Longitudinal aerodynamic coefficients modeling from simulated flight data of fixed wing aircraft based on three methods

The equation error method, output error method, and neural network method are used to model the aircraft longitudinal aerodynamic coefficients from simulated flight data. The equation error method and the output error method are based on postulated mathematical models of the longitudinal aerodynamic coefficients, and the least squared principle and the maximum likelihood principle are used to estimate the model parameters respectively. The neural network method uses the BP network output to fit the aerodynamic coefficients without assumptions of mathematical model of aerodynamic coefficients. Two sets of simulation tests were conducted, one set of data is used for model identification, and the other set is used for model validation.