Data Clustering for the Identification of the Bifurcation Behaviour in Non-Linear Aeroelastic Systems using a Coupled Harmonic Balance/Genetic Algorithm Approach

摘要

This paper describes an efficient method for calculating the bifurcation behaviour of an aeroelastic system using a Harmonic Balance expansion coupled with a Genetic Algorithm, combined with a clustering algorithm in order to determine all the solutions at every single flight condition. It will be shown how it is possible to obtain all the bifurcation branches in one step. Two clustering algorithms, K-Means and PAM, together with a number of cluster index techniques, such as Davies-Boulding, Calinski-Harabasz are investigated. The method is applied to an aeroelastic galloping problem as this phenomenon presents a number of co-existing limit cycles at a range of airspeeds.