A deep learning framework for defect prediction based on thermographic in-situ monitoring in laser powder bed fusion

Abstract

The prediction of porosity is a crucial task for metal based additive manufacturing techniques such as laser powder bed fusion. Short wave infrared thermography as an in-situ monitoring tool enables the measurement of the surface radiosity during the laser exposure. Based on the thermogram data, the thermal history of the component can be reconstructed which is closely related to the resulting mechanical properties and to the formation of porosity in the part. In this study, we present a novel framework for the local prediction of porosity based on extracted features from thermogram data. The framework consists of a data pre-processing workflow and a supervised deep learning classifier architecture. The data pre-processing workflow generates samples from thermogram feature data by including feature information from multiple subsequent layers. Thereby, the prediction of the occurrence of complex process phenomena such as keyhole pores is enabled. A custom convolutional neural network model is used for classification. The model is trained and tested on a dataset from thermographic in-situ monitoring of the manufacturing of an AISI 316L stainless steel test component. The impact of the pre-processing parameters and the local void distribution on the classification performance is studied in detail. The presented model achieves an accuracy of 0.96 and an f1-Score of 0.86 for predicting keyhole porosity in small sub-volumes with a dimension of (700 × 700 × 50) µm3. Furthermore, we show that pre-processing parameters such as the porosity threshold for sample labeling and the number of included subsequent layers are influential for the model performance. Moreover, the model prediction is shown to be sensitive to local porosity changes although it is trained on binary labeled data that disregards the actual sample porosity.

Mitigation of Gas Porosity in Additive Manufacturing Using Experimental Data Analysis and Mechanistic Modeling

“…[34] Ti6Al4V A deep learning architecture was employed using heat signals to predict and minimize porosity. [35] Ti6Al4V A deep learning technique was used for porosity reduction and monitoring that used Convolutional Neural Networks.…”

“…Several attempts have been made to mitigate gas porosity in LPBF parts (Table 1) using experimental techniques [18][19][20][21][22][23][24][25], mechanistic modeling [26][27][28][29][30][31][32][33][34], and machine learning [35][36][37][38][39][40][41]. However, experimental trial-and-error to adjust many process variables for reducing gas porosity is expensive and time-consuming.…”

“…However, these models are computationally intensive and often difficult to use in real time. Machine learning models [35][36][37][38][39][40][41] can be used in real-time; however, they are often unable to capture the important physical factors causing porosity and need a large volume of high-quality data for a reliable prediction. Thus, the existing approaches (Table 1) using experimental, modeling, and machine learning methods are inadequate to reduce gas porosity in LPBF.…”

Complementary Methods for the Assessment of the Porosity of Laser Additive-Manufactured Titanium Alloy

“…The tests were accomplished according to ASTM 1476-04, using a SPECTRO xSORT spectrometer. The prediction of porosity, which strongly affects the mechanical performance of a product, becomes of major importance [30], and it depends on the pre-processing parameters [31], the prediction of the model being sensitive to local changes even under the conditions of training on labeled data. The SLM samples were obtained using the Arcam Q20 + system Concept Laser M2 PBF machine (General Electric, Gothenburg, Sweden), and were then processed directly, without introducing additional stress.…”