RDE.jl provides a solver for the rotating detonation engine (RDE) model equations presented in Koch et al. (2020):
The package is registered in a custom package registry, KristianHolmeRegistry. To install the package through the registry, first add the registry.
From the Julia REPL, type ] to enter the Pkg REPL mode and run:
pkg> registry add https://github.com/KristianHolme/KristianHolmeRegistryand then install with
pkg> add RDEIf you don't want to add the custom registry, you can install RDE.jl directly from github.
pkg> add https://github.com/KristianHolme/RDE.jlusing RDE
using GLMakie
# Create and solve a basic RDE problem
params = RDEParam()
rde_prob = RDEProblem(params)
solve_pde!(rde_prob)
plot_solution(rde_prob)The solver defaults to a conservative finite-volume discretization with an explicit SSPRK integrator and CFL-based steps. You can override the integrator and time stepping options:
using OrdinaryDiffEqSSPRK
solve_pde!(rde_prob; alg = SSPRK33(), adaptive = false)Injection u_p(x, t) and s(x, t) live on RDEProblem as prob.injection. The RHS samples them each step with update_injection!. Construction without an injection argument builds a uniform C∞ multi-step hold at params.u_p / params.s. Reset then fills that same constant hold, so time-dependent control is installed afterwards with commit_schedule!.
params = RDEParam(; tmax = 10)
prob = RDEProblem(params) # UniformMultiStepPressureProfile at params.u_p
commit_schedule!(prob.injection, 0.0f0, 2.0f0, 0.8f0) # live value at t=0 → 0.8 at t=2
solve_pde!(prob)A vector of knots is a multi-step schedule. The default smoother is CinftySmoother(τ) (τ_smooth = 1): C∞ step-then-hold on each segment. Construct the profile with LinearSmoother() for linear interpolation between knots.
commit_schedule!(prob.injection, 0.0f0, [2.0f0, 5.0f0], [0.8f0, 0.3f0])Pass spatial_kernel_width > 0 for a spatial profile. New knots are smoothed at commit time only; the previous live field is not re-smoothed.
params = RDEParam(; N = 64)
prob = RDEProblem(params; spatial_kernel_width = 9)
target = fill(0.8f0, params.N)
target[1:(params.N ÷ 2)] .= 0.3f0
commit_schedule!(prob.injection, 0.0f0, 1.0f0, target)control_shift_strategy wraps the inner profile in a moving frame (MovingFrameInjectionProfile). You can also pass a prebuilt injection.
prob = RDEProblem(params; control_shift_strategy = LinearControlShift(1.0f0))
prob = RDEProblem(params; injection = ConstantInjectionProfile(0.5f0, 3.5f0))params = RDEParam() # or RDEParam(; N = 32, tmax = 0.01) for a quick run
prob = RDEProblem(params; reset_strategy = NShock(2)) # Initialize with 2 shocks
solve_pde!(prob)
# Or use random shock initialization (1-4 shocks)
reset_strategy = RandomShock()
params = RDEParam()
prob = RDEProblem(params; reset_strategy)
# or use a custom function for the u-variable (x is the spatial grid vector)
reset_strategy = CustomPressureReset() do x
T = eltype(x)
return abs.(x) ./ T(π)
end
prob = RDEProblem(params; reset_strategy)
solve_pde!(prob)
plot_solution(prob)Some reset strategies and predictors rely on artifact data (shock profiles and speed model). These are initialized at package load time; if the artifacts are missing, the package will warn and the related features will throw a clear error when used.
@article{PhysRevE.101.013106,
title = {Mode-locked rotating detonation waves: Experiments and a model equation},
author = {Koch, James and Kurosaka, Mitsuru and Knowlen, Carl and Kutz, J. Nathan},
journal = {Phys. Rev. E},
volume = {101},
issue = {1},
pages = {013106},
numpages = {11},
year = {2020},
month = {Jan},
publisher = {American Physical Society},
doi = {10.1103/PhysRevE.101.013106},
url = {https://link.aps.org/doi/10.1103/PhysRevE.101.013106}
}
@article{Koch_2021,
title={Multiscale physics of rotating detonation waves: Autosolitons and modulational instabilities},
volume={104},
ISSN={2470-0053},
url={http://dx.doi.org/10.1103/PhysRevE.104.024210},
DOI={10.1103/physreve.104.024210},
number={2},
journal={Physical Review E},
publisher={American Physical Society (APS)},
author={Koch, James and Kurosaka, Mitsuru and Knowlen, Carl and Kutz, J. Nathan},
year={2021},
month=aug }