameters and post-processing. The effect and interaction of the parameters on the formation of surface profile comprising tension, morphology and roughness in various directions have been analysed. The main contribution of this paper is developing a model to approximate the melting pool temperature and surface tension based on the process parameters. Other contributions are an analysis of process parameters to determine the formation and variation of surface tension and roughness and explain the governing mechanisms through rheological phenomena. Results showed that the main driving factors in the variation of surface tension and formation of the surface profile are thermophysical properties of the feedstock, rheology and the temperature of the melting pool. Also, the results showed that while the value of surface tension is the same for each test case, morphology and the value of roughness are different when analysing the surface in perpendicular, parallel and angled directions to laser movement.

Download Full-text

Powder Bed Fusion of Biomedical Co-Cr-Mo and Ti-6Al-4V Alloys: Microstructure and Mechanical Properties

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.1151.3 ◽

2019 ◽

Vol 1151 ◽

pp. 3-7 ◽

Cited By ~ 1

Author(s):

Eleonora Santecchia ◽

Paolo Mengucci ◽

Andrea Gatto ◽

Elena Bassoli ◽

Lucia Denti ◽

...

Keyword(s):

Computer Aided Design ◽

Tensile Tests ◽

Layer By Layer ◽

Powder Bed Fusion ◽

X Ray Diffraction ◽

Powder Bed ◽

Mechanical Parts ◽

Aided Design ◽

Microscopy Techniques ◽

High Degree

Powder bed fusion (PBF) is an additive manufacturing technique, which allows to build complex functional mechanical parts layer-by-layer, starting from a computer-aided design (CAD) model. PBF is particularly attractive for biomedical applications, where a high degree of individualization is required. In this work, the microstructure of two biomedical alloys, namely Co-Cr-Mo and Ti-6Al-4V, were studied by X-ray diffraction and electron microscopy techniques. Hardness and tensile tests were performed on the sintered parts.

Download Full-text

Measured Data Alignments for Monitoring Metal Additive Manufacturing Processes Using Laser Powder Bed Fusion Methods

Volume 9: 40th Computers and Information in Engineering Conference (CIE) ◽

10.1115/detc2020-22478 ◽

2020 ◽

Author(s):

Shaw C. Feng ◽

Yan Lu ◽

Albert T. Jones

Keyword(s):

Coordinate System ◽

Computer Aided Design ◽

Three Dimensional ◽

Ex Situ ◽

Powder Bed Fusion ◽

Powder Bed ◽

Alignment Procedure ◽

3D Cad ◽

Data Alignment ◽

Cad Models

Abstract The number and types of measurement devices used for monitoring and controlling Laser-Based Powder Bed Fusion of Metals (PBF-LB/M) processes and inspecting the resulting AM metal parts have increased rapidly in recent years. The variety of the data collected by such devices has increased, and the veracity of the data has decreased simultaneously. Each measurement device generates data in a unique coordinate system and in a unique data type. Data alignment, however, is required before 1) monitoring and controlling PBF-LB/M processes, 2) predicting the material properties of the final part, and 3) qualifying the resulting AM parts can be done. Aligned means all data must be transformed into a single coordinate system. In this paper, we describe a new, general data-alignment procedure and an example based on PBF-LB/M processes. The specific data objects used in this example include in-situ photogrammetry, thermography, ex-situ X-ray computed tomography (XCT), coordinate metrology, and computer-aided design (CAD) models. We propose a data-alignment procedure to align the data from melt pool images, scan paths, layer images, XCT three-dimensional (3D) model, coordinate measurements, and the 3D CAD model.

Download Full-text

Laser Powder Bed Fusion of Potential Superalloys: A Review

Metals ◽

10.3390/met11010058 ◽

2020 ◽

Vol 11 (1) ◽

pp. 58

Author(s):

Prince Valentine Cobbinah ◽

Rivel Armil Nzeukou ◽

Omoyemi Temitope Onawale ◽

Wallace Rwisayi Matizamhuka

Keyword(s):

Computer Aided Design ◽

Value Added ◽

Powder Bed Fusion ◽

Laser Powder Bed Fusion ◽

Powder Bed ◽

High Entropy ◽

Controlled Processing ◽

Structural Applications ◽

Aided Design ◽

Sintering Processes

The laser powder bed fusion (LPBF) is an additive manufacturing technology involving a gradual build-on of layers to form a complete component according to a computer-aided design. The LPBF process boasts of manufacturing value-added parts with higher accuracy and complex geometries for the transport, aviation, energy, and biomedical industries. TiAl-based alloys and high-entropy alloys (HEAs) are two materials envisaged as potential replacements of nickel-based superalloys for high temperature structural applications. The success of these materials hinge on optimization and implementation of tailored microstructures through controlled processing and appropriate alloy manipulations that can promote and stabilize new microstructures. Therefore, it is important to understand the LPBF technique, and its associated microstructure-mechanical property relationships. This paper discusses the metallurgical sintering processes of LPBF, the effects of process parameters on densification, microstructures, and mechanical properties of LPBFed TiAl-based alloys and HEAs. This paper also, presents updates and future studies recommendations on the LPBFed TiAl-based alloys and HEAs.

Download Full-text

Toward Sub-Surface Pore Prediction Capabilities for Laser Powder Bed Fusion Using Data Science

Journal of Manufacturing Science and Engineering ◽

10.1115/1.4050461 ◽

2021 ◽

Vol 143 (7) ◽

Author(s):

Deniz Sera Ertay ◽

Shima Kamyab ◽

Mihaela Vlasea ◽

Zohreh Azimifar ◽

Thanh Ma ◽

...

Keyword(s):

Process Parameters ◽

Data Science ◽

Pore Space ◽

Synthetic Data ◽

Superior Performance ◽

Powder Bed Fusion ◽

Laser Powder Bed Fusion ◽

Data Set ◽

Powder Bed ◽

Scan Path

Abstract Achieving defect-free parts is traditionally challenging in laser powder bed fusion (LPBF). The mechanical properties of additively manufactured parts are highly affected by their density; as such, research in defect detection and pore prediction has gained significant interest. The process parameters, the powder characteristics, and the process environment conditions play an important role in defect occurrence. Moreover, the laser scan path affects density, especially at scan path discontinuities. In this work, the complex interaction between the process parameters and the scan path on the occurrence of subsurface pores is investigated. In the data preparation step, a synthetic data set is generated to model the melt pool morphology along the scan path. A secondary data set containing the pore space of the resulting parts is obtained via X-ray computed tomography (CT) and is registered with the synthetic data set. Machine learning models, namely, a Conditional Variational AutoEncoder (CVAE) and a Convolutional Neural Network (CNN), are then trained based on the input features to predict pore occurrence. The performance evaluation of both CNN and CVAE models on synthetic data indicates that the scan path and process parameters can be utilized in predicting pore locations. Quantitative results show that employing offline CT images a priori in training the CVAE, without the need to have CT information in the test phase, leads the CVAE model to superior performance over the CNN.

Download Full-text

Analysis of Spatter Removal by Sieving during a Powder-Bed Fusion Manufacturing Campaign in Grade 23 Titanium Alloy

Metals ◽

10.3390/met11030399 ◽

2021 ◽

Vol 11 (3) ◽

pp. 399

Author(s):

Ryan Harkin ◽

Hao Wu ◽

Sagar Nikam ◽

Justin Quinn ◽

Shaun McFadden

Keyword(s):

Computer Aided Design ◽

Compositional Analysis ◽

Mesh Size ◽

Particle Morphology ◽

Particle Size Analysis ◽

Size Analysis ◽

Powder Bed Fusion ◽

Powder Bed ◽

Powder Particles ◽

Sieving Process

The Laser-based Powder Bed Fusion (L-PBF) process uses a laser beam to selectively melt powder particles deposited in a layer-wise fashion to manufacture components derived from Computer-Aided Design (CAD) information. During laser processing, material is ejected from the melt pool and is known as spatter. Spatter particles can have undesirable geometries for the L-PBF process, thereby compromising the quality of the powder for further reuse. An integral step in any powder replenishing and reuse procedure is the sieving process. The sieving process captures spatter particles within the exposed powder that have a diameter larger than a defined mesh size. This manuscript reports on Ti6Al4V (Grade 23) alloy powder that had been subjected to seven reuse iterations, focusing on the characterisation of powder particles that had been captured (i.e., removed) by the sieving processes. Characterisation included chemical composition focusing upon interstitial elements O, N and H (wt.%), particle morphology and particle size analysis. On review of the compositional analysis, the oxygen contents were 0.43 wt.% and 0.40 wt.% within the 63 µm and 50 µm sieve-captured powder, respectively. Additionally, it was found that a minimum of 79% and 63% of spatter particles were present within the captured powder removed by the 63 µm and 50 µm sieves, respectively.

Download Full-text

Automated Finite Element Model Mesh Updating Scheme Applicable to Mistuning Analysis

Volume 7B: Structures and Dynamics ◽

10.1115/gt2014-26925 ◽

2014 ◽

Cited By ~ 12

Author(s):

Alexander A. Kaszynski ◽

Joseph A. Beck ◽

Jeffrey M. Brown

Keyword(s):

Finite Element ◽

Computer Aided Design ◽

User Involvement ◽

Element Model ◽

Cad Model ◽

Data Set ◽

Optical Scanner ◽

Wide Range ◽

Cad Models ◽

Scan Data

Advancement of optical geometric measurement hardware has enabled the construction of accurate 3D tessellated models for a wide range of turbomachinery components. These tessellated models can be reverse-engineered into computer-aided design (CAD) models and input into grid generation software for finite element analyses. However, generating a CAD model from scan data is a time consuming and cumbersome process requiring significant user-involvement for even a single model. While it is possible to generate finite element models (FEMs) directly from tessellated data, current direct-grid methods produce unstructured grids that can introduce fictitious, numerical mistuning in these models, obscuring geometric mistuning. Nonetheless, as-measured scan data captured in a structured grid is essential for accurate geometric mistuning analyses, provided the tessellated scan data can be rapidly and accurately transformed into a FEM. This paper outlines and demonstrates an approach for rapidly generating structured FEMs for a population of integrally bladed rotors (IBRs) without requiring the arduous task of generating a CAD model for each as-measured IBR. This is accomplished by morphing the structured mesh of a nominal model to the tessellated data set collected from an optical scanner. It is shown that the fidelity and structure of these FEMs can be utilized for accurate mistuning analyses.

Download Full-text