Aerodynamic Parameter Identification of Aircraft Based on GA-BP Algorithm

Aiming at the problem of aerodynamic parameter identification of a certain type aircraft, the GA-BP algorithm is used to identify the coefficients of aerodynamic force. Firstly, the coefficients are solved by using flight trajectory data obtained from wind tunnel simulation experiments. Then, the relationships between aerodynamic force coefficients and Mach, angle of attack, rudder pendulum angle, etc. are fitted by BP neural network. The thresholds and weights between neurons are optimized by genetic algorithm during the training of neural network, so that the optimal structure of neural network can be obtained through training of multiple sets of flight trajectory data under different flight conditions. Simulation results show that the aerodynamic force coefficients obtained by the GA-BP algorithm have higher accuracy compared with traditionally used engineering methods.