Output details

Fault diagnosis of an industrial gas turbine prototype using a system identification approach

This paper presents a highly novel black box approach to Fault Detection and Isolation (FDI) requiring no physical system dynamic information. This study was funded by Alstom (Leicester) to investigate the potential of model-based FDI for a generic non-linear simulation of a single shaft gas turbine with control system, turbine faults and hidden dynamics. The black box model was developed with robust diagnostic algorithms to detect and isolate compressor and nozzle faults, solely using input-output measurement data. Analysis of the results from the system’s application showed excellent FDI performance. Work continues at Hull with Professor Tan (Beihang University).