Login | Register

GPU Based Real-time Welding Simulation with Smoothed-Particle Hydrodynamics


GPU Based Real-time Welding Simulation with Smoothed-Particle Hydrodynamics

Qing, Gu (2016) GPU Based Real-time Welding Simulation with Smoothed-Particle Hydrodynamics. Masters thesis, Concordia University.

Text (application/pdf)
Gu_MCompSc_S2016.pdf - Accepted Version
Available under License Creative Commons GNU GPL (Software).


Welding training is essential in the development of industrialization. A good welder will build robust workpieces that ensure the safety and stability of the product. However, training a welder requires lots of time and access professional welding equipment. Therefore, it is desirable to have a training system that is economical and easy to use. After decades development of computer graphics, sophisticated methodologies are developed in simulation fields, along the advanced hardware, enables the possibility of simulation welding with software. In this thesis, a novel prototype of welding training system is proposed. We use smoothed-particle hydrodynamics (SPH) method to simulate fluid as well as heat transfer and phase changing. In order to accelerate the processing to reach the level of real-time, we adopt CUDA to implement the SPH solver on GPU. Plus, Leap Motion is utilized as the input device to control the welding gun. As the result, the simulation reaches decent frame rate that allows the user control the simulation system interactively. The input device permits the user to adapt to the system in less than 5 minutes. This prototype shows a new direction in the training system that combines VR, graphics, and physics simulation. The further development of VR output device like Oculus Rift will enable the training system to a more immersive level.

Divisions:Concordia University > Gina Cody School of Engineering and Computer Science > Computer Science and Software Engineering
Item Type:Thesis (Masters)
Authors:Qing, Gu
Institution:Concordia University
Degree Name:M. Comp. Sc.
Program:Computer Science
Date:January 2016
Thesis Supervisor(s):Mudur, Sudhir and Popa, Tiberiu
ID Code:980845
Deposited By: QING GU
Deposited On:16 Jun 2016 14:39
Last Modified:18 Jan 2018 17:52
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Back to top Back to top