Towards intelligent monitoring system in wire arc additive manufacturing: a surface anomaly detector on a small dataset

Abstract

Rapid developments in artificial intelligence and image processing have presented many new opportunities for defect detection in manufacturing processes. In this work, an intelligent image processing system has been developed to monitor inter-layer deposition quality during a wire arc additive manufacturing (WAAM) process. This system reveals the feasibility and future potential of using computer vision knowledge in WAAM. Information produced from this system is to be used in conjunction with other quality monitoring systems to verify the quality of fabricated components. It is tailored to identify the presence of defects relating to lack of fusion and voids immediately after the deposition of a given layer. The image processing system is built upon the YOLOv3 architecture and through moderate changes on anchor settings and achieves 53% precision on surface anomaly detection and 100% accuracy in identifying the fabricated components’ location, providing a prerequisite for high-precision assessment of welding quality. The work presented in this paper presents an inter-layer vision-based defect monitoring system in WAAM and serves to highlight the feasibility of developing such intelligent computer vision systems for monitoring the WAAM process for defects.

Simultaneous Pore Detection and Morphological Features Extraction in Laser Powder Bed Fusion with Image Processing

“…On the other hand, compared with general models, specialized parameter adjustments and structural improvements to the model will achieve better detection results. However, the cost of investment in this attempt cannot be ignored [ 42 , 43 ]. The method in this paper can label most of the pore defects in the image, even the smaller gas pores.…”

Synopsis of Industry 5.0 Paradigm for Human-Robot Collaboration

“…Consequently, engineering education should emphasize human-robot interaction (HRI), particularly focusing on the diverse methods of interaction and cooperation with emerging CPS [68]. Human-assisted learning strategies should also be implemented to manage and regulate automated additive manufacturing systems and error detection mechanisms within manufacturing processes [69,70].…”

Intelligent Monitoring System for Real-Time Optimization of Ultra-Precision Manufacturing Processes

“…Key components include real-time monitoring, data collection, analysis, and decision-making capabilities. References (Jiang et al, 2022;, Wang et al, 2022;, Cheng et al, 2021;, Andronie et al, 2021;, Shi et al, 2019), andWu et al (2021) provide insights into various aspects of intelligent monitoring systems in manufacturing. These systems leverage technologies such as multi-sensor fusion, deep learning, cloud-based remote data collection, IoT, digital twins, and artificial intelligence to monitor and optimize manufacturing processes in real-time.…”

“…These systems enable features like state perception, real-time analysis, independent decision-making, and precise execution, which are essential for achieving efficient and high-quality manufacturing operations (Lan & Chen, 2021). Moreover, the utilization of technologies like machine learning, IIoT, high-precision sensors, and real-time data processing tools has become imperative for fault diagnosis, predictive maintenance, and energy efficiency in complex product manufacturing (Wang et al, 2022). Real-time optimization strategies, such as those involving metamodeling for process optimization (Morais & Araújo, 2023), are gaining significant industrial interest due to their ability to iteratively operate in closed loops and provide optimal set points for process control systems.…”