Skip to content

Commit

Permalink
docs fix (#333)
Browse files Browse the repository at this point in the history
  • Loading branch information
JonasKoziorek committed Jun 7, 2024
1 parent 4053d18 commit 850e1f4
Show file tree
Hide file tree
Showing 2 changed files with 10 additions and 7 deletions.
2 changes: 1 addition & 1 deletion docs/src/periodicity.md
Original file line number Diff line number Diff line change
Expand Up @@ -169,7 +169,7 @@ davidchacklai
#### Logistic Map example

The idea of periodic orbits can be illustrated easily on 1D maps. Finding all periodic orbits of period
$n$ is equivalent to finding all points $x$ such that $f^{n}(x)=x$, where $f^{n}$ is $n$-th composition of $f$. Hence, solving $f^{n}(x)-x=0$ yields such points. However, this is impossible analytically.
$n$ is equivalent to finding all points $x$ such that $f^{n}(x)=x$, where $f^{n}$ is $n$-th composition of $f$. Hence, solving $f^{n}(x)-x=0$ yields such points. However, this is often impossible analytically.
Let's see how `davidchacklai` deals with it:

First let's start with finding first $9$ periodic orbits of the logistic map for parameter $3.72$.
Expand Down
15 changes: 9 additions & 6 deletions src/periodicity/davidchacklai.jl
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,7 @@ periodic orbits will be used to detect periodic orbits of order `m+1` to `n`.
is an `n`-periodic point.
* `abstol = 1e-8`: A detected periodic point isn't stored if it is in `abstol`
neighborhood of some previously detected point. Distance is measured by
euclidian norm. If you are getting duplicate periodic points, decrease this value.
euclidian norm. If you are getting duplicate periodic points, increase this value.
## Description
Expand All @@ -37,7 +37,7 @@ by turning fixed points of the original
map `ds` to stable ones, through the transformation
```math
\\mathbf{x}_{n+1} = \\mathbf{x}_{n} +
[\\beta |g(\\mathbf{x}_{n}| C^{T} - J(\\mathbf{x}_{n})]^{-1} g(\\mathbf{x}_{n})
[\\beta |g(\\mathbf{x}_{n})| C^{T} - J(\\mathbf{x}_{n})]^{-1} g(\\mathbf{x}_{n})
```
where
```math
Expand All @@ -46,7 +46,7 @@ g(\\mathbf{x}_{n}) = f^{n}(\\mathbf{x}_{n}) - \\mathbf{x}_{n}
and
```math
J(\\mathbf{x}_{n}) = \\frac{\\partial g(\\mathbf{x}_{n})}{\\partial \\mathbf{x}_{n}}
````
```
The main difference between Schmelcher & Diakonos[Schmelcher1997](@cite) and
Davidchack & Lai[Davidchack1999](@cite) is that the latter uses periodic points of
Expand All @@ -58,9 +58,12 @@ while `davidchacklai` detects periodic points of all orders up to `n`.
## Important note
For low periods `n` circa less than 6, you should select `m = n` otherwise the algorithm
detect periodic orbits correctly. For higher periods, you can select `m` as 6.
You can use initial grid of points for `ics`. Increase `m` in case the orbits are
not being detected correctly.
won't detect periodic orbits correctly. For higher periods, you can select `m` as 6.
We recommend experimenting with `m` as it may depend on the specific problem.
Increase `m` in case the orbits are not being detected correctly.
Initial conditions `ics` can be selected as a uniform grid of points in the state space or
subset of a chaotic trajectory.
"""
function davidchacklai(
Expand Down

0 comments on commit 850e1f4

Please sign in to comment.