PID control of Gaussian distribution input in MATLAB

A random array of matlab

s=1:1:500;
in = 0.1*randn (1,500) +1;
Plot (s,in, ' * ');

hist (in,20);

Second, PID control

Online Source:

Clear All;close ALL;TS=0.001;SYS=TF (5.235e005,[1,87.35,1.047e004,0]);% build transfer function dsys=c2d (sys,ts, ' z ');% The continuous time model is converted into a discrete time model, the sampling time is ts=0.001[num,den]=tfdata (Dsys, ' V '), and the numerator denominator u_1=0.0;u_2=0.0;u_3=0.0;y_1=0.0;y_2=0.0 is obtained after the dispersion; y_3=0.0;x=[0,0,0] '; error_1=0;for k=1:1:500time (k) =k*ts;rin (k) = 1; The% input is (0.8,1.2) the random number of the kp=0.6;ki=0.001;kd=0.001 distribution; The p,i,d parameter of the% setting is U (k) =kp*x (1) +kd*x (2) +ki*x (3); %pid Controller%linear Model Linear models yout (k) =-den (2) *y_1-den (3) *y_2-den (4) *y_3+num (2) *u_1+num (3) *u_2+num (4) *u_3;error (k) =rin (k)-yout (k);%return of Parametersu_3=u_2;u_2=u_1;u_1=u (k); Y_3=y_2;y_2=y_1;y_1=yout (k); X (1) =error (k); %calculating Px (2) = (Error (k)-error_1)/ts; %calculating x (3) =x (3) +error (k) *ts; %calculating Ixi (k) =x (3); Error_1=error (k); Endfigure (1);p lot (Time,rin, ' B ', time,yout, ' R '); Xlabel (' Time (s) '); Ylabel (' Rin,yout ')

　　

Three, Gaussian distribution of input

%pid controlerclear All;close all;s=1:1:500;in = 0.1*randn (1,500) +1; % of 500 (0.95,1.05) of the random number of the distribution of the plot (S,in, ' * '); grid ONTS=0.001;SYS=TF (5.235e005,[1,87.35,1.047e004,0]);% Build transfer function dsys= C2D (sys,ts, ' z ');% converts a continuous time model into a discrete time model with a sampling time of ts=0.001[num,den]=tfdata (Dsys, ' V '), and a percentage of the numerator denominator u_1=0.0;u_2=0.0;u_3=0.0 ; y_1=0.0;y_2=0.0;y_3=0.0;x=[0,0,0] '; error_1=0;for k=1:1:500time (k) =k*ts;rin (k) = in (k); The% input is (0.95,1.05) the random number of the kp=0.6;ki=0.001;kd=0.001 distribution; The p,i,d parameter of the% setting is U (k) =kp*x (1) +kd*x (2) +ki*x (3); %pid Controller%linear Model Linear models yout (k) =-den (2) *y_1-den (3) *y_2-den (4) *y_3+num (2) *u_1+num (3) *u_2+num (4) *u_3;error (k) =rin (k)-yout (k);%return of Parametersu_3=u_2;u_2=u_1;u_1=u (k); Y_3=y_2;y_2=y_1;y_1=yout (k); X (1) =error (k); %calculating Px (2) = (Error (k)-error_1)/ts; %calculating x (3) =x (3) +error (k) *ts; %calculating Ixi (k) =x (3); Error_1=error (k); Endfigure (1);p lot (Time,rin, ' B ', time,yout, ' R '); Xlabel (' Time (s) '); Ylabel (' rin,yout ');

　　

The effect is good, but also very happy. Next time, do a self-tuning control system based on annealing

PID control of Gaussian distribution input in MATLAB