Search

Metal-based Laser Powder Bed Fusion (LPBF) suffers from a lack of repeatability and is challenging to model, making their quality monitoring essential and demanding. The reason lies in the high dynamics taking place during the interaction of the laser with metallic powders. To bring this technology to mass production, industries are only interested in the process regime where the built layer's quality meets their standards. All other process regimes leading to poor mechanical properties and/or defect formation such as balling, Lack of Fusion (LoF) pores, keyhole pores, delamination, and crack propagation irrespective of their different regimes are considered anomalies. Today, the common methodology for monitoring uses conventional/supervised Machine Learning (ML) algorithms for the classification task. However, it requires collecting a balanced dataset corresponding to each investigated regime from the sensors, which is very expensive and time-consuming. As an alternative, the article proposes a semi-supervised approach where the defect-free regime can be differentiated from the anomalies by familiarising the ML algorithms only with the distribution of acoustic signatures corresponding to the defect-free regime. This work presents two generative Convolutional Neural Network architectures based on Variational Auto-Encoder and General Adversarial Network. As a result, we could classify the anomaly regimes with 96 and 97% accuracy, respectively. [ABSTRACT FROM AUTHOR]

The defective regimes in metal-based Laser Powder Bed Fusion (LPBF) processes can be minimized by deploying in-situ monitoring strategies comprising Machine learning (ML) algorithms and sensing techniques. So far, algorithms trained for monitoring a particular material type cannot be re-used to monitor another material in Additive Manufacturing (AM). This is a topic rarely researched in AM. Inspired by the idea of transfer learning in ML, we demonstrate the knowledge learned by the two native Deep Learning (DL) networks, namely VGG and ResNets, on four LPBF process mechanisms such as balling, Lack of Fusion (LoF) pores, conduction mode, and keyhole pores in stainless steel (316L) can be transferred to bronze (CuSn8). In this work, the spectrograms computed using Wavelet Transforms (WT) on Acoustic Emissions (AE) during the LBPF process of stainless steel and bronze are used for training the two DL networks. Either network is first trained for classification by spectrograms representing four mechanisms during the processing of stainless steel. The trained model is then re-trained using transfer learning with spectrograms from bronze data for a similar classification task. The accuracy of the two networks during transfer learning shows that it is effectively possible to learn transferable features from one material to another with minimum network training time and dataset collection. [Display omitted] • In situ monitoring for Metal-based LPBF is presented by combining airborne acoustic signatures with machine learning. • Spectrograms computed from acoustic emissions are used to train the VGG-16 and ResNet-18 for classifying mechanisms. • The classification model trained on one material is selectively retrained for classification on another material. • Accuracies based on transfer learning shows that it is possible to transfer features from stainless steel to bronze. • The performance of the two CNN architectures during transfer learning is compared and discussed. [ABSTRACT FROM AUTHOR]

Metal-based Laser Powder Bed Fusion (LPBF) has made fabricating intricate components easier. Yet, assessing part quality is inefficient, relying on costly Computed Tomography (CT) scans or time-consuming destructive tests. Also, intermittent inspection of layers also hampers machine productivity. The Additive Manufacturing (AM) field explores real-time quality monitoring using sensor signatures and Machine Learning (ML) to tackle this. One such approach is sensing airborne Acoustic Emissions (AE) from process zone perturbations and comprehending flaw formation for monitoring the LPBF process. This study emphasizes the importance of selecting airborne AE sensors for accurately classifying LPBF dynamics in 316 L, utilizing a flat response sensor to capture AE's during three regimes: Lack of Fusion, conduction mode, and keyhole. To comprehensively understand AE from a broad process space, the data was collected for two different 316 L stainless steel powder distributions (> 45 µm and < 45 µm) using two different parameter sets. Frequency analysis unveiled distinct LPBF dynamics as dominant and correlated in specific frequency ranges. Empirical Mode Decomposition was used to examine the periodicity of AE signals by separating them into constituent signals for comparison. Transformed AE signals were trained to distinguish regimes using ML classifiers (Convolutional Neural Networks, eXtreme Gradient Boosting, and Support Vector Machines). Sensitivity analysis using saliency maps and feature importance scores identified frequency information below 40 kHz relevant for decision-making. This study highlights interpretable machine learning's potential to identify critical frequency ranges for distinguishing LPBF regimes and underscores the importance of sensor selection for enhanced process monitoring. [Display omitted] [ABSTRACT FROM AUTHOR]

[Display omitted] • This study addresses LPBF monitoring robustness amid diverse data distributions across process parameters. • To label complex datasets into discrete process dynamics, this study suggests an ML strategy using Acoustic Emission from the process zone. • The study suggests a self-supervised Bayesian Neural Network for LPBF process dynamics identification without ground-truth data. • The framework excels in classification, anomaly detection, and transfer learning, even with varying AE data distribution due to LPBF parameters. • The study shows improved ML model generalizability through the self-supervised learning method's accurate predictions in a different environment. This study presents a self-supervised Bayesian Neural Network (BNN) framework using air-borne Acoustic Emission (AE) to identify different Laser Powder Bed Fusion (LPBF) process regimes such as Lack of Fusion , conduction mode , and keyhole without ground-truth information. The proposed framework addresses the challenge of labelling datasets with semantic complexities into discrete process dynamics. This novel AE-based in-situ monitoring approach provides a promising alternative to quantify part density in LPBF process. The study demonstrates the effectiveness of a Bayesian encoder backbone for learning the manifold representations of LPBF regimes, which were visually separable in a lower-dimensional representation using t-distributed stochastic neighbour embedding. The generalized representations learned by the Bayesian backbone allowed traditional classifiers trained on smaller datasets to exhibit high classification accuracy. The feature map computed using pre-trained Bayesian encoder on other datasets was also effective in anomaly detection, achieving 92% accuracy with one-class Support Vector Machine. Additionally, the representation learned by the BNN facilitates transfer learning, where it can be fine-tuned for classification tasks on different process maps, which is also demonstrated in this work. Our proposed framework improves the generalization and robustness of the LPBF monitoring, particularly in the face of varying data distribution across multiple process parameter spaces. [ABSTRACT FROM AUTHOR]