plotting the bifurcation diagram of Duffing oscillations in the (x1, omega) plane?

6 Ansichten (letzte 30 Tage)
function dx = duffing(t,x);
global alpha delta beta F omega
dx=[x(2);
-delta*x(2)-alpha*x(1)-beta*x(1)^3+F*cos(x(3));
omega];
% Define parameters
global alpha delta beta F omega
alpha=0.1;
delta=0.2;
beta=0.3;
F=0.5;
omega_range = 0.8:0.1:1; % Range of omega values
x0 = [0.1, 0.1, 0.1]; % Initial conditions
% Set simulation step
dt = 0.001;
% Presave a matrix of NaNs to save points of intersection
M = NaN * zeros(1000, length(omega_range));
pos = 0; % Indexing dummy variable
% Set ode options
options = odeset('RelTol', 1e-5, 'AbsTol', 1e-5);
for omega = omega_range
omega % Print omega, just to track progress
pos = pos + 1; % Increase index
% Simulate the system
[t, x] = ode45(@duffing, 0:dt:1000, x0, options);
% Discard transient
index = t > 400;
X = x(index, :); % Save states without transient in a new vector
l = length(X);
% Save the final x(1) value
M(1:l, pos) = X(:, 1);
end
% Plot the bifurcation diagram
figure;
plot(omega_range, M, '.b', 'MarkerSize', 2);
xlabel('\omega');
ylabel('x_1');
title('Bifurcation Diagram of Duffing System (x_1 vs. \omega)');
grid on;
Can anyone help troubleshoot my code for plotting the bifurcation diagram of Duffing oscillations in the (x1, omega) plane? It's not functioning as expected. Thanks!

Antworten (1)

Athanasios Paraskevopoulos
Bearbeitet: Athanasios Paraskevopoulos am 16 Mai 2024
function dx = duffing(t, x)
global alpha delta beta F omega
dx = [x(2);
-delta * x(2) - alpha * x(1) - beta * x(1)^3 + F * cos(omega * t);
omega];
end
% Define parameters
global alpha delta beta F omega
alpha = 0.1;
delta = 0.2;
beta = 0.3;
F = 0.5;
omega_range = 0.8:0.01:1.0; % Finer range of omega values
x0 = [0.1, 0.1, 0.1]; % Initial conditions
% Set simulation step and time
dt = 0.01;
simulation_time = 1000; % Total simulation time
transient_time = 400; % Time to discard transient
% Preallocate matrix for results
num_points = floor((simulation_time - transient_time) / dt) + 1;
M = NaN(num_points, length(omega_range));
% Set ODE options
options = odeset('RelTol', 1e-5, 'AbsTol', 1e-5);
% Loop over omega values
for pos = 1:length(omega_range)
omega = omega_range(pos);
fprintf('Processing omega = %.2f\n', omega); % Track progress
% Simulate the system
[t, x] = ode45(@duffing, 0:dt:simulation_time, x0, options);
% Remove transients
index = t > transient_time;
X = x(index, :);
% Save the x(1) values after transient
M(1:length(X), pos) = X(:, 1);
end
Processing omega = 0.80 Processing omega = 0.81 Processing omega = 0.82 Processing omega = 0.83 Processing omega = 0.84 Processing omega = 0.85 Processing omega = 0.86 Processing omega = 0.87 Processing omega = 0.88 Processing omega = 0.89 Processing omega = 0.90 Processing omega = 0.91 Processing omega = 0.92 Processing omega = 0.93 Processing omega = 0.94 Processing omega = 0.95 Processing omega = 0.96 Processing omega = 0.97 Processing omega = 0.98 Processing omega = 0.99 Processing omega = 1.00
% Plot the bifurcation diagram
figure;
hold on;
for pos = 1:length(omega_range)
omega_vals = omega_range(pos) * ones(sum(~isnan(M(:, pos))), 1);
plot(omega_vals, M(~isnan(M(:, pos)), pos), '.b', 'MarkerSize', 2);
end
xlabel('\omega');
ylabel('x_1');
title('Bifurcation Diagram of Duffing System (x_1 vs. \omega)');
grid on;
hold off;

Produkte


Version

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by