Identification model of aeroengine based on improved LS-SVM

Because of aeroengine properties such as the strong nonlinearity and time-varying uncertainty, a new identification algorithm of aeroengine model based on improved LS-SVM was brought forward. In the method, LS-SVM robustness was improved by adding weighed values to errors and its sparseness was improved by clipping algorithm. In terms of the recorded flight data on some turbofan engine, improved LS-SVM identification model of aeroengine was set up. Through the identification of the recorded flight data, the results show that the improved LS-SVM identification model has the advantages of high identification precision, good self-adaptability and strong robustness. It is effective that the improved LS-SVM identification model is used in aeroengine.