Hello,
I have trouble understanding the modeling scheme of modern MTK with pins /connectors and so on. As a toy example, I try to implement this simple system of two coupled oscillators:
I tried to model it using the following code:
using Pkg
pkg"activate --temp"
pkg"add ModelingToolkit"
using ModelingToolkit
using ModelingToolkit: t_nounits as t, D_nounits as Dt
@connector θPPin begin
θ(t), [description = "voltage angle"]
P(t), [connect = Flow, description = "Power"]
end
@mtkmodel Swing begin
@components begin
terminal = θPPin()
end
@variables begin
ω(t) = 0.0, [description = "Rotor frequency"]
end
@parameters begin
M = 1, [description = "Inertia"]
D = 0.1, [description = "Damping"]
Pmech, [description = "Mechanical Power"]
end
@equations begin
Dt(terminal.θ) ~ ω
Dt(ω) ~ 1/M * (Pmech - D*ω - terminal.P)
end
end
@mtkmodel Line begin
@components begin
dst = θPPin()
src = θPPin()
end
@parameters begin
K = 1.0, [description = "Line conductance"]
end
@variables begin
P(t), [description = "Power flow"]
end
@equations begin
P ~ -K*sin(dst.θ - src.θ)
src.P ~ -P
dst.P ~ P
end
end
@named swing1 = Swing(Pmech=1)
@named swing2 = Swing(Pmech=-1)
@named line = Line()
eqs = [connect(swing1.terminal, line.src),
connect(swing2.terminal, line.dst)]
@named twoswingmodel = ODESystem(eqs, t, systems = [swing1, swing2, line])
equations(twoswingmodel) # output not helpful because I don't know what happens in `connect()`
sys = structural_simplify(twoswingmodel)
equations(sys)
I know the system is completely reducable to a 4-state ODE. However there seems to be something wrong with my implementation. structural_simplify gets rid of the coupling entirely.
julia> equations(sys)
4-element Vector{Equation}:
Differential(t)(swing1₊terminal₊θ(t)) ~ swing1₊ω(t)
Differential(t)(swing1₊ω(t)) ~ (swing1₊Pmech - swing1₊terminal₊P(t) - swing1₊D*swing1₊ω(t)) / swing1₊M
Differential(t)(swing2₊terminal₊θ(t)) ~ swing2₊ω(t)
Differential(t)(swing2₊ω(t)) ~ (swing2₊Pmech - swing2₊terminal₊P(t) - swing2₊D*swing2₊ω(t)) / swing2₊M
