Model-Predictive-Control/scripts/dmc/pulse.m at main · MRGilak/Model-Predictive-Control · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
run(fullfile(fileparts(fileparts(fileparts(mfilename('fullpath')))), 'initDemo.m'));
clear; close all; clc;

%% Define parameters
params = struct();
params.tf = 100;
params.Ts = 0.2;
params.N = params.tf / params.Ts;
params.u0_nonlinear = -0.897;
params.u0_linear = -1.7863;

bias = -1;
span = 1;

%% Define input
levels = bias * ones(1, 5) + span * [0, 0.95, 0, -0.95, 0];
spans = 1.5 * [8, 12, 12, 12, 12];

span_steps = round(spans ./ params.Ts);
ref_signal = [];
for i = 1:length(levels)
    ref_signal = [ref_signal; repmat(levels(i), span_steps(i), 1)];
end
params.Ref = [ref_signal; repmat(levels(end), max(0, params.N - length(ref_signal)), 1)];

%% Define parameters
params.P = 11;
params.M = 8;
params.alpha = 0.5;
params.Q = 1 * ones(params.P, 1);
params.R = 0.3107 * ones(params.M, 1);
params.N_model = 55;
params.disturbance_amp = 0;
params.noise_power = -200000;
params.disturbance_start_time = 50;
params.x0 = [-1; 0];
params.is_programmed = true;
params.is_open_loop = false;

%% Run simulations
[results_lin, sr_lin] = simulate_linear_system(params);
[results_nonlin, ~] = simulate_nonlinear_system(params);

Hankel = results_lin.F;
Toeplitz = results_lin.G;
gainMatrix = results_lin.gainMatrix;
sr_all = results_lin.sr_all;

%% Plot results
plot_simulation_results(results_lin, results_nonlin, params, false);
% plot_step_response(params, sr_all);
% plot_static_gain(-1);