Observers are usually used in model-based FDI system for aero engines, especially for control system sensors. However, when a bank of traditional observers is used to detect gas path faults, it seems difficult to locate or isolate them, which is not a problem for sensor FDI. As a solution, a modified design method for a kind of robust observer called UIO (Unknown Input Observer) is proposed in this paper. An augmented linear model of a big turbofan is built at first. The degradations of compressor and HP turbine efficiencies are integrated as model inputs. Then some functions reflecting the FDI performance of UIO for these degradations can be obtained. DE (Differential Evolution) algorithm is introduced to solve these equations, whose best solution means the residual of observer is only sensitive for the specified gas path fault and not effected by other faults and unknown inputs. Some simulation results for UIOs designed by modified method are given to verify their FDI performance for compressor and HP turbine efficiencies.
- International Gas Turbine Institute
Model-Based Fault Diagnosis for Performance Degradations of Turbofan Gas Path via Optimal Robust Residuals
- Views Icon Views
- Share Icon Share
- Search Site
Yang, C, Kong, X, & Wang, X. "Model-Based Fault Diagnosis for Performance Degradations of Turbofan Gas Path via Optimal Robust Residuals." Proceedings of the ASME Turbo Expo 2016: Turbomachinery Technical Conference and Exposition. Volume 6: Ceramics; Controls, Diagnostics and Instrumentation; Education; Manufacturing Materials and Metallurgy. Seoul, South Korea. June 13–17, 2016. V006T05A004. ASME. https://doi.org/10.1115/GT2016-56300
Download citation file: