AI-based defect detection and self-healing in metal additive manufacturing

2021

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...

All Academics Research

Defect detection in Additive Manufacturing (AM) is a critical aspect of ensuring product quality, particularly in industries such as renewable energy and biomedical engineering, where reliability and precision are paramount. This study conducted a systematic review of 152 peerreviewed articles, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, to analyze the adoption of Artificial Intelligence (AI) techniques in defect detection within AM processes. The review revealed that machine learning (ML) and deep learning (DL) techniques, such as Convolutional Neural Networks (CNNs) and Support Vector Machines (SVMs), are widely employed for identifying common defects like porosity, delamination, and dimensional inaccuracies. Hybrid AI models, integrating ML and DL, demonstrated superior performance in detecting complex, multi-dimensional defects across various AM applications. Additionally, the integration of multimodal data, including thermal imaging, acoustic signals, and optical measurements, was found to improve defect detection rates by an average of 22%, enhancing the robustness and accuracy of AI models. The study also identified significant challenges, including dataset scarcity and annotation inconsistencies, which limit the generalizability and scalability of AI solutions. Comparative analyses further highlighted the distinct advantages of tailored AI approaches for specific applications, with renewable energy and biomedical engineering being key focus areas. This review underscores the transformative potential of AI in advancing defect detection in AM, providing a comprehensive understanding of its capabilities, challenges, and implications for highstakes manufacturing industries.

Metals

Laser Melting Deposition (LMD) is a metal printing technique that allows for the manufacturing of large objects by Directed Energy Deposition. Due to its versatility in variation of parameters, the possibility to use two or more materials, to create alloys in situ or produce multi-layer structures, LMD is still being scientifically researched and is still far from industrial maturity. The structural testing of obtained samples can be time consuming and solutions that can decrease the samples analysis time are constantly proposed in the scientific literature. In this manuscript we present a quality improvement study for obtaining defect-free bulk samples of Ti6Al4V under X-Ray Computed Tomography (XCT) by varying the hatch spacing and distance between planes. Based on information provided by XCT, the experimental conditions were changed until complete elimination of porosity. Information on the defects in the bulk of the samples by XCT was used for feedback during parameters tuning i...

Materials

Surface deformation is a multi-factor, laser powder-bed fusion (LPBF) defect that cannot be avoided entirely using current monitoring systems. Distortion and warping, if left unchecked, can compromise the mechanical and physical properties resulting in a build with an undesired geometry. Increasing dwell time, pre-heating the substrate, and selecting appropriate values for the printing parameters are common ways to combat surface deformation. However, the absence of real-time detection and correction of surface deformation is a crucial LPBF problem. In this work, we propose a novel approach to identifying surface deformation problems from powder-bed images in real time by employing a convolutional neural network-based solution. Identifying surface deformation from powder-bed images is a significant step toward real-time monitoring of LPBF. Thirteen bars, with overhangs, were printed to simulate surface deformation defects naturally. The carefully chosen geometric design overcomes pr...

Technologies, 2019

In this study, micro-computed tomography (CT) is utilized to detect defects of Ti-6Al-4V specimens fabricated by selective laser melting (SLM) and electron beam melting (EBM), which are two popular metal additive manufacturing methods. SLM and EBM specimens were fabricated with random defects at a specific porosity. The capability of micro-CT to evaluate inclusion defects in the SLM and EBM specimens is discussed. The porosity of EBM specimens was analyzed through image processing of CT single slices. An empirical method is also proposed to estimate the porosity of reconstructed models of the CT scan.

e-Journal of Nondestructive Testing, 2022

Despite the capability of fabricating complex and customized components, metal laser powder bed fusion (LPBF) is still affected by manufacturing issues, which can lead to significant geometrical and dimensional errors and internal defects. These aspects can represent a major barrier to a wider industrial application of LPBF, particularly if considering that relevant applications of additive manufacturing are in sectors such as biomedical and aerospace, which have stringent requirements in terms of defects and product quality. The requests for precision improvement are orienting research activities towards the development of inprocess monitoring systems able to perform accurate analyses during the fabrication itself, hence providing useful information for improving the quality of produced parts. To this aim, several in-process monitoring methods have been proposed in the literature to identify and correct out-of-control process conditions. In spite of the aforementioned research efforts, work is still needed to reliably correlate in-process measurements to actual defects. The focus of this experimental study is the definition of a robust methodology to compare in-process optical acquisitions to post-process X-ray computed tomography (XCT) measurements of actual defects. XCT unique capabilities are therefore exploited to support and improve the LPBF process through the implementation of an accurate comparison methodology.

Journal of Intelligent Manufacturing

Part quality manufactured by the laser powder bed fusion process is significantly affected by porosity. Existing works of process–property relationships for porosity prediction require many experiments or computationally expensive simulations without considering environmental variations. While efforts that adopt real-time monitoring sensors can only detect porosity after its occurrence rather than predicting it ahead of time. In this study, a novel porosity detection-prediction framework is proposed based on deep learning that predicts porosity in the next layer based on thermal signatures of the previous layers. The proposed framework is validated in terms of its ability to accurately predict lack of fusion porosity using computerized tomography (CT) scans, which achieves a F1-score of 0.75. The framework presented in this work can be effectively applied to quality control in additive manufacturing. As a function of the predicted porosity positions, laser process parameters in the ...

DergiPark (Istanbul University), 2024

Additive manufacturing technologies present a wide array of benefits, including the capacity to manufacture components with complex geometric forms, reduced production expenses, minimized material usage, and time efficiency. This research constitutes a significant effort to pinpoint geometric defects and dimensional irregularities as well as surface quality imperfections in the Fused Deposition Modeling process through the development of a deep learning model utilizing multi-scale convolutional neural networks. The proposed methodology encompasses three distinct scales, each capable of identifying defects of varying dimensions. The model underwent extensive hybridizing procedures for precisely training through diverse datasets, and the training process is repeated numerous times until the desired level of accuracy was attained. A sufficiently extensive image datasets are employed to train the models, leading to the precise calibration of the network. As a result, the necessity for prolonged time and intricate computations to identify large-scale defects is eliminated. The highest validation accuracy for defect detection in this study reached 94%.

Materials & Design, 2021

h i g h l i g h t s A novel digital twin method for real-time flaw detection in laser powder bed fusion. Approach combines in-situ meltpool temperature measurements with computational predictions. Three types of flaws considered from processing, machine, and cyber intrusions. Digital twin approach detects all the three types of flaws. Characterization of flaws and microstructure analysis confirms the proposed approach.

Production Engineering, 2022

Despite the potential of additive manufacturing and specifically of selective laser melting, several considerable barriers exist to widespread utilization, especially in specific industries that produce high-value components. Quality control and mechanical characterization remain the most expensive challenge. The quality and mechanical properties of the manufactured parts are influenced by potential defects; the characteristics of these defects, such as size, shape, location, and distribution, have shown to play key roles in mechanical properties. This work proposes a methodology for providing the identification of powder bed anomalies and consequent part defects through a synchronized analysis of the powder layers via digital image processing. This method can be used to study the critical defects formation during the layerwise process, providing important information about their location without the use of expensive or destructive measurements.

Scientific Reports, 2021

In this study, the effects of surface roughness and pore characteristics on fatigue lives of laser powder bed fusion (LPBF) Ti–6Al–4V parts were investigated. The 197 fatigue bars were printed using the same laser power but with varied scanning speeds. These actions led to variations in the geometries of microscale pores, and such variations were characterized using micro-computed tomography. To generate differences in surface roughness in fatigue bars, half of the samples were grit-blasted and the other half were machined. Fatigue behaviors were analyzed with respect to surface roughness and statistics of the pores. For the grit-blasted samples, the contour laser scan in the LPBF strategy led to a pore-depletion zone isolating surface and internal pores with different features. For the machined samples, where surface pores resemble internal pores, the fatigue life was highly correlated with the average pore size and projected pore area in the plane perpendicular to the stress direc...

Additive Manufacturing, 2019

Risk-averse areas such as the medical, aerospace and energy sectors have been somewhat slow towards accepting and applying Additive Manufacturing (AM) in many of their value chains. This is partly because there are still significant uncertainties concerning the quality of AM builds. This paper introduces a machine learning algorithm for the automatic detection of faults in AM products. The approach is semi-supervised in that, during training, it is able to use data from both builds where the resulting components were certified and builds where the quality of the resulting components is unknown. This makes the approach cost efficient, particularly in scenarios where part certification is costly and time consuming. The study specifically analyses Laser Powder-Bed Fusion (L-PBF) builds. Key features are extracted from large sets of photodiode data, obtained during the building of 49 tensile test bars. Ultimate tensile strength (UTS) tests were then used to categorise each bar as 'faulty' or 'acceptable'. Using a variety of approaches (Receiver Operating Characteristic (ROC) curves and 2-fold cross-validation), it is shown that, despite utilising a fraction of the available certification data, the semi-supervised approach can achieve results comparable to a benchmark case where all data points are labelled. The results show that semi-supervised learning is a promising approach for the

Journal of Advanced Manufacturing and Processing, 2019

Metal additive manufacturing (AM) is an innovative manufacturing technique that can build complex and high value metal parts layer by layer using a computerized three-dimensional solid model. Powder bed fusion (PBF) is one of the most common AM techniques. It sequentially processes a powdered feedstock in thin layers and solidifies it by either a laser beam or an electron beam. However, PBF induces microstructural defects that can adversely affect the performance of the manufactured components. These undesirable defects including voids and porosities must be avoided or minimized to limit their negative effects on mechanical properties and to ensure the consistency and repeatability of AM parts. The present article provides an overview of the formation mechanisms of pores in AM metals and some emerging techniques for the detection and quantification of pores. The review also highlights other common defects in PBF parts such as microstructural features associated with keyhole mode of melting. K E Y W O R D S additive manufacturing, defect, metal, porosity, powder bed fusion

Risk-averse areas such as the medical, aerospace and energy sectors have been somewhat slow towards accepting and applying Additive Manufacturing (AM) in many of their value chains. This is partly because there are still signicant uncertainties concerning the quality of AM builds. This paper introduces a machine learning algorithm for the automatic detection of faults in AM products. The approach is semi-supervised in that, during training, it is able to use data from both builds where the resulting components were certied and builds where the quality of the resulting components is unknown. This makes the approach cost ecient, particularly in scenarios where part certication is costly and time consuming. The study specically analyses Selective Laser Melting (SLM) builds. Key features are extracted from large sets of photodiode data, obtained during the building of 49 tensile test bars. Ultimate tensile strength (UTS) tests were then used to categorise each bar as…

2021

This paper presents an investigation of the rapid variations in the temperature of metal melt pool for Additive Manufacturing (AM) processes. The melt pool is created by scanning a high-power laser beam across a metal powder bed. Rapid heating and cooling processes are involved in the layer-by-layer fabrication of the metal part. Recent advances in Machine Learning and Deep Learning algorithms provide efficient ways to analyze large sets of data in search of correlations that would otherwise be extremely time-consuming. The use of Machine Learning and Deep Learning algorithms to understand temperature variations in AM fabrication process will allow to predict the formation of porosity before it occurs. The objective of this research is to advance the AM technology using enhanced Deep Learning techniques to provide in-situ analysis of the melt pool temperature that can lead to a reliable manufacturing of Three-Dimensional (3D) metal parts/components. In specific, Deep Learning based ...