Abstract A research activity has been initiated to study the development of a diagnostic methodology, for the optimization of energy efficiency and the maximization of the operational time in those conditions, based on artificial intelligence (AI) techniques such as artificial neural network (ANN) and fuzzy logic. The diagnostic procedure, developed specifically for the cogeneration plant located at the Engineering Department of the University of Perugia, must be characterized by a modular architecture to obtain a flexible architecture applicable to different systems. The first part of the study deals with the identifying the principal modules and the corresponding variables necessary to evaluate the module “health state”. Also the consequent upgrade of the monitoring system is described in this paper. Moreover it describes the structure proposed for the diagnostic procedure, consisting of a procedure for measurement validation and a fuzzy logic-based inference system. The first reveals the presence of abnormal conditions and localizes their source distinguishing between system failure and instrumentation malfunctions. The second provides an evaluation of module health state and the classification of the failures which have possibly occurred. The procedure was implemented in C++.
[1]
Linda Barelli,et al.
Thermodynamic Diagnosis Of A Small Scale Gas Turbine With Neural Networks
,
2001
.
[2]
Paul M. Frank,et al.
Fault diagnosis in dynamic systems: theory and application
,
1989
.
[3]
Hans R. Depold,et al.
The Application of Expert Systems and Neural Networks to Gas Turbine Prognostics and Diagnostics
,
1998
.
[4]
R. N. Clark,et al.
Instrument Fault Detection in a Pressurized Water Reactor Pressurizer
,
1982
.
[5]
Ehsan Mesbahi.
An intelligent sensor validation and fault diagnostic technique for marine diesel engines
,
2001
.
[6]
Ndaona Chokani,et al.
Engine monitoring using neural networks
,
1998
.
[7]
R. Singh,et al.
Fault diagnosis of a turbofan engine using neural networks - A quantitative approach
,
1998
.