Modeling, Diagnostics and Prognostics of a Two-Spool Turbofan Engine

摘要

Model-based diagnostic/prognostic techniques have the potential to predict, within reasonable bounds, the remaining useful life of critical system components. Due to the numerous uncertainties in the operation of a turbine engine and unavailability of accurate engine models, prognostics continue to pose a significant challenge. There is a need to develop an engine prognostic approach that can accommodate different damage modes, sensor failures, material properties, dynamic load histories and damage accumulation. Using an accurate physics-based model of the engine one can develop such a prognostic approach. We present a nonlinear dynamical model of a two-spool turbine engine developed from first principles. The simulation model has been implemented using MATLAB/Simulink. It is used with the Kalman Filter-based diagnostic technique previously discussed in literature to detect and isolate sensor faults. A literature review of the developments in the area of prognostics is also presented, along with the problems and challenges.