Abstract: The model-based and data-driven-based (DDB) aero-engine component fault fusion diagnosis method is present-ed in this paper. The extreme learning machine (ELM) is used to DDB component fault diagnosis. However, the ELM input weights and hidden layer biases are generated randomly, which usually leads to more nodes in hidden layer and poor generalization ability. To overcome these drawbacks of ELM, the improved differential evolution (IDE) is used to optimize the input weights and the hidden layer biases. The optimization is carried out according to the root mean squared error (RMSE) of validation set and the norm of the output weights. Meantime, SVD-based Reduced-dimensional Kalman filters are used to estimate the health parameters which can solve the problem of limited number of sensors in model based diagnosis method. To improve the accuracy of fault diagnosis for aero-engine component, the prediction results of health parameters based on both methods are fused by Improved Re-cursive Reduced Least Squares Support Vector Regression (IRR-LSSVR). Simulation results show that compared with either the model-based or DDB approaches, the proposed fusion method improves the accuracy of fault diag-nosis significantly.

Key words: aeroengine, extreme learning machine, component fault, differential evolution, Kalman filter, information fusion