Deep Learning of Variant Geometry in Layerwise Imaging Profiles for Additive Manufacturing Quality Control

2019 ◽  
Vol 141 (11) ◽  
Author(s):  
Farhad Imani ◽  
Ruimin Chen ◽  
Evan Diewald ◽  
Edward Reutzel ◽  
Hui Yang
Keyword(s):  
Quality Control ◽  
Deep Learning ◽  
Additive Manufacturing ◽  
Computer Aided Design ◽  
Flaw Detection ◽  
Regions Of Interest ◽  
Monitoring And Control ◽  
New Paradigm ◽  
Image Guided ◽  
Process Monitoring And Control

Abstract Additive manufacturing (AM) is a new paradigm in design-driven build of customized products. Nonetheless, mass customization and low-volume production make the AM quality assurance extremely challenging. Advanced imaging provides an unprecedented opportunity to increase information visibility, cope with the product complexity, and enable on-the-fly quality control in AM. However, in situ images of a customized AM build show a high level of layer-to-layer geometry variation, which hampers the use of powerful image-based learning methods such as deep neural networks (DNNs) for flaw detection. Very little has been done on deep learning of variant geometry for image-guided process monitoring and control. The proposed research is aimed at filling this gap by developing a novel machine learning approach that is focused on variant geometry in each layer of the AM build, namely region of interests, for the characterization and detection of layerwise flaws. Specifically, we leverage the computer-aided design (CAD) file to perform shape-to-image registration and to delineate the regions of interest in layerwise images. Next, a hierarchical dyadic partitioning methodology is developed to split layer-to-layer regions of interest into subregions with the same number of pixels to provide freeform geometry analysis. Then, we propose a semiparametric model to characterize the complex spatial patterns in each customized subregion and boost the computational speed. Finally, a DNN model is designed to learn variant geometry in layerwise imaging profiles and detect fine-grained information of flaws. Experimental results show that the proposed deep learning methodology is highly effective to detect flaws in each layer with an accuracy of 92.50 ± 1.03%. This provides a significant opportunity to reduce interlayer variation in AM prior to completion of a build. The proposed methodology can also be generally applicable in a variety of engineering and medical domains that entail customized design, variant geometry, and image-guided process control.

Image-Guided Variant Geometry Analysis of Layerwise Build Quality in Additive Manufacturing

2019 ◽  
Author(s):  
Farhad Imani ◽  
Ruimin Chen ◽  
Evan Diewald ◽  
Edward Reutzel ◽  
Hui Yang
Keyword(s):  
Quality Control ◽  
Additive Manufacturing ◽  
Process Monitoring ◽  
Flaw Detection ◽  
Monitoring And Control ◽  
Image Guided ◽  
Process Monitoring And Control ◽  
And Control ◽  
Geometry Analysis ◽  
The Impact

Abstract Additive manufacturing (AM) is a new paradigm in design-driven build of customized products. Nonetheless, mass customization and low volume production make the AM quality assurance extremely challenging. Advanced imaging provides an unprecedented opportunity to increase information visibility, cope with the product complexity, and enable on-the-fly quality control in AM. However, in-situ images of a customized AM build show a high level of layer-to-layer geometry variation, which hampers the use of powerful image-based learning methods such as deep neural networks (DNNs) for flaw detection. Few, if any, previous works investigated how to tackle the impact of AM customization on image-guided process monitoring and control. The proposed research is aimed at filling this gap by developing a novel real-time and multi-scale process monitoring methodology for quality control of customized AM builds. Specifically, we leverage the computer-aided design (CAD) file to perform shape-to-image registration and delineate the regions of interests in lay-erwise images. Next, a hierarchical dyadic partitioning methodology is developed to split layer-to-layer regions of interest into subregions with the same number of pixels to provide freeform geometry analysis. Then, we propose a semiparametric model to characterize the complex spatial patterns in each customized subregion and boost the computational speed. Finally, a DNN model is designed to learn and detect fine-grained information of flaws. Experimental results show that the proposed process monitoring and control methodology detects flaws in each layer with an accuracy of 92.50±1.03%. This provides an opportunity to reduce inter-layer variation in AM prior to completion of the build. The proposed methodology can also be generally applicable in a variety of engineering and medical domains that entail image-based process monitoring and control with customized designs.

Fractal Pattern Recognition of Image Profiles for Manufacturing Process Monitoring and Control

2018 ◽  
Author(s):  
Farhad Imani ◽  
Bing Yao ◽  
Ruimin Chen ◽  
Prahalada Rao ◽  
Hui Yang
Keyword(s):  
Additive Manufacturing ◽  
Process Monitoring ◽  
Manufacturing Process ◽  
Manufacturing Systems ◽  
Experimental Studies ◽  
Precision Machining ◽  
Monitoring And Control ◽  
Process Monitoring And Control ◽  
Ultra Precision ◽  
And Control

Nowadays manufacturing industry faces increasing demands to customize products according to personal needs. This trend leads to a proliferation of complex product designs. To cope with this complexity, manufacturing systems are equipped with advanced sensing capabilities. However, traditional statistical process control methods are not concerned with the stream of in-process imaging data. Also, very little has been done to investigate nonlinearity, irregularity, and inhomogeneity in image stream collected from manufacturing processes. This paper presents the multifractal spectrum and lacunarity measures to characterize irregular and inhomogeneous patterns of image profiles, as well as detect the hidden dynamics of the underlying manufacturing process. Experimental studies show that the proposed method not only effectively characterizes the surface finishes for quality control of ultra-precision machining but also provides an effective model to link process parameters with fractal characteristics of in-process images acquired from additive manufacturing. This, in turn, will allow a swift response to processes changes and consequently reduce the number of defective products. The proposed fractal method has strong potentials to be applied for process monitoring and control in a variety of domains such as ultra-precision machining, additive manufacturing, and biomanufacturing.

CubeSat Fabrication through Additive Manufacturing and Micro-Dispensing

2011 ◽  
Vol 2011 (1) ◽  
pp. 001021-001027 ◽  
Author(s):  
Cassie Gutierrez ◽  
Rudy Salas ◽  
Gustavo Hernandez ◽  
Dan Muse ◽  
Richard Olivas ◽  
...  
Keyword(s):  
Additive Manufacturing ◽  
Computer Aided Design ◽  
Printed Circuit Boards ◽  
Fused Deposition Modeling ◽  
Integrated Approach ◽  
Layer By Layer ◽  
New Paradigm ◽  
Spatial Gradients ◽  
Fused Deposition ◽  
On Demand

Fabricating entire systems with both electrical and mechanical content through on-demand 3D printing is the future for high value manufacturing. In this new paradigm, conformal and complex shapes with a diversity of materials in spatial gradients can be built layer-by-layer using hybrid Additive Manufacturing (AM). A design can be conceived in Computer Aided Design (CAD) and printed on-demand. This new integrated approach enables the fabrication of sophisticated electronics in mechanical structures by avoiding the restrictions of traditional fabrication techniques, which result in stiff, two dimensional printed circuit boards (PCB) fabricated using many disparate and wasteful processes. The integration of Additive Manufacturing (AM) combined with Direct Print (DP) micro-dispensing and robotic pick-and-place for component placement can 1) provide the capability to print-on-demand fabrication, 2) enable the use of micron-resolution cavities for press fitting electronic components and 3) integrate conductive traces for electrical interconnect between components. The fabrication freedom introduced by AM techniques such as stereolithography (SL), ultrasonic consolidation (UC), and fused deposition modeling (FDM) have only recently been explored in the context of electronics integration and 3D packaging. This paper describes a process that provides a novel approach for the fabrication of stiff conformal structures with integrated electronics and describes a prototype demonstration: a volumetrically-efficient sensor and microcontroller subsystem scheduled to launch in a CubeSat designed with the CubeFlow methodology.

scholarly journals Real-Time 3D Surface Measurement in Additive Manufacturing Using Deep Learning

2021 ◽  
Author(s):  
Liu Chenang ◽  
Wang Rongxuan ◽  
Zhenyu Kong ◽  
Babu Suresh ◽  
Joslin Chase ◽  
...  
Keyword(s):  
Deep Learning ◽  
Additive Manufacturing ◽  
Surface Morphology ◽  
Real Time ◽  
Learning Algorithm ◽  
Surface Measurement ◽  
Monitoring And Control ◽  
Deep Learning Algorithm ◽  
3D Surface ◽  
3D Surface Morphology

Layer-wise 3D surface morphology information is critical for the quality monitoring and control of additive manufacturing (AM) processes. However, most of the existing 3D scan technologies are either contact or time consuming, which are not capable of obtaining the 3D surface morphology data in a real-time manner during the process. Therefore, the objective of this study is to achieve real-time 3D surface data acquisition in AM, which is achieved by a supervised deep learning-based image analysis approach. The key idea of this proposed method is to capture the correlation between 2D image and 3D point cloud, and then quantify this relationship by using a deep learning algorithm, namely, convolutional neural network (CNN). To validate the effectiveness and efficiency of the proposed method, both simulation and real-world case studies were performed. The results demonstrate that this method has strong potential to be applied for real-time surface morphology measurement in AM, as well as other advanced manufacturing processes.

Joint Multifractal and Lacunarity Analysis of Image Profiles for Manufacturing Quality Control

2019 ◽  
Vol 141 (4) ◽  
Author(s):  
Farhad Imani ◽  
Bing Yao ◽  
Ruimin Chen ◽  
Prahalad Rao ◽  
Hui Yang
Keyword(s):  
Quality Control ◽  
Additive Manufacturing ◽  
Manufacturing Systems ◽  
Manufacturing Industry ◽  
Experimental Studies ◽  
Imaging Data ◽  
Monitoring And Control ◽  
Ultraprecision Machining ◽  
Statistical Process ◽  
Lacunarity Analysis

The modern manufacturing industry faces increasing demands to customize products according to personal needs, thereby leading to the proliferation of complex designs. To cope with design complexity, manufacturing systems are increasingly equipped with advanced sensing and imaging capabilities. However, traditional statistical process control methods are not concerned with the stream of in-process imaging data. Also, very little has been done to investigate nonlinearity, irregularity, and inhomogeneity in the image stream collected from manufacturing processes. This paper presents the joint multifractal and lacunarity analysis to characterize irregular and inhomogeneous patterns of image profiles, as well as detect the hidden dynamics in the manufacturing process. Experimental studies show that the proposed method not only effectively characterizes surface finishes for quality control of ultraprecision machining but also provides an effective model to link process parameters with fractal characteristics of in-process images acquired from additive manufacturing. This, in turn, will allow a swift response to processes changes and consequently reduce the number of defective products. The proposed multifractal method shows strong potentials to be applied for process monitoring and control in a variety of domains such as ultraprecision machining and additive manufacturing.

A Review on Process Monitoring and Control in Metal-Based Additive Manufacturing

2014 ◽  
Vol 136 (6) ◽  
Author(s):  
Gustavo Tapia ◽  
Alaa Elwany
Keyword(s):  
Additive Manufacturing ◽  
Process Monitoring ◽  
Monitoring And Control ◽  
Part Quality ◽  
Industrial Sectors ◽  
Process Monitoring And Control ◽  
Traditional Manufacturing ◽  
Manufacturing Technologies ◽  
And Control ◽  
Am Processes

There is consensus among both the research and industrial communities, and even the general public, that additive manufacturing (AM) processes capable of processing metallic materials are a set of game changing technologies that offer unique capabilities with tremendous application potential that cannot be matched by traditional manufacturing technologies. Unfortunately, with all what AM has to offer, the quality and repeatability of metal parts still hamper significantly their widespread as viable manufacturing processes. This is particularly true in industrial sectors with stringent requirements on part quality such as the aerospace and healthcare sectors. One approach to overcome this challenge that has recently been receiving increasing attention is process monitoring and real-time process control to enhance part quality and repeatability. This has been addressed by numerous research efforts in the past decade and continues to be identified as a high priority research goal. In this review paper, we fill an important gap in the literature represented by the absence of one single source that comprehensively describes what has been achieved and provides insight on what still needs to be achieved in the field of process monitoring and control for metal-based AM processes.

Characteristics and Technologies of Advanced CNC Systems

2011 ◽  
pp. 519-526 ◽  
Author(s):  
M. Minhat ◽  
X.W. Xu
Keyword(s):  
Computer Aided Design ◽  
Manufacturing Industry ◽  
Numerical Control ◽  
Monitoring And Control ◽  
Cnc Machine ◽  
Computer Aided ◽  
Process Monitoring And Control ◽  
Key Issues ◽  
Aided Design ◽  
And Control

Computer Numerical Control (CNC) systems are the “backbones” of modern manufacturing industry for over the last 50 years and the machine tools have evolved from simple machines with controllers that had no memory and were driven by punched tape, to today’s highly sophisticated, multiprocess workstations. These CNC systems are still being worked and improved on. The key issues center on autonomous planning, decision making, process monitoring and control systems that can adjust automatically to the changeable requirements. Introduction of CNC systems has made it possible to produce goods with consistent qualities, apart from enabling the industry to enhance productivity with a high degree of flexibility in a manufacturing system. CNC systems sit at the end of the process starting from product design using Computer Aided Design (CAD) tools to the generation of machining instructions that instruct a CNC machine to produce the final product. This process chain also includes Computer Aided Process Planning (CAPP) and Computer Aided Manufacturing (CAM).

Sign in / Sign up

Export Citation Format

Share Document