Deep Learning Approach for Controlling Additive Manufacturing Process

مرکز و کتابخانه مطالعات اسلامی به زبان های اروپایی

منو

" Deep Learning Approach for Controlling Additive Manufacturing Process "
Razavi Arab, Nariman Banadaki, Yaser

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	1106503
Doc. No	:	TLpq2397555289
Main Entry	:	Banadaki, Yaser
	:	Razavi Arab, Nariman
Title & Author	:	Deep Learning Approach for Controlling Additive Manufacturing Process\ Razavi Arab, NarimanBanadaki, Yaser
College	:	Southern University and Agricultural and Mechanical College
Date	:	2019
student score	:	2019
Degree	:	M.S.
Page No	:	106
Abstract	:	Additive manufacturing (AM) is a crucial component of smart manufacturing systems that disrupts traditional supply chains. However, the parts built using the state-of-the-art powder-bed 3D printers have noticeable unpredictable mechanical properties. The attempts to improve the geometry and mechanical properties of final products in the 3D printing process were usually limited to the development of a typical control system using feedback from sensor measurement. However, a smarter control system is required to learn and adapt as the AM machine is operating. In this thesis, I propose a machine learning (ML) algorithm as a promising way of improving the underlying failure phenomena in 3D printing. I employ a Deep Convolutional Neural Network (DCNN) to automatically detect the defects in printing the layers, thereby turning 3D printers into essentially their own inspectors. We expect that the proposed DCNN model generates a precise feedback signal for a smart 3D printer to recognize any issues with the build itself to make proper adjustments and corrections without operator intervention. This can enhance the quality of the AM process, leading to manufacturing better parts with fewer quality hiccups, limiting waste of time and materials.
Subject	:	Artificial intelligence
	:	Computer science