Session: 21-08 Diagnostic and prognostic methods

Paper Number: 82915

82915 - Statistical Rule Extraction for Gas Turbine Trip Prediction 

Gas turbine trip is an operational event that arises when undesirable operating conditions are approached or exceeded, and predicting its onset is a largely unexplored area. The application of novel artificial intelligence methods to this problem is interesting both from the computer science and the engineering point of view, and the results may be relevant in both the academia and the industry. In this paper we consider data gathered from a fleet of Siemens industrial gas turbines in operation that include several thermodynamic variables observed during a long period of operation. To assess the possibility of predicting trip events, we first apply a new, systematic statistical analysis to identify the most important variables, then we use a novel machine learning technique known as temporal decision trees, which differs from canonical ones because they allow a native treatment of the temporal component, and have an elegant logic interpretation that eases the post-hoc validation of the results, and, finally, we use the learned models to extract statistical rules. Our results are encouraging and consistent, and, to our knowledge, this is a first attempt to use such an approach not only in the gas turbine field, but also in the whole industry domain.

Presenting Author: Ionel Eduard Stan Universitá degli Studi di Ferrara

Presenting Author Biography: Ionel Eduard Stan is a PhD student at the University of Ferrara. His research interests include temporal reasoning and learning from temporal data.

Authors:

Giovanni Bechini Siemens Energy
Enzo Losi Universitá degli Studi di Ferrara
Lucrezia Manservigi Università degli Studi di Ferrara
Giovanni Pagliarini Universitá degli Studi di Ferrara
Guido Sciavicco Universitá degli Studi di Ferrara
Ionel Eduard Stan Universitá degli Studi di Ferrara
Mauro Venturini Universitá degli Studi di Ferrara