Vibration Fault Diagnosis of Rotating Machinery Using PSO-Based Radial Basis Function Network

2013 ◽  
Vol 302 ◽  
pp. 474-480
Author(s):  
Huo Ching Sun ◽  
Chao Ming Huang ◽  
Yann Chang Huang ◽  
Hsing Feng Chen
Keyword(s):  
Fault Diagnosis ◽  
Radial Basis Function ◽  
Basis Function ◽  
Radial Basis Function Network ◽  
Back Propagation ◽  
Pso Algorithm ◽  
Back Propagation Algorithm ◽  
Error Back Propagation ◽  
Radial Basis ◽  
Function Network

A particle swarm optimization-based radial basis function network (PSO-RBFN) is presented to diagnose vibration faults of steam turbine-generator sets (STGS) in a power plant. The proposed PSO algorithm is used to automatically tune the control parameters of the RBFN. The test results demonstrate that the proposed PSO-RBFN has a higher diagnostic accuracy than the RBFN and multilayer perceptron network (MLPN) trained by error back-propagation algorithm. Moreover, this paper has demonstrated that the proposed PSO-RBFN can be as a reliable tool for vibration fault diagnosis of STGS.

Intelligent Parallel Networks for Combustion Quality Monitoring in Power Station Boilers

2013 ◽  
Vol 699 ◽  
pp. 893-899 ◽  
Author(s):  
K. Sujatha ◽  
N. Pappa ◽  
U. Siddharth Nambi ◽  
C.R. Raja Dinakaran ◽  
K. Senthil Kumar
Keyword(s):  
Radial Basis Function ◽  
Basis Function ◽  
Radial Basis Function Network ◽  
Research Work ◽  
Back Propagation ◽  
Average Intensity ◽  
Back Propagation Algorithm ◽  
Feature Size ◽  
Propagation Algorithm ◽  
Radial Basis

This research work includes a combination of Fisher’s Linear Discriminant (FLD) analysis by combining Radial Basis Function Network (RBF) and Back Propagation Algorithm (BPA) for monitoring the combustion conditions of a coal fired boiler so as to control the air/fuel ratio. For this two dimensional flame images are required which was captured with CCD camera whose features of the images, average intensity, area, brightness and orientation etc., of the flame are extracted after pre-processing the images. The FLD is applied to reduce the n-dimensional feature size to 2 dimensional feature size for faster learning of the RBF. Also three classes of images corresponding to different burning conditions of the flames have been extracted from a continuous video processing. In this the corresponding temperatures, the Carbon monoxide (CO) emissions and other flue gases have been obtained through measurement. Further the training and testing of Parallel architecture of Radial Basis Function and Back Propagation Algorithm (PRBFBPA) with the data collected have been done and the performance of the algorithms is presented.

Sign in / Sign up

Export Citation Format

Share Document