Standard Article

Machine Learning Techniques

Signal Processing

  1. Fulei Chu,
  2. Shengfa Yuan,
  3. Zhike Peng

Published Online: 15 SEP 2009

DOI: 10.1002/9780470061626.shm184

Encyclopedia of Structural Health Monitoring

Encyclopedia of Structural Health Monitoring

How to Cite

Chu, F., Yuan, S. and Peng, Z. 2009. Machine Learning Techniques. Encyclopedia of Structural Health Monitoring. .

Author Information

  1. Tsinghua University, Department of Precision Instruments and Mechanology, Beijing, China

Publication History

  1. Published Online: 15 SEP 2009

Abstract

Machine learning is a hot topic of artificial intelligence, and the interests of researchers on the same have grown fast in recent years. Especially, the great potential of machine learning techniques in classification makes it more popular in the research areas where pattern recognition is involved, including the development of condition monitoring and fault diagnosis for machines and structures. This article gives an outline to machine learning including the background and some key algorithms and theories that form the core of machine learning. The use of machine learning techniques in condition monitoring and fault diagnosis has also been discussed along with a basic framework of the machine-learning-based fault diagnosis system. More efforts have been made to introduce support vector machine (SVM), which is a significant development in machine learning. The basic procedure of the machine-learning-based fault diagnosis system used to perform the fault diagnosis on a turbo-pump rotor has been illustrated by using SVM. The article is mainly intended for the readers who want a simple but exhaustive introduction to this new learning system.

Keywords:

  • machine learning;
  • artificial intelligence;
  • neural network;
  • support vector machines;
  • fault diagnosis;
  • structural health monitoring