International Journal of Theoretical and Applied Mechanics

  • Year: 2010
  • Volume: 5
  • Issue: 1

Vibration based Gearbox Fault Detection Using Artificial Neural Networks

  • Author:
  • P.V. Srihari1, K. Ramachandra2, K. Govindarajulu3
  • Total Page Count: 9
  • DOI:
  • Page Number: 53 to 61

1Department of Mechanical Engineering, R.V. College of Engineering, Bangalore-560 059, India, Email: .

2Department of Mechanical Engineering, R.V. College of Engineering, Bangalore-560 059, India.

3Department of Mechanical Engineering, J.N.T.University, Anantapur, A.P., India.

null

Abstract

Fault diagnosis of gearboxes play an important role in increasing availability of machinery in condition monitoring. An effort has been made in this work to develop an ANN based diagnosis system to increase the reliability of fault diagnosis. Two prominent fault conditions of gears worn-out and broken tooth are being simulated and five feature parameters are extracted based on vibration signals and used as input features to the ANN diagnosis system developed in MATLAB, a three layered feed forward network using back propagation algorithm. The ANN system has been trained with 30 sets of data and tested with 10 sets of data. The learning rate, number of hidden layer neurons are varied one after the other and fixed optimal training parameters based on number of epochs. Among the five different learning rates used the 0.15 is deduced to be optimal one. And at that learning rate the number of hidden layer neurons of 9 found the optimal one out of three values of consideration. Then keeping the training parameters fixed, number of hidden layers is varied by comparing the performance of the networks and results show the two and three hidden layers have best classification accuracy.

Keywords

Gearbox Fault Diagnosis, Vibration signal, Artificial Neural Networks, Training data, optimization, Back propagation Algorithm