Fault detection and diagnosis algorithms for transient state of an open-cycle liquid rocket engine using nonlinear Kalman filter methods. (October 2019)

• Record Type: Journal Article
• Authors: Cha, Jihyoung; Ko, Sangho; Park, Soon-Young; Jeong, Eunhwan
• Is Part Of: Acta astronautica. Volume 163(2019)Part A
• Page Start: 147
• Page End: 156
• Publication Date: 2019-10
• Subjects: Open-cycle liquid propellent rocket engine -- Fault detection and diagnosis -- Extended Kalman filter -- Unscented Kalman filter -- Multiple-model method -- Numerical simulation; Astronautics -- Periodicals; Outer space -- Exploration -- Periodicals; Astronautics; Periodicals; 629.405
• DOI: 10.1016/j.actaastro.2019.03.075
• Languages: English
• ISSNs: 0094-5765

Abstract

Abstract: This paper deals with an application of the fault detection and diagnosis algorithm based on nonlinear Kalman filter methods for transient state of an open-cycle liquid propellant rocket engine. In order to develop the algorithm, we designed two types of the nonlinear Kalman filter which are the extended Kalman filter and unscented Kalman filter with non-linear model of the liquid propellant rocket engine. Then using the measurement data of some important parameters of the engine, the residuals from the nonlinear Kalman filters are obtained. Using the residuals, we can detect and diagnose faults in the components of the engine by using the multiple model method. To confirm the fault detection and diagnosis algorithm, we developed mathematical model of an open-cycle liquid propellant rocket engine and artificially injected various faults such as decreasing turbopump efficiency. And then, perform the fault detection and diagnosis algorithm and check the performance of the algorithm. This process is numerically demonstrated for the open-cycle liquid propellant rocket engine under start-up process by using the simulated measurement data from the mathematical model of the engine. Highlights: A mathematical model for the transient state of an open-cycle liquid propellant rocket engine (OPLPRE) is developed. Two variations of the nonlinear Kalman filter are used for fault detection and diagnosis (FDD) algorithm of the OPLPRE. The FDD algorithms can be applied to the transient state of the OPLPRE. The performances of the FDD algorithms are demonstrated by numerical simulations.