Research on WNN aerodynamic modeling from flight data based on improved PSO algorithm

To depict the aerodynamic characteristics of flight vehicle accurately, a Wavelet Neural Network (WNN) method, based on improved Particle Swarm Optimization (IPSO) algorithm, is proposed for aerodynamic modeling from flight data. First the multi-particle information share strategy and mutation operation are introduced into Simple PSO algorithm in order to improve the modeling capability of WNN, and then according to modeling flow the aerodynamic model from flight data for flight vehicles is established by WNN based on IPSO algorithm. Simulation results show that the method proposed has a good capability with features of precision, convergence and surmounting prematurity or local optimum, and is also effective and feasible for aerodynamic modeling from flight data.

► An improved PSO algorithm sharing multi-particle information with mutation operator is proposed.
► A WNN modeling method based on improved PSO algorithm, namely IPSO–WNN, is given.
► IPSO–WNN is used to establish the aerodynamic model for flight vehicles from flight data.
► For aerodynamic modeling, IPSO–WNN shows a good capability on precision, convergence and so on.