Method for identifying fixed-order parameter model of aircraft based on modal segmentation and genetic algorithm

Abstract

The invention discloses a method for identifying a fixed-order parameter model of an aircraft based on modal segmentation and genetic algorithm, which belongs to the field of aircraft identification modeling. The invention is characterized in that the method comprises four stages, namely, model structure determination, flight data acquisition, model identification and model validation, wherein the stage of model structure determination is used for establishing a modal segmentation model of the aircraft, and comprises three steps, namely, kinetic analysis, model order determination and modal segmentation model determination; the stage of flight data acquisition is used for obtaining the frequency-domain response data of the aircraft, and comprises three steps, namely, sweep-frequency flight experiment, frequency domain transformation and data response of frequency domain; the stage of model identification is used for identifying the obtained dynamic model of the aircraft, and comprises a step of identifying model by using the genetic algorithm; and the stage of model validation is used for verifying the obtained dynamic model, and comprises a step of model validation, wherein, in the stage of model structure determination, the complicated high-order dynamic models are simplified by using the modal segmentation model, so that the precision of the model is not restricted by the model order in the process of identification while all unknown parameters are kept; in the stage of flight data acquisition, the true dynamic frequency response of the aircraft is obtained; in the stage of model identification, the modal segmentation model is upon the true dynamic frequency response to the greatest extent by using the genetic algorithm; and in the stage of model validation, the obtained dynamic model is inspected, in case of meeting the requirements, all identification models meet the requirements, otherwise, an experiment is carried out again.

Description

technical field [0001] The invention is an identification method for dynamic modeling of an aircraft, which can identify the dynamic model of an aircraft, especially a helicopter, with high precision. It is mainly used in technical fields such as aircraft identification modeling and control. Background technique [0002] The dynamic model of the aircraft is the premise of flight control. Only by obtaining an accurate dynamic model can an excellent flight control effect be obtained. Conversely, many advanced flight control algorithms cannot even be implemented without precise dynamic model accuracy. [0003] There are mainly three traditional aircraft modeling methods, wind tunnel modeling, mechanism modeling and identification modeling. However, for aircraft, especially helicopters, its structure, flow field and flight principle are very complicated. It is difficult to accurately quantify the turbulent flow field on the surface and the complex dynamic and control mechani...

AI Technical Summary

Problems solved by technology

In addition, the least squares, Levy, and numerical optimization methods used in traditional identification methods are also difficult to model aircraft dynamics with multiple parameters, multiple ranges, and high precision.

Based on the above factors, traditional identification methods (including CIFER ) has limited accuracy in modeling the dynamics of aircraft, especially helicopters

Images