|
| next >> |
Journal description
All volumes of the corresponding journal
All issues of the corresponding volume
All articles of the corresponding issues
Details for article 1 of 6 found articles
| Title: |
An Improved Neural Network Procedure for Assessing Accelerated Aging of Hydrogenerator Stator Bars |
| Author: |
Gleizer, H. Mannheimer, W. A. Nelson, J. K. |
| Appeared in: |
Electric power components & systems |
| Paging: | Volume 31 (2003) nr. 3 pages 275-284 |
| Year: |
2003-03 |
| Contents: |
This work presents an assessment of the capability of artificial neural networks to analyze the aging of hydrogenerator stator bar insulation. Four stator bars were artificially aged with thermal, electrical, and mechanical stresses in a laboratory setup. Aging was monitored by partial discharge measurements. A procedure was developed to preprocess the acquired data, which brought substantial improvement, with reduction in training time and increase in the efficiency of the neural networks analysis. The network showed good performance with error levels for the test data typically less than 10%, and was able to detect the singular aging behavior presented by one of the tested bars. |
| Publisher: |
Taylor & Francis |
| Source file: |
Elektronische Wetenschappelijke Tijdschriften |
Details for article 1 of 6 found articles
| next >> |