Heterogeneous sensor data fusion for multiscale, shape agnostic flaw detection in laser powder bed fusion additive manufacturing

Bevans, Benjamin; Barrett, Christopher B.; Spears, Thomas; Gaikwad, Aniruddha; Riensche, Alex; Smoqi, Ziyad; Halliday, Harold; Rao, Prahalada

doi:10.1080/17452759.2023.2196266

Cited by 11 publications

(5 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jacobsmühlen [10] Inconel 625, Inconel 718 SVM, RF, SGD Structure defects Zhang [11] Inconel 625 Topography analysis Flatness of powder bed Wang [13] 30CrMnSiNi2A Topography analysis Flatness of powder bed Xiong [14] / Visual binocular vision Thin-walled components Scime [15] Inconel 718 MsCNN Flatness of powder bed Li [12] Metal Topography analysis Flatness of powder bed Gaikwrad [17] Ti-6Al-4V CNN Geometric integrity Liu [18] Ti-6Al-4 V alloy CNN EBM powder bed Bevans [19] Inconel 718 Multi-sensor fusion Part-level and micro defects Repossini [21] Nickel alloy Statistics Melt pool Rodriguez [22] Ti-6Al-4V Regression Melt pool Li [23] ZrO 2 , SiO 2 Finite element simulation Slurry flow Zhao [24] Alumina ceramics Finite element simulation lamellar structure…”

Section: Materials Methods Targetmentioning

confidence: 99%

See 1 more Smart Citation

High-Performance Defect Detection Methods for Real-Time Monitoring of Ceramic Additive Manufacturing Process Based on Small-Scale Datasets

Jia,

Li,

Wang

et al. 2024

Processes

View full text Add to dashboard Cite

Vat photopolymerization is renowned for its high flexibility, efficiency, and precision in ceramic additive manufacturing. However, due to the impact of random defects during the recoating process, ensuring the yield of finished products is challenging. At present, the industry mainly relies on manual visual inspection to detect defects; this is an inefficient method. To address this limitation, this paper presents a method for ceramic vat photopolymerization defect detection based on a deep learning framework. The framework innovatively adopts a dual-branch object detection approach, where one branch utilizes a fully convolution network to extract the features from fused images and the other branch employs a differential Siamese network to extract the differential information between two consecutive layer images. Through the design of the dual branches, the decoupling of image feature layers and image spatial attention weights is achieved, thereby alleviating the impact of a few abnormal points on training results and playing a crucial role in stabilizing the training process, which is suitable for training on small-scale datasets. Comparative experiments are implemented and the results show that using a Resnet50 backbone for feature extraction and a HED network for the differential Siamese network module yields the best detection performance, with an obtained F1 score of 0.89. Additionally, as a single-stage defect object detector, the model achieves a detection frame rate of 54.01 frames per second, which meets the real-time detection requirements. By monitoring the recoating process in real-time, the manufacturing fluency of industrial equipment can be effectively enhanced, contributing to the improvement of the yield of ceramic additive manufacturing products.

show abstract

Section: Materials Methods Targetmentioning

confidence: 99%

“…However, this method may not perform well on surfaces with significant fluctuations. Bevans [19] utilized various types of optical sensors to capture images during the manufacturing process of Inconel 718 nickel alloy parts. By analyzing spectral images, they detected defects at part level, medium scale, and micro scale.…”

Section: Related Workmentioning

confidence: 99%

High-Performance Defect Detection Methods for Real-Time Monitoring of Ceramic Additive Manufacturing Process Based on Small-Scale Datasets

Jia,

Li,

Wang

et al. 2024

Processes

View full text Add to dashboard Cite

show abstract

“…Recent literature has proposed more cost-effective, flexible solutions, such as in-situ acoustic monitoring, demonstrating promising potential for defect detection tasks [167][168][169]. Another important future direction is the development of multi-sensor monitoring systems, which combine the strengths of various sensing techniques to enhance defect detection outcomes [170]. This approach can help overcome the limitations of individual sensors and provide more comprehensive and reliable information for in-situ monitoring.…”

Section: Machine Learning Assisted Lam Of Ti Alloysmentioning

confidence: 99%

“…Current research in machine learning for understanding process-microstructure-property relationships in LAMproduced Ti alloys is still in its early stages. Future efforts are needed to develop more advanced machine learning models that can decipher the complex interdependencies among processing parameters, microstructural features, and material properties, as well as uncover causal relationships for improved process control and alloy performance [170]. In addition, constructing and sharing Ti-6Al-4V material PSP relationship datasets, such as the one presented by Luo et al [171], is of particular importance in stimulating research progress in this area.…”

Section: Machine Learning Assisted Lam Of Ti Alloysmentioning

confidence: 99%

Recent innovations in laser additive manufacturing of titanium alloys

Su,

Jiang,

Teng

et al. 2024

Int. J. Extrem. Manuf.

View full text Add to dashboard Cite

Titanium (Ti) alloys are widely used in frontier fields like aerospace and biomedical engineering. Laser additive manufacturing (LAM), as an innovative technology, is the key driver for the development of Ti alloys. Despite the significant advancements in LAM of Ti alloys, there remain challenges that need further research and development efforts. To recap the potential of LAM high-performance Ti alloy, this article systematically reviews LAM Ti alloys with up-to-date information on process, materials, and properties. Several feasible solutions to advance LAM Ti alloys were reviewed, including intelligent process parameters optimization, LAM process innovation with auxiliary fields and novel Ti alloys customization for LAM. The auxiliary energy fields (e.g., thermal, acoustic, mechanical deformation and magnetic fields) that can be applied during LAM Ti alloys affect the melt pool dynamics and solidification behaviour, altering microstructures and mechanical performances. Different kinds of novel Ti alloys customized for LAM, like peritectic α-Ti, eutectoid (α+β)-Ti, hybrid (α+β)-Ti, isomorphous β-Ti and eutectic β-Ti alloys are reviewed in detail. Furthermore, machine learning in accelerating the LAM process optimization and new materials development is also outlooked. The review summarizes the material properties and performance envelops and benchmarks the research achievements in LAM of Ti alloys. In addition, the perspectives and further trends in LAM of Ti alloys are also highlighted.

show abstract

“…For example, M. Abdelrahman [ 18 ] and others, utilized a high-resolution optical imaging monitoring system to photograph the powder bed before and after laser scanning, which used multiple light sources from different directions to construct the image, and then created a binary template from a sliced 3D model of the part, which was utilized to index the optical image data to the part geometry, which ultimately allowed for the detection of defects in the part defects in the area of the part; B. Shi et al [ 19 ], proposed to build a powder bed inspection system using an industrial camera and multiple illumination sources, and proposed a better illumination strategy by investigating the expression of defective features under different illumination, and also utilized image feature enhancement and adaptive threshold segmentation algorithm based on the grayscale features of the powder bed image for separating defective regions and based on the three convolutional neural network algorithms, namely, AlexNet, RexNet50, and VGG16—three kinds of convolutional neural network algorithms on the current powder layer exist in the stripe, ultra-high and incomplete powder laying three types of defective regions were experimentally compared and analyzed, and the results showed that the three kinds of defective data are prone to overfitting under the complex model. Other scholars have identified and detected defects in the powder laying process by using industrial cameras, infrared cameras, thermal cameras, and other devices combined with depth algorithms [ 20 , 21 , 22 , 23 , 24 ]. The above research has realized the acquisition of scraper motion signals and powder bed images in the powder spreading process by installing piezoelectric accelerometers on the scraper, installing industrial cameras, etc.…”

Section: Introductionmentioning

confidence: 99%

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

Wei,

Liu,

et al. 2024

Micromachines

View full text Add to dashboard Cite

Improving the quality of metal additive manufacturing parts requires online monitoring of the powder bed laying procedure during laser powder bed fusion production. In this article, a visual online monitoring tool for flaws in the powder laying process is examined, and machine vision technology is applied to LPBF manufacture. A multiscale improvement and model channel pruning optimization method based on convolutional neural networks is proposed, which makes up for the deficiencies of the defect recognition method of small-scale powder laying, reduces the redundant parameters of the model, and enhances the processing speed of the model under the premise of guaranteeing the accuracy of the model. Finally, we developed an LPBF manufacturing process laying powder defect recognition algorithm. Test experiments show the performance of the method: the minimum size of the detected defects is 0.54 mm, the accuracy rate of the feedback results is 98.63%, and the single-layer laying powder detection time is 3.516 s, which can realize the effective detection and control of common laying powder defects in the additive manufacturing process, avoids the breakage of the scraper, and ensures the safe operation of the LPBF equipment.

show abstract

Heterogeneous sensor data fusion for multiscale, shape agnostic flaw detection in laser powder bed fusion additive manufacturing

Cited by 11 publications

References 59 publications

High-Performance Defect Detection Methods for Real-Time Monitoring of Ceramic Additive Manufacturing Process Based on Small-Scale Datasets

High-Performance Defect Detection Methods for Real-Time Monitoring of Ceramic Additive Manufacturing Process Based on Small-Scale Datasets

Recent innovations in laser additive manufacturing of titanium alloys

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

Contact Info

Product

Resources

About