Application of Kalman Filters for the Fault Diagnoses of Aircraft Engine

Abstract

Fault detection and isolation (FDI) logic plays a crucial role in enhancing the safety and reliability, and reducing the operating cost of aircraft propulsion systems. However, it is a challenging problem achieving the FDI task with high reliability. For this purpose, various approaches have been proposed in the literature. In an on-line engine fault diagnoses, two tasks may use Kalman filter to carry out: 1) evaluation of on-line engine state variables to renew the on-board model; 2) diagnoses of online aircraft engine sensor/actuator fault. How to solve the above problems through application of Kalman filter is discussed in this paper. A challenge in developing an on-line fault detection algorithm is making it adaptive to engine health degradation. If the algorithm has no adaptation capability, it will eventually lose its diagnostic effectiveness. To address this problem, the integration of on-line diagnostic algorithms was investigated. The Kalman filter estimates engine health condition over the course of engine’s life. Then the on-board model could be re-constructer based on the estimated values of Kalman filter. After all of the above, A Robust Kalman filter and a bank of Kalman filters are applied in fault detection and isolation (FDI) of sensor and actuator for aircraft gas turbine engine. A bank of Kalman filters are used to detect and isolate sensor fault, each of Kalman filter is designed based on a specific hypothesis for detecting a specific sensor fault. In the event that a fault does occur, all filters except the one using the correct hypothesis will produce large estimation errors, from which a specific fault is isolated. When the Kalman filter is used, failures in the sensors and actuators affect the characteristics of the residual signals of the Kalman filter. While a Robust Kalman filter is used, the decision statistics changes regardless the faults in the sensors or in the actuators, because it is sensitive to sensor fault and insensitive to actuator fault. W. C. Merrill, J. C. Delaat, and W. M. Bruton used a bank of Kalman filters for aircraft engine sensor FDI. This study successfully improved control loop tolerance to sensor failures, which were considered the most likely engine failures to happen under the harsh operating environment. In this study, actuator failure was not considered. In the study done by T. Kobayashi and D. Simon, a fault detection and isolation (FDI) system which utilizes a bank of Kalman filters is developed for aircraft engine sensor and actuator FDI in conjunction with the detection of component faults. The results indicate that the proposed Source: Kalman Filter, Book edited by: Vedran Kordic, ISBN 978-953-307-094-0, pp. 390, May 2010, INTECH, Croatia, downloaded from SCIYO.COM