Machine learning with DSP for condition monitoring system

Abstract

Condition monitoring is the process of monitoring a parameter of condition in a system in order to identify a significant change which is indicative of a developing fault. It has a unique benefit in that conditions that would shorten normal lifespan can be addressed before they develop into a major failure. Using machine learning techniques, the big data gathered around a system can be analysed as a single coherent whole to draw conclusions about its current state of health. This project will develop a condition monitoring method using machine learning to detect defects on a real life system. A test jig will be used to mimic a real life system to collect sufficient data for machine learning. A DSP will be used to implement the machine learning algorithm.