Fault diagnosis for the Space Shuttle main engine

Abstract

A conceptual design of a model-based fault detection and diagnosis system is developed for the Space Shuttle main engine. The design approach consists of process modeling, residual generation, and fault detection and diagnosis. The engine is modeled using a discrete time, quasilinear state-space representation. Model parameters are determined by identification. Residuals generated from the model are used by a neural network to detect and diagnose engine component faults. Fault diagnosis is accomplished by training the neural network to recognize the pattern of the respective fault signatures. Preliminary results for a failed valve, generated using a full, nonlinear simulation of the engine, are presented. These results indicate that the developed approach can be used for fault detection and diagnosis. The results also show that the developed model is an accurate and reliable predictor of the highly nonlinear and very complex engine.

Diverse neural net solutions to a fault diagnosis problem

“…The results reported here indicate the feasibility of applying neural nets to the monitoring of combustion condition, and as such, compliment the results of other neural net applications to fault diagnosis and condition monitoring [4][5][6][7]. Duyar and Merrill [4] report a system, based on simulated data, which uses two levels of neural nets for fault diagnosis of the Space Shuttle main engine.…”

“…[4][5][6][7]). The most interesting novel aspect of the present study is its reliance on the concept of diversity; a reliance which has interesting implications for neural net applications in general.…”

“…Duyar and Merrill [4] report a system, based on simulated data, which uses two levels of neural nets for fault diagnosis of the Space Shuttle main engine. A classifier level classifies faulty input data as two types of fault, and a second 'severity' level classifies the severity of the fault.…”

Fault diagnosis for a turbine engine

“…In Merrill (1985), Merrill, DeLaat, and Bruton (1988), and Merrill, DeLaat, and Abdelwahab (1991), the authors studied sensor failure detection for jet engines using a Kalman filter with a generalized likelihood ratio testing-based scheme. In Duyar and Merrill (1992) and Duyar, Eldem, Merrill, and Guo (1994) the authors derived linearized models of jet engine systems via the a-canonical form parameterization identification method and applied a parameter estimation approach in fault detection and isolation (FDI) for the space shuttle main engine. In Patton andChen (1992, 1997) and Patton, Chen, and Zhang (1997), the authors studied fault detection of jet engine sensor systems using an eigenstructure assignment technique to design observer-based residual generators, and they also studied its robustness.…”

Stable fault-tolerant adaptive fuzzy/neural control for a turbine engine

“…In [3]- [5] the authors studied sensor failure detection for jet engines using a Kalman filter with a generalized likelihood ratio testing based scheme. In [6], [7] the authors derived linearized models of jet engine systems via the -canonical form parameterization identification method and applied a parameter estimation approach in fault detection and isolation for the space shuttle main engine. In [8]- [10] the authors studied fault detection of jet engine sensor systems using an eigenstructure assignment technique to design observer based residual generators, and they also studied its robustness.…”