Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Is the initial value problem of an ODE considered as a dynamic system?

A dynamic system is defined as

In the most general sense, a dynamical system is a tuple (T, M, Φ) where T is a monoid, written additively, M is a set and Φ is a function $$ \Phi: U \subset T \times M \to M $$ with $$ I(x) = \{ t \in T : (t,x) \in U \}\,$$ $$ \Phi(0,x) = x\,$$ $$ \Phi(t_2,\Phi(t_1,x)) = \Phi(t_1 + t_2, x),\, for \, t_1, t_2, t_1 + t_2 \in I(x)\, $$

For an initial value problem $$ \dot{\boldsymbol{x}}=\boldsymbol{v}(t,\boldsymbol{x}) $$ $$ \boldsymbol{x}|_{{t=0}}=\boldsymbol{x}_0 $$

the solution is an evolution function $$ \boldsymbol{{x}}(t)=\Phi(t,\boldsymbol{{x}}_0) $$

I don't think $\Phi$ satisfy $\Phi(t_2,\Phi(t_1,x)) = \Phi(t_1 + t_2, x),\, for \, t_1, t_2, t_1 + t_2 \in I(x)\, $. For example, when $v(t,x) = f(t)$, $\Phi(t,\boldsymbol{{x}}_0) = \int_0^t f(s) ds + x_0$.

But I often heard of the initial value problem of ODE and dynamic system together. If the former is not an example of the latter, I wonder why they are mentioned together?

Thanks!

There is also a related question posted before here Construction of dynamical systems from ODEs with initial values.

share|improve this question

1 Answer 1

up vote 3 down vote accepted

Your guess that the flow map $\Phi(t,x_0) =x(t)$ in the ODE with initial point $x_0$ satisfies the dynamical system property is actually correct, if the ODE is autonomous (time-independent), and if solutions exist locally in time. In your example, $\dot{x} = f(t)$ depends on time, so at face value the flow map does not satisfy the semigroup property. However, if $\dot{x} = v(x,t)$ is time-dependent, then we can relabel $t$ to a new variable $y$ and form the system $$ \dot{x} = v(x,y), \, \dot{y} = 1$$ and this can be interpreted as a time-independent system.

Edit: To answer the questions posed in the comments:

  1. An ODE of the form $\dot{x} = v(x)$ does not necessarily even have a solution; the existence and uniqueness of such solutions, which is one of the main items proven in a course on the subject, is required for the idea of an ODE being a dynamical system to even make sense. A sufficient condition for existence/uniqueness is that the vector field $v(x)$ is a locally Lipschitz continuous map.

  2. From this, the semigroup property for the flow map $\Phi$ can be demonstrated. The flow $\Phi(t,x_0)$ gives the value of $x(t)$ as the solution to the initial value problem $x(0) = x_0$. Now, $\Phi(t_2, \Phi(t_1, x_0))$ gives the unique value after $t_2$ time after the initial condition given by $x(0) = \Phi(t_1, x_0)$. In turn, $\Phi(t_1, x_0)$ is the unique value $x(t_1)$ to the initial condition given by $x(0) = x_0$. Then $\Phi(t_1 + t_2, x_0)$ gives the solution to up to time $t_1 + t_2$; by uniqueness the corresponding solution curve $x(t)$ passes through the point $x(t_1)$ and $x(t_1 + t_2)$. Then, by uniqueness, $x(t_1 + t_2)$ must also be the unique value which is $t_2$-time after $x(t_1)$; therefore we have $$ \Phi(t_1 + t_2, x_0) = \Phi(t_2, \Phi(t_1, x_0))$$

share|improve this answer
    
Thanks! But isn't it $Φ(t_2,Φ(t_1,x_0))=∫^{t_2}_0 f(s)ds+Φ(t_1,x_0) = ∫^{t_2}_0 f(s)ds+ ∫^{t_1}_0 f(s)ds + x_0$? –  Tim Nov 18 '12 at 21:49
    
That's a good catch, my answer will be edited shortly. –  Christopher A. Wong Nov 18 '12 at 21:54
    
Thanks for the edit! (1) What does "solutions exist locally in time" mean? (2) Can you show a little bit that the initial problem of ODE is a dynamic system when the ODE is autonomous (time-independent), and if solutions exist locally in time? –  Tim Nov 19 '12 at 0:00
    
Some references for my questions are also appreciated. Thanks! –  Tim Nov 19 '12 at 1:58

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.