A Novel Data-Driven Approach to Parameter Identification for Flight Simulation

At present, the empirical formulas of flight parameters are not accurately fitted for different aircraft types, and the traditional black-box algorithms don't express the flight parameter correlation quantitatively. To address the above problems, a flight simulation parameter identification method bases on Quick Access Recorder (QAR) data is proposed. Based on the polynomial fitting algorithm, the method adds weight coefficients to assign the weight of input parameters in the algorithm and calculates the weight coefficients of different input parameters using Spearman Rank correlation coefficients. Designing the binary polynomial and solving the optimal polynomial coefficients based on the improved algorithm. The simulation results show that the method effectively reduce the simulation error.