EARLY MAINTENANCE AND DIAGNOSIS OF CONNECTED MACHINES USING MACHINE LEARNING
Journal: International Journal of Electrical Engineering and Technology (IJEET) (Vol.11, No. 4)Publication Date: 2020-06-30
Authors : R Srivatsan; Sharath Cherian Thomas Venugopal P Ravi Kumar C. V;
Page : 412-421
Keywords : Sensors; Temperature; vibration; health prediction; microcontroller; machine learning; 5G; IoT; M2M.;
Abstract
In this paper a novel system is proposed to monitor the health of industrial machines thus helping in their maintenance and early failure detection. This will help in prediction of when an industrial machine or its part will malfunction based on the data extracted from it. We will be able to replace the part or the machine in advance before any production lines get affected. This way resources will be saved and also cost of maintenance will be reduced. On top of this it is not possible to always manually monitor machines placed in remote areas like motors and pumps in water supply systems, sewage plants and chemical plants but our system would be able to not just monitor the machines but also predict the machine's current health. The proposal encompasses a lightweight machine monitoring system for next generation M2M ecosystem for on the fly fault detection and diagnosis.
Other Latest Articles
- HIGH VOLTAGE ENGINEERING USING ACTIVITY BASED LEARNING
- COMPARATIVE ANALYSIS OF UPQC WITH VARIOUS HYSTERESIS CURRENT CONTROLLERS
- PHC CONTROL METHOD FOR SHUNT ACTIVE POWER FILTERS IN THREE-PHASE FOURWIRE SYSTEMS
- IMPLEMENTATION OF A SMART FLASK USING INDUCTION HEATING AND ARDUINO MICROCONTROLLER
- DESIGN OF SHARED BUFFER ARCHITECTURE FOR CPU-GPU ON CHIP NETWORK
Last modified: 2021-03-04 16:13:52