Date of Award

Summer 2009

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical & Computer Engineering

Program/Concentration

Electrical Engineering

Committee Director

Vishnu K. Lakdawala

Committee Member

Prathap Basappa

Committee Member

Linda Vahala

Call Number for Print

Special Collections LD4331.E55 N655 2009

Abstract

Partial Discharges (PD) have been traditionally used to assess the state of any insulation system and its remnant life. In earlier work, Perspex (PMMA) samples with a needle plane gap have been aged with AC voltage. Their tree growth was monitored simultaneously by collecting PD at regular intervals of time and taking microphotographs in real time without interrupting the aging voltage. The obtained partial discharge pulse amplitude records were clustered together into groups of class intervals. The sequence of PD pulse height records was quantified as a time series of shape (η), and scale (σ) parameters of a Weibull distribution. This thesis describes two new techniques to analyze and predict the pulse height distribution parameters of PD (η and σ): Linear prediction and artificial neural networks. To test these techniques, we have analyzed the experimental results for the two samples of data previously obtained. Simulation results in MATLAB show that both methods predict the future values of each sample with optimal mean square errors. The relative advantages and limitations of each approach are discussed. A state of the art experimental system to conduct PD measurements and analysis was built as part of the present work. This system will be used for future research work.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/3dg5-z743

Download

Share

COinS