Title :
The fault diagnosis for centrifugal compressor based on time series analysis with neutral network
Author :
Hong-Jian Liao ; Sheng-Zhong Huang
Author_Institution :
Dept. of Math. & Comput. Sci., Liuzhou Teachers Coll., Liuzhou, China
Abstract :
In order to have fault diagnosis for centrifugal compressor correctly, the times series analysis with neural network was carried out for it. The theory model of time series analysis method was put forward firstly, and the autoregressive coefficients diagnosis method was discussed then. And the neural network diagnosis method with AR coefficients was put forward at last. And the fault diagnosis results of centrifugal compressor were acquired based on BPNN and distance diagnosis method, and diagnosis results was good.
Keywords :
autoregressive processes; backpropagation; compressors; fault diagnosis; mechanical engineering computing; neural nets; time series; BPNN; autoregressive coefficients; centrifugal compressor; distance diagnosis method; fault diagnosis; neutral network; time series analysis; Humans; BPNN; centrifugal compressor; fault diagnosis;
Conference_Titel :
Advanced Computer Theory and Engineering (ICACTE), 2010 3rd International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-6539-2
DOI :
10.1109/ICACTE.2010.5579406