30.06.2013 Views

multiple time scale dynamics with two fast variables and one slow ...

multiple time scale dynamics with two fast variables and one slow ...

multiple time scale dynamics with two fast variables and one slow ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

nates. The computer algebra system Maple [66] was used in this case:<br />

l Ku<br />

1<br />

l MC<br />

1<br />

= 4K<br />

√ ǫ + O( √ ǫ)<br />

= 4Kω0 √<br />

√ + O(ω0 ǫ)<br />

ǫ<br />

l GH<br />

1 = K+ O(ǫ) (6.21)<br />

l CLW<br />

1<br />

l Pe<br />

1<br />

= K+ O(ǫ)<br />

3πK<br />

= + O(ǫ/ω0)<br />

64ω0<br />

Using the results (6.21) we now have a direct strategy how to analyze a canard<br />

explosion generated in a singular Hopf bifurcation.<br />

1. Compute the location of the Hopf bifurcation. This givesλH.<br />

2. Find the first Lyapunov coefficient at the Hopf bifurcation, e.g. we get<br />

l MC<br />

1<br />

≈ 4Kω0/ √ ǫ.<br />

3. Compute the location of the maximal canard, <strong>and</strong> hence the canard explo-<br />

sion, byλc≈λH− Kǫ. For example, using MatCont we would get<br />

λc≈λH− lMC<br />

1<br />

ǫ<br />

4ω0<br />

3/2<br />

(6.22)<br />

Observe that the previous calculation may require calculating the eigenvalues at<br />

the Hopf bifurcation to determineω0 but does not require any center manifold<br />

calculations nor additional normal form transformations; these have basically<br />

been encoded in the calculation of the Lyapunov coefficient.<br />

158

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!