The Adaptive Control And Application Of A Class Non-Strict Feedback Stochastic System

In the practical engineering application,the stochastic perturbation not only affects the stability of the control system,but also greatly improves the design difficulty of the system controller.Therefore,the study of stochastic system control is of great significance.However,the current research results mainly focus on the nonlinear system with strict feedback or pure feedback form,which requires the research system to have a relatively strict lower-triangular structure.How to break through this framework to further promote the above research results to a more general control system,the relevant scientific research and engineering practice are of very important significance.This paper studies the adaptive control problem of a class of non-strict feedback stochastic nonlinear systems.And the research method was applied to the Marine search and rescue helicopter level stabilization system.The main contents of this paper include the following three parts:1.For a class of non-strict feedback nonlinear stochastic system,based on the lyapunov stability analysis theory,an adaptive neural network control method was proposed,using the function separation method and coordinate transformation method,to separate variables of the whole state function of the nonlinear system,using radial basis function neural network approximate the unknown nonlinear function in the system,ultimately ensures that all the signals in the closed-loop system in probability of bounded,but the large amount of calculation,unfavorable engineering implementation.2.An adaptive neural network control scheme based on dynamic surface control was proposed.In the process of controller design,the virtual control law is processed by the filter to overcome the circulation structure of the controller,using the Minimum Learning Parameter method to avoid the problem of "dimension explosion" at the same time,greatly reduces the computation burden of the controller.Simulation results demonstrate the proposed method.3.At last,the proposed algorithm is applied to the marine search and rescue helicopter.on the nonlinear dynamic model,this paper proposes a simple structure,small amount of calculation,good effect of adaptive neural network controller design method.the level stability control of the marine search and rescue helicopter.was realized through the numerical simulation test,and the simulation results verify the effectiveness of the proposed scheme.