Abstract:Laser powder bed fusion (LPBF) process can incur defects due to melt pool instabilities, spattering, temperature increase, and powder spread anomalies. Identifying defects through in-situ monitoring typically requires collecting, storing, and analyzing large amounts of data generated. The first goal of this work is to propose a new approach to accurately map in-situ data to a three-dimensional (3D) geometry, aiming to reduce the amount of storage. The second goal of this work is to introduce several new IR features for defect detection or process model calibration, which include laser scan order, local preheat temperature, maximum pre-laser scanning temperature, and number of spatters generated locally and their landing locations. For completeness, processing of other common IR features, such as interpass temperature, heat intensity, cooling rates, and melt pool area, are also presented with the underlying algorithm and Python implementation. A number of different parts are printed, monitored, and characterized to provide evidence of process defects and anomalies that different IR features are capable of detecting.

What problem does this paper attempt to address?

This paper attempts to solve the problem of defect detection in the Laser Powder Bed Fusion (LPBF) process. Specifically, the objectives of the paper include: 1. **Propose a new method to accurately map in - situ data to a three - dimensional geometric model**: - Through this method, storage requirements can be reduced, while paving the way for real - time defect detection. Existing methods usually require manual processing, involving threshold setting and generating point clouds to align with 3D geometric models. These methods have limitations when dealing with complex geometries, especially when different features or components may exhibit significant temperature changes. The method proposed in this paper achieves automation and a significant reduction in storage space by correcting image distortion and placing each pixel in the correct position on the 3D geometric model. 2. **Introduce new infrared features for defect detection or process model calibration**: - These new features include laser scanning order, local preheating temperature, maximum pre - laser scanning temperature, the number of locally generated spatters and their landing positions. These features are helpful for identifying lack - of - fusion defects, which are usually caused by large - particle spatters. In addition, the paper also discusses the processing methods of other common infrared features, such as inter - layer temperature, thermal intensity, cooling rate and molten pool area, and provides corresponding algorithms and Python implementations. ### Specific problem - solving methods - **Data mapping**: - The error problems caused by camera angles and rotations are solved by using the `getPerspectiveTransformation()` function in OpenCV for perspective correction. This method can accurately map each pixel in the image to the 3D geometric model, thereby reducing storage requirements and improving accuracy. - **Infrared feature extraction**: - Multiple new infrared features are proposed, including laser scanning order, local preheating temperature, maximum pre - laser scanning temperature, the number of locally generated spatters and their landing positions. The extraction and processing methods of these features are described in detail in the paper, aiming to improve the accuracy and reliability of defect detection. ### Experimental verification - **Experimental setup**: - Experiments are carried out using a FLIR A700 infrared camera and an EOS M290 DMLS system. The camera is fixed at a certain distance to record the infrared images of each layer. Temperature calibration is carried out through a customized heating module and thermocouples to ensure the measurement accuracy in different temperature ranges. - **Feature verification**: - The effectiveness of the proposed feature extraction and data mapping methods is verified by printing test pieces with complex geometric structures (such as Qualification Test Artifact, QTA). The results show that even in very complex geometric structures, this method can accurately identify and map infrared features. ### Conclusion - This paper proposes a new method. Through accurate data mapping and new infrared feature extraction techniques, the accuracy and efficiency of defect detection in the LPBF process are significantly improved. These methods not only reduce storage requirements but also provide technical support for real - time defect detection.

In-Situ Infrared Camera Monitoring for Defect and Anomaly Detection in Laser Powder Bed Fusion: Calibration, Data Mapping, and Feature Extraction

Multiple Sensor Detection of Process Phenomena in Laser Powder Bed Fusion

In-situ visual monitoring for multi-scale defects detection in laser powder bed fusion

Using machine learning to identify in-situ melt pool signatures indicative of flaw formation in a laser powder bed fusion additive manufacturing process

Thermal Based Process Monitoring for Laser Powder Bed Fusion (LPBF)

Multi Phenomena Melt Pool Sensor Data Fusion for Enhanced Process Monitoring of Laser Powder Bed Fusion Additive Manufacturing

Predicting Laser Powder Bed Fusion Defects Through In-Process Monitoring Data and Machine Learning

Inference of highly time-resolved melt pool visual characteristics and spatially-dependent lack-of-fusion defects in laser powder bed fusion using acoustic and thermal emission data

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

Machine‐Learning‐Based Monitoring of Laser Powder Bed Fusion

Towards Online Monitoring and Data-driven Control: A Study of Segmentation Algorithms for Laser Powder Bed Fusion Processes

Development of instrumentation for eddy current in situ monitoring of laser powder bed fusion

Research on an Online Monitoring Device for the Powder Laying Process of Laser Powder Bed Fusion

High-Fidelity Optical Monitoring of Laser Powder Bed Fusion via Aperture Division Multiplexing

Online Quality Monitoring in Material Extrusion Additive Manufacturing Processes Based on Laser Scanning Technology

Methodology to Determine Melt Pool Anomalies in Powder Bed Fusion of Metals Using a Laser Beam by Means of Process Monitoring and Sensor Data Fusion

Defect detection by multi-axis infrared process monitoring of laser beam directed energy deposition

Spatter feature-based in-situ identification of disturbances in process parameters during laser powder bed fusion

Semi-supervised machine learning of optical in-situ monitoring data for anomaly detection in laser powder bed fusion

Deep semi-supervised learning of dynamics for anomaly detection in laser powder bed fusion