UTHM Institutional Repository

Gas turbine performance based creep life estimation using soft computing technique

Mohamed Zarti , Almehdi (2012) Gas turbine performance based creep life estimation using soft computing technique. Masters thesis, Universiti Tun Hussein Onn Malaysia.

[img]
Preview
PDF
ALMEHDI_MOHAMED_ZARTI.pdf

Download (804kB)

Abstract

Accurate and simple prediction system has become an urgent need in most disciplines. Having the accurate prediction system for gas turbine components will allow the user to produce reliable creep life prediction. Focusing on the turbine blades and its life, the current method to calculate its creep life is complex and consumes a lot of time. For this reason, the aim of this research is to use an alternative performance–based creep life estimation that is able to provide a quick solution and obtain accurate creep life prediction. By the use of an artificial neural network to predict creep life, a neural network architecture called Sensor Life Based (SLB) architecture that produces a direct mapping from gas path sensor to predict the blade creep life was created by using the gas turbine simulation performance software. The performance of gas turbine and the effects of multiple operations on the blade are studied. The result of the study is used to establish the input and output to train the Sensor Life Based network. The result shows that the Sensor Life-Based architecture is able to produce accurate creep life predictions yet performing rapid calculations. The result also shows that the accuracy of prediction depends on the way, how the gas path sensor is grouped together.

Item Type: Thesis (Masters)
Subjects: T Technology > TJ Mechanical engineering and machinery > TJ751-805 Miscellaneous motors and engines
Depositing User: Normajihan Abd. Rahman
Date Deposited: 05 Jan 2016 01:03
Last Modified: 05 Jan 2016 01:03
URI: http://eprints.uthm.edu.my/id/eprint/3644
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year