Aero-Engine Sensor Fault Diagnosis Based On Kalman.pdf

AERO-ENGINE SENSOR FAULT DIAGNOSIS BASED ON KALMAN FILTER Yuying Guo1,2, Bin Jiang1, Youmin Zhang3,4, Zhengwei Zhu2 1 College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China E-mail: binjiang@ 2 College of Information Engineering, Southwest University of Science and Technology, Mianyang 621010, China E-mail: polarisguo@163.com 3 Dept. of Computer Science and Engineering, Aalborg University Esbjerg, Niels Bohrs Vej 8, 6700 Esbjerg, Denmark Email: ymzhang@cs.aaue.dk 4 Dept. of Mechanical Industrial Engineering, Concordia University, Montreal, Quebec H3G 1M8, Canada Email: ymzhang@encs.concordia.ca Abstract: In this paper, an aero-engine augmented state variable model and a Kalman filter based method are developed for the purpose of aero-engine sensor fault diagnosis. A bank of Kalman filters method is proposed to detect and isolate the sensor faults, and an adaptive model is further reconstructed based on the remaining non-faulty sensor information. Simulation results show the effectiveness of the proposed method. Keywords: Aero-engine system, a bank of Kalman filters, fault diagnosis 1. INTRODUTION Due to the increasing demand for high performance and reliability in modern aero-engine systems, control and monitoring of such systems have become a challenging issue and become also more complex due to the multi-loop, multi-variable and highly non-linear nature of the system. At the same time, the engine usually runs under the high-temperature, high-pressure conditions, sensor faults could occur, which makes the measured parameter inaccurate and may lead to severe deterioration of system control performance. Therefore, there is a strong need for sensor fault detection and diagnosis (FDD) and provide correct estimation for engine states and parameters under fault conditions. For developing an effective FDD scheme to aero-engine systems, a state