Title :
Realtime diagnostic prognostic solution for life cycle management of thermomechanical system
Author :
Saxena, Bhavaye ; Kumar, Ajit ; Srivastava, Anurag ; Goe, Alok
Author_Institution :
Univ. of Ottawa, Ottawa, ON, Canada
Abstract :
Data based approach and methodology for real time diagnosis and prognosis solutions for thermomechanical systems is discussed. Coated turbine blade operation is emulated to sensor online temperature data as the real-time inputs for the software code developed. An algorithm is presented first and extended sampling based statistical hypothesis tests are used for anomaly detection tests. Paired t-test and rank sum hypotheses test are found to be appropriate for different data set combinations. Matlab statistical package is used for the software code. The algorithm and methodology work well with laboratory temperature data and seeded faults. Few limitations of the developed system code are identified.
Keywords :
blades; condition monitoring; fault diagnosis; mathematics computing; mechanical engineering computing; product life cycle management; sampling methods; software engineering; turbines; Matlab statistical package; anomaly detection tests; databases; rank sum hypotheses test; realtime diagnostic prognostic solution; sampling; sensor online temperature data; software code development; statistical hypothesis tests; t-test hypotheses; thermomechanical system life cycle management; turbine blade operation; Algorithm design and analysis; Engines; Real time systems; Temperature distribution; Temperature measurement; Temperature sensors; Turbines; Temperature; anomaly; diagnosis; prognosis; reference; t-test;
Conference_Titel :
Electrical and Computer Engineering (CCECE), 2011 24th Canadian Conference on
Conference_Location :
Niagara Falls, ON
Print_ISBN :
978-1-4244-9788-1
Electronic_ISBN :
0840-7789
DOI :
10.1109/CCECE.2011.6030610
