论文信息 - On-Line Aircraft Engine Diagnostic Using a Soft-Constrained Kalman Filter

On-Line Aircraft Engine Diagnostic Using a Soft-Constrained Kalman Filter

The purpose of this contribution is to apply ridge regression to Kalman filtering in order to stabilize a health parameter identification under low or negative redundancy. The resulting algorithm achieves a so-called soft-constrained recursive health parameter identification, i.e. constraints are applied to parameters in a statistical way, contrary to hard-constrained algorithms based on strong equality or inequality constrains. The method is tested on data generated by a steady state turbofan engine model and representing typical component faults. The benefits that can be realized in terms of stability and accuracy are highlighted and some limits of the method are also mentioned.Copyright © 2004 by ASME

[1] Anastassios G. Stamatis,et al. Real Time Engine Model Implementation for Adaptive Control and Performance Monitoring of Large Civil Turbofans , 2001 .

[2] Marc Grodent,et al. Engine physical diagnosis using a robust parameter estimation method , 2001 .

[3] S. Haykin. Kalman Filtering and Neural Networks , 2001 .

[4] N. Aretakis,et al. Non-Linear Engine Component Fault Diagnosis From a Limited Number of Measurements Using a Combinatorial Approach , 2002 .

[5] Eric A. Wan,et al. Nonlinear estimation and modeling of noisy time series by dual kalman filtering methods , 2000 .

[6] John E. Moody,et al. The Effective Number of Parameters: An Analysis of Generalization and Regularization in Nonlinear Learning Systems , 1991, NIPS.

[7] Donald L. Simon,et al. Application of a Bank of Kalman Filters for Aircraft Engine Fault Diagnostics , 2003 .

[8] Mahesan Niranjan,et al. Hierarchical Bayesian-Kalman models for regularisation and ARD in sequential learning , 1997 .

[9] A. N. Tikhonov,et al. Solutions of ill-posed problems , 1977 .

[10] Donald L. Simon,et al. Aircraft Turbofan Engine Health Estimation Using Constrained Kalman Filtering , 2005 .