Skip to content

Commit

Permalink
rewrite nonlinear solve tutorial
Browse files Browse the repository at this point in the history
  • Loading branch information
ArnoStrouwen committed Jul 31, 2024
1 parent a8b341f commit a44bbea
Showing 1 changed file with 16 additions and 8 deletions.
24 changes: 16 additions & 8 deletions docs/src/tutorials/nonlinear.md
Original file line number Diff line number Diff line change
@@ -1,10 +1,18 @@
# Modeling Nonlinear Systems

In this example, we will go one step deeper and showcase the direct function
generation capabilities in ModelingToolkit.jl to build nonlinear systems.
Let's say we wanted to solve for the steady state of an ODE. This steady state
is reached when the nonlinear system of differential equations equals zero.
We use (unknown) variables for our nonlinear system.
ModelingToolkit.jl is not only useful for generating initial value problems (`ODEProblem`).
The package can also build nonlinear systems.
This is, for example, useful for finding the steady state of an ODE.
This steady state is reached when the nonlinear system of differential equations equals zero.

!!! note

The high level `@mtkmodel` macro used in the
[getting started tutorial](@ref getting_started)
is not yet compatible with `NonlinearSystem`.
We thus have to use a lower level interface to define nonlinear systems.
For an introduction to this interface, read the
[programmatically generating ODESystems tutorial](@ref programmatically).

```@example nonlinear
using ModelingToolkit, NonlinearSolve
Expand All @@ -15,8 +23,6 @@ using ModelingToolkit, NonlinearSolve
eqs = [0 ~ σ * (y - x)
0 ~ x * (ρ - z) - y
0 ~ x * y - β * z]
guesses = [x => 1.0, y => 0.0, z => 0.0]
ps = [σ => 10.0, ρ => 26.0, β => 8 / 3]
@mtkbuild ns = NonlinearSystem(eqs)
guesses = [x => 1.0, y => 0.0, z => 0.0]
Expand All @@ -26,7 +32,9 @@ prob = NonlinearProblem(ns, guesses, ps)
sol = solve(prob, NewtonRaphson())
```

We can similarly ask to generate the `NonlinearProblem` with the analytical
We found the `x`, `y` and `z` for which the right hand sides of `eqs` are all equal to zero.

Just like with `ODEProblem`s we can generate the `NonlinearProblem` with its analytical
Jacobian function:

```@example nonlinear
Expand Down

0 comments on commit a44bbea

Please sign in to comment.