Hypersonic aircraft self-adaptation time-varying preset performance control method based on LS-SVM

Abstract

The invention discloses a hypersonic aircraft self-adaptation time-varying preset performance control method based on LS-SVM. The method includes the steps of firstly, designing LS-SVM identifier and a time-varying preset performance function according to the longitudinal model of an air-breathing hypersonic aircraft to obtain an error transformation model; secondly, designing a speed subsystem self-adaptation time-varying preset performance controller according to the error transformation model; thirdly, designing an altitude subsystem self-adaptation time-varying preset performance controller according to the error transformation model; fourthly, performing stabilizing analysis and performance analysis according to the speed subsystem self-adaptation time-varying preset performance controller and the altitude subsystem self-adaptation time-varying preset performance controller. The hypersonic aircraft self-adaptation time-varying preset performance control method has the advantages that the online approximation algorithm complexity of an unknown nonlinear power model is lowered while the saturation constraint conditions of an aircraft actuating mechanism are not violated; the self-adaptation high-precision control of a hypersonic aircraft is achieved while the stability of the control system of the aircraft is guaranteed.

Description

technical field [0001] The invention relates to aircraft adaptive control technology, in particular to an LS-SVM-based hypersonic aircraft adaptive time-varying preset performance control method. Background technique [0002] For the adaptive control of hypersonic vehicles with unknown nonlinear dynamics, the existing control methods basically use the following two steps to design related control systems [1],[2] : [0003] A. With the help of the neural network's good approximation characteristics for any nonlinearity, first estimate the unknown nonlinear dynamic model of the hypersonic vehicle; [0004] B. Non-linear dynamic model based on identification, using regressive control [3] , sliding mode control [4] Design the corresponding adaptive control method to realize the adaptive control of the hypersonic vehicle. [0005] Although the existing control methods can achieve adaptive robust control of hypersonic vehicles, the adaptive control method based on neural netwo...

AI Technical Summary

Problems solved by technology

[0018] The present invention aims at the disadvantages of the traditional hypersonic vehicle adaptive control method that the unknown nonlinear model identification/approximation algorithm has high complexity and non-optimum, and cannot guarantee the (transient/steady state) performance of the whole flight process, and provides a method based on The LS-SVM

Images