Intelligent Rocket Condition Monitoring and Fault Detection using Deep Learning
J Jagadeesan1, Hardik Agrawal2, Aditya Iyer3, Vishal Deo Mahto4, Harshit Soni5

1Dr. J Jagadeesan, P.hD, Professor and Head, Department of Computer Science and Engineering, SRM IST, Chennai (Tamil Nadu), India.
2Hardik Agrawal B.Tech, Department of Computer Science and Engineering, SRM IST, Chennai (Tamil Nadu), India.
3Aditya Iyer B.Tech, Department of Computer Science and Engineering, SRM IST, Chennai (Tamil Nadu), India.
4Vishal Deo Mahto B.Tech, Department of Computer Science and Engineering, SRM IST, Chennai (Tamil Nadu), India.
5Harshit Soni B.Tech, Department of Computer Science and Engineering, SRM IST, Chennai (Tamil Nadu), India.

Manuscript received on 01 May 2019 | Revised Manuscript received on 15 May 2019 | Manuscript published on 30 May 2019 | PP: 2807-2811 | Volume-8 Issue-7, May 2019 | Retrieval Number: F3656048619/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Machines, including rockets can develop fatigue over a specific course of time, due to being subjected to constant pressure, and physical functioning. This can cause heat to be built up in certain parts of the machine, or rocket. This puddled up heat can cause leakage in energy, and may result in the loss of energy, due to heat dissipation. The motive of the article is to monitor the health of a rocket. In this, we investigate how the new age technology of deep learning can be applied to images, more specifically, infrared thermal images, to automatically check and to accurately determine the condition of a rocket/machine. We use two cases, i.e. machine fault detection and oil/coolant level prediction, and show that the proposed system is able to detect numerous conditions in normal machines, and even in rockets, very accurately without requiring any detailed knowledge about its general physics, or its functionality, or structure, by taking thermal images as the general input, and taking the excessive heating issues in the rocket engines into consideration. This system incorporates CNN, machine learning, and to a certain extent, deep learning for producing necessary results.
Keyword:  Thermal Images, Rocket, CNN, Neural Network, Deep Learning.
Scope of the Article: Deep Learning