OpenAIR @ RGU >
Design and Technology >
Engineering >
Theses (Engineering) >

Please use this identifier to cite or link to this item:
This item has been viewed 102 times in the last year. View Statistics

Files in This Item:

File Description SizeFormat
Stewart Elder PhD.pdf10.73 MBAdobe PDFView/Open
Title: The monitoring of induction motor starting transients with a view to early fault detection.
Authors: Elder, Stewart
Supervisors: Watson, John
Thomson, W. T.
Maxwell, Grant M.
Issue Date: Mar-1992
Publisher: The Robert Gordon University
Citation: ELDER, S., WATSON, J. F. and THOMSON, W. T., 1989. Fault detection in induction motors as a result of transient analysis. 4th IEE International Conference on Electrical Machines. London. pp. 182-184
ELDER, S., WATSON, J. F. and THOMSON, W. T., 1989. The monitoring and analysis of transient currents for the purpose of fault detection in induction motors.. 24th Universities Power Engineering Conference (UPEC). Belfast.
ELDER, S., WATSON, J. F. and THOMSON, W. T., 1990. The analysis of transient currents in induction motors for the purpose of detecting motor faults.
Abstract: The aim of this work is to investigate the possibility of detecting faults in a 3 phase Induction motor by monitoring and analysing the transient line current waveform during the starting period. This is a particularly onerous time for the machine and the inter-relationships between parameters such as current, torque, speed and time are very complex. As a result two parallel paths of investigation have been followed, by methods of experimentation and computer simulation. Transient line current signals have been obtained from purpose built test rigs and these signals have been analysed in both the time and frequency domains. In order to assist with the comprehension of this data a sophisticated computer simulation of the induction motor during the starting period has also been developed. Computer simulation of the induction motor has been developed initially using the two and then three phase induction motor voltage equations which are solved by numerical integration. Using these techniques it has been possible to detect small degrees of fault level for both wound and cage rotor machines by analysing the line current waveform during the starting period. Good agreement has been found between the real and simulated data. A range of Digital Signal Processing techniques have been utilised to extract the components indicative of rotor faults. These techniques were at first wideband and highly numerically intensive, some originating from Speech Processing. The final processing techniques were far simpler and selected by analysis of the results from experimental data, both real and simulated.
Appears in Collections:Theses (Engineering)

All items in OpenAIR are protected by copyright, with all rights reserved.


   Disclaimer | Freedom of Information | Privacy Statement |Copyright ©2012 Robert Gordon University, Garthdee House, Garthdee Road, Aberdeen, AB10 7QB, Scotland, UK: a Scottish charity, registration No. SC013781