Lotka Volterra with an additional parameter

Question

Khushi Patel 2020-9-25

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/599620-lotka-volterra-with-an-additional-parameter

评论： Star Strider 2020-9-25

I have attempted to model the equations provided below. the code keeps giving me an error when I try to plot the figure using the parameter k instead of t. I am trying to model the effect of varying k on the population dynamics for both prey and predator.

note: if I use plot(k, P) it gives me a complicated and entirely different graph that is unreadable.

Script Window:

tspan = [0, 400];

P0 = [200, 50];

k = [-1, 1];

[t,P] = ode45(@(t,P)lotkavolterra(t,P),tspan,P0);

plot(k,P(:,1),k,P(:,2)) %This section gives me the error. It works, however, if I use t instead of k.

function [dPdt] = lotkawithk(t,P)

alpha = 1;

beta = 0.05;

gamma = 0.05;

delta = 0.0025;

x = P(1);

y = P(2);

dPdt = zeroes(2,1);

dPdt(1) = alpha*x - (beta*x*y)/((log(exp(1)+t))^k);

dPdt(2) = (delta*x*y)/((log(exp(1)+t))^k) - (gamma*y);

end

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Star Strider 2020-9-25

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/599620-lotka-volterra-with-an-additional-parameter#answer_500341

在 MATLAB Online 中打开

I cannot imagine how you got that code to work at all, considering that the function you call in ode45 is not the function you posted!

Try this:

tspan = linspace(0, 400, 100);;
P0 = [200, 50];
k = [-1, 1];
for k1 = 1:numel(k)
    [t,P] = ode45(@(t,P)lotkawithk(t,P,k(k1)),tspan,P0);
    Pm{k1} = P;
end
figure
hold on
for k1 = 1:numel(k)
    plot(t,Pm{k1}(:,1), t,Pm{k1}(:,2), 'DisplayName',sprintf('k = %d',k(k1)))  %This section gives me the error. It works, however, if I use t instead of k.
end
grid
legend
    function [dPdt] = lotkawithk(t,P,k)
        alpha = 1;
        beta = 0.05;
        gamma = 0.05;
        delta = 0.0025;
        x = P(1);
        y = P(2);
        dPdt = zeros(2,1);
        dPdt(1) = alpha*x - (beta*x*y)/((log(exp(1)+t))^k);
        dPdt(2) = (delta*x*y)/((log(exp(1)+t))^k) - (gamma*y);
    end

Note that ‘k’ is passed as a parameter.

Make appropriate changes to get different results.

6 个评论
显示 4更早的评论隐藏 4更早的评论

Star Strider 2020-9-25

在 MATLAB Online 中打开

My pleasure!

It is possible. However plotting a (100x1) vector dependent variables against a scalar independent variable produces uninterpretable results, so it’s not worth the effort:

figure
hold on
for k1 = 1:numel(k)
    plot(k(k1),Pm{k1}(:,1), '.', 'LineWidth',1.5, 'DisplayName',sprintf('Predator (k = %d)',k(k1)))  %This section gives me the error. It works, however, if I use t instead of k.
    plot(k(k1),Pm{k1}(:,2), '.', 'LineWidth',1.5, 'DisplayName',sprintf('Prey (k = %d)',k(k1)))
end
hold off
grid
xlim([-1.1  1.1])
legend('Location','N')

Note that this uses a marker to plot the points, because they are plotted as individual points, not lines. If you didn’t plot markers, that’s likely the reason that nothing showed up when you attempted to plot it.

Still another option is to plot them in 3D:

figure
hold on
for k1 = 1:numel(k)
    plot3(k(k1)*ones(size(t)),t,Pm{k1}(:,1), 'LineWidth',1.5, 'DisplayName',sprintf('Predator (k = %d)',k(k1)))  %This section gives me the error. It works, however, if I use t instead of k.
    plot3(k(k1)*ones(size(t)),t,Pm{k1}(:,2), 'LineWidth',1.5, 'DisplayName',sprintf('Prey (k = %d)',k(k1)))
end
hold off
grid
xlim([-1.1  1.1])
legend('Location','NW')
view(30, 30)