I try to solve a Modelingtoolkit ODEProblem. Apparently it runs, but I can only access the initial condition.
@mtkbuild m = Setup()
prob = ODEProblem(m, [], (0.,100.), [])
sol = solve(prob)
print(sol[m.c.w])
# get: [0.01165680598011326]
# expected: [0.01, 0.005, ...]
Full code single output time · GitHub
I started with ModelingToolkit.jl/docs/src/tutorials/acausal_components.md at master · SciML/ModelingToolkit.jl · GitHub
If you print the full solution object:
julia> sol
retcode: InitialFailure
Interpolation: 3rd order Hermite
t: 1-element Vector{Float64}:
0.0
u: 1-element Vector{Vector{Float64}}:
[760.0, 0.01165680598011326, 0.01165680598011326, 0.01165680598011326, 0.01165680598011326, 0.01165680598011326, 0.01165680598011326, 760.0, 0.01165680598011326, 760.0]
The retcode field indicates that consistent initial conditions could not be found so the solver never started. Check your model carefully and maybe try modeling smaller pieces of it and building up to the full system.
Are you sure all of your initial values are consistent? It doesn’t look to be the case. Did you mean for some of them to be guesses?
I’ve updated the gist. Now I get Success and a vector of values, but the system is at steady state no matter where I start it.
D(w)*P ~ ifelse(a.F <= 0, 0, 760 * (a.F * a.w) / V)
Seems to be the problem. It looks like a.F is a single value (net standard cm3/s flow into the tank). I need to refer to the vector of flows connected to the chamber flange c.a. This ModelingToolkit.jl/src/systems/connectors.jl at master · SciML/ModelingToolkit.jl · GitHub seems to do something like that, but I have no idea how to use it. The alternative seems to be making PressureTank with 4 flanges so that connect() only joins two flanges.
check the equations of your initialization system: are there any initial equations you have that are accidentally forcing you to be instead state? Some stray velocity = 0 condition you have around showing up in that equation list?