A Neural Network-Based Intelligent Control Method of Time-Scale Separation Aircraft Elastic Body

Abstract

The invention discloses a timescale-separated aircraft elastomer intelligent control method based on a neural network and belongs to the field of aircraft control. The method is used for solving the problem that existing elastic hypersonic aircraft control methods cannot achieve high-precision control under the influences of uncertain information and elastic vibration. According to the method, firstly, based on the capability of solving a multi-timescale problem through a singular perturbation theory, fast-slow timescale decomposition is conducted on a hypersonic aircraft to separate a rigid modal from a soft modal; secondly, control algorithm design is conducted on the rigid modal and the soft model separately, uncertainty existing in a slowly varying timescale part representing the rigidmodal of a system is taken into account, the neural network is adopted for online study, then a construction weight updating rule of a Lyapunov method is adopted for ensuring the stability of the system, an intelligent control strategy is designed, and a sliding-mode control method is designed on the basis of a fast varying timescale part representing an elastic modal of the system; finally, thetwo control input modes are combined into a whole part to serve as a general rudder deviation instruction to achieve effective tracking of an altitude instruction, and meanwhile effective suppressionof the elastic modal is achieved.

Description

technical field [0001] The invention relates to a hypersonic vehicle elastic body intelligent control technology, in particular to a neural network-based time scale separation hypersonic vehicle elastic body intelligent control strategy, which belongs to the field of aircraft control. Background technique [0002] Hypersonic vehicles are developing towards lighter materials, and the influence of elastic deformation increases accordingly, making it easy for sensors to capture elastic deformation information and further transmit this information to the control system. If the control system lacks a corresponding elastic mode control mechanism, This information will be processed as pseudo-attitude information, which will reduce the control accuracy of the system, which is not conducive to the safety of the aircraft. The singular perturbation theory is used to separate the time scale of the hypersonic vehicle dynamics model, and the control algorithm is designed for the elastic m...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem of the influence of uncertainty and elastic vibration on the control of high precision during the flight of a hypersonic vehicle, the present invention proposes a neural network-based intelligent control strategy for time scale separation of hypersonic vehicle elastic body

Images