Publication Date


Document Type


First Advisor

Gupta, Abhijit

Degree Name

M.S. (Master of Science)

Legacy Department

Department of Mechanical Engineering


Mechanical engineering; Statistics; Gas-turbines--Performance--Research; Big data--Statistical methods


Gas turbines have many important variables such as the load, turbine speed, fuel gas flow, and inlet and outlet pressures. The volume, velocity, variability and complexity of the data from various sensors are huge. Monitoring of gas turbines consequently needs big data analytics which is the process of collecting, organizing and analyzing large data sets. The need for big data analytics stems from the need to increase the efficiency, improve operations and predict various trends and comment on the performance. Big data implies that the data sets are too large to be analyzed or even viewed by conventional methods and software. The gas turbine data set is a time series data.;In this thesis, a large data set from gas turbines is first made readable by converting it into the CSV format, as it is beyond the dimensional capability of Microsoft excel. The data is then analyzed using various statistical tools such as R-software. Combustion instabilities have been observed in the units and units with high dynamics have been determined. Data quality issues and missing data were observed. The limits of the blade path temperature spreads have been determined and the correlation of operational parameters were determined. Principle.;Component analysis was performed to reduce the dimensionality of the data and observe health of the gas turbine operation in terms of dynamic behavior.


Advisors: Abhijit Gupta.||Committee members: Sanjib Basu; Jenn-Terng Gau.


57 pages




Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type