"

Cookies ussage consent

Our site saves small pieces of text information (cookies) on your device in order to deliver better content and for statistical purposes. You can disable the usage of cookies by changing the settings of your browser. By browsing our site without changing the browser settings you grant us permission to store that information on your device.

I agree, do not show this message again.

Simulated annealing wavelet neural network for vibration fault diagnosis of hydro turbine generating unit

XIAO ZHIHUAI1,* , SUN ZHAOHUI1, SONG LIBO1, ZHANG XIAOJING1, O. P. MALIK2

Affiliation

  1. Wuhan University, School of Power and Mechanical Engineering, Wuhan 430072, China
  2. University of Calgary, Department of Electrical and Computer Engineering, University of Calgary, Canada

Abstract

In view of non-linear characteristics between fault symptoms and fault types of hydro-turbine generating unit and defects of traditional wavelet neural network learning method, a wavelet neural network fault diagnosis model based on simulated annealing algorithm is designed and applied to the hydro turbine fault diagnosis . Instead of gradient descent method, the simulated annealing algorithm is applied to optimiz e parameters of wavelet neural network Example results show that the designed model has higher convergence precision and faster convergence speed compared wi th wavelet neural network and additional momentum BP neural network . The simulat ed annealing algorithm wavelet neural network can be effectively applied to hydro-turbine fault diagnosis, and it provides a new way for hydro-turbine fault diagnosis..

Keywords

Simulated annealing algorith mm; Wavelet neural network; Gradient descent Hydro-turbine generating unit; Fault diagnosis.

Submitted at: April 15, 2015
Accepted at: May 7, 2015

Citation

XIAO ZHIHUAI, SUN ZHAOHUI, SONG LIBO, ZHANG XIAOJING, O. P. MALIK, Simulated annealing wavelet neural network for vibration fault diagnosis of hydro turbine generating unit, Journal of Optoelectronics and Advanced Materials Vol. 17, Iss. 5-6, pp. 734-740 (2015)