mirror of
https://github.com/ohmyjesus/RBF_NeuralNetwork.git
synced 2026-02-05 11:09:47 +08:00
89 lines
2.4 KiB
Matlab
89 lines
2.4 KiB
Matlab
function [sys,x0,str,ts] = Book4341_Controller(t,x,u,flag)
|
||
% 以下程序是 基于RBF神经网络的直接鲁棒自适应控制
|
||
switch flag
|
||
case 0
|
||
[sys,x0,str,ts]=mdlInitializeSizes;
|
||
case 1
|
||
sys=mdlDerivatives(t,x,u);
|
||
case {2,4,9}
|
||
sys=[];
|
||
case 3
|
||
sys=mdlOutputs(t,x,u);
|
||
otherwise
|
||
DAStudio.error('Simulink:blocks:unhandledFlag', num2str(flag));
|
||
end
|
||
|
||
function [sys,x0,str,ts]=mdlInitializeSizes %系统的初始化
|
||
sizes = simsizes;
|
||
sizes.NumContStates = 0; %设置系统连续状态的变量
|
||
sizes.NumDiscStates = 0; %设置系统离散状态的变量
|
||
sizes.NumOutputs = 1; %设置系统输出的变量
|
||
sizes.NumInputs = 4; %设置系统输入的变量
|
||
sizes.DirFeedthrough = 1; %如果在输出方程中显含输入变量u，则应该将本参数设置为1
|
||
sizes.NumSampleTimes = 0; % 模块采样周期的个数
|
||
% 需要的样本时间，一般为1.
|
||
% 猜测为如果为n，则下一时刻的状态需要知道前n个状态的系统状态
|
||
sys = simsizes(sizes);
|
||
x0 = []; % 系统初始状态变量
|
||
str = []; % 保留变量，保持为空
|
||
ts = []; % 采样时间[t1 t2] t1为采样周期，如果取t1=-1则将继承输入信号的采样周期；参数t2为偏移量，一般取为0
|
||
global W
|
||
% 神经网络采用5-9-1结构 IN = 5 MID = 9 OUT = 1
|
||
% 初始权值
|
||
W = [0 0 0 0 0 0 0 0 0]' ; %MID * OUT矩阵 9*1
|
||
|
||
|
||
function sys = mdlOutputs(t,x,u) %产生（传递）系统输出
|
||
global W
|
||
% 神经网络采用5-9-1结构
|
||
b = 20; % 高斯函数的基宽 维度MID * 1 1*1 b的选择很重要 b越大 网路对输入的映射能力越大
|
||
c = [-2 -1.5 -1 -0.5 0 0.5 1 1.5 2;
|
||
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2;
|
||
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2;
|
||
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2;
|
||
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2]; % 高斯函数的中心点矢量 维度 IN * MID 5*9
|
||
% 仿真中应根据网络输入值的有效映射范围来设计 c和b 从而保证有效的高斯映射 不合适的b或c均会导致结果不正确
|
||
IN = 5;
|
||
Mid = 9;
|
||
Out = 1;
|
||
|
||
lambda = 5;
|
||
ita = 500 * eye(9);
|
||
xite = 0.005;
|
||
If = 0.25;
|
||
|
||
e = -u(1); % e = x - xd; 实际-期望
|
||
de = -u(2);
|
||
x_1 = u(3);
|
||
x_2 = u(4);
|
||
s = lambda * e + de;
|
||
s_if = s/If;
|
||
|
||
yd = sin(t);
|
||
dyd = cos(t);
|
||
ddyd = -sin(t);
|
||
|
||
v = -ddyd + lambda * de;
|
||
|
||
Input = [x_1; x_2; s; s_if ; v];
|
||
h = zeros(Mid , 1); %9*1矩阵
|
||
for i =1:Mid
|
||
h(i) = exp(-(norm(Input - c(:,i))^2) / (2*b^2));
|
||
end
|
||
% 控制率ut
|
||
belta = 150;
|
||
ut = 1/belta * W' * h;
|
||
sys(1) = ut;
|
||
|
||
% W权值的更新
|
||
dw = -ita * (h * s + xite * W); % 9*9 * (9*1 + 9*1)
|
||
dt = 0.001; % 仿真步长
|
||
W = W + dw * dt; % W的自适应律
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|