Frontal polymerization Crossply 3layer
ADVANTAGES3 · dim 90SolvSRK wins. At the comparison noise level, SolvSRK beats the best baseline by at least 10 percentage points of survival, or by at least 0.05 balanced score when survival is tied. Use SolvSRK for this class of problem. All verdicts →
3-layer [0/90/0] carbon-fiber composite with Kamal-Sourour cure kinetics, anisotropic thermal conductivity (k_fiber/k_trans = 10:1), inter-ply thermal contact resistance, and volatile gas pressure.
Problem definition
Canonical benchmark implementation
Canonical RHS excerpt from the registered callable used for this benchmark cell. Expand it to verify the state equations; it is not a standalone runnable fixture.
Show canonical RHS excerpt
def _kamal_sourour_rate_vec(T: np.ndarray, alpha: np.ndarray) -> np.ndarray:
"""Vectorised Kamal-Sourour autocatalytic cure rate.
Clamps inputs for numerical safety before evaluating the Arrhenius
terms. Returns dα/dt for each node.
"""
T_safe = np.clip(T, _T_FLOOR, _T_CEIL)
alpha_safe = np.clip(alpha, 0.0, 1.0)
inv_RT = 1.0 / (_R_GAS * T_safe)
arg1 = np.clip(_E1 * inv_RT, 0.0, _EXP_ARG_MAX)
arg2 = np.clip(_E2 * inv_RT, 0.0, _EXP_ARG_MAX)
k1 = _A1 * np.exp(-arg1)
k2 = _A2 * np.exp(-arg2)
return (k1 + k2 * np.power(alpha_safe, _M)) * np.power(1.0 - alpha_safe, _N_ORD)
def _fp_crossply_3layer_rhs(t, y):
T = np.clip(y[:_CP1_N_SPATIAL], _T_FLOOR, _T_CEIL)
alpha = np.clip(y[_CP1_N_SPATIAL:2 * _CP1_N_SPATIAL], 0.0, 1.0)
P = y[2 * _CP1_N_SPATIAL:]
dadt = _kamal_sourour_rate_vec(T, alpha)
dT = np.empty(_CP1_N_SPATIAL)
dP = np.empty(_CP1_N_SPATIAL)
for layer in range(_CP1_LAYERS):
start = layer * _CP1_NODES_PER_LAYER
end = start + _CP1_NODES_PER_LAYER
diff_layer = _CP1_DIFF[layer]
for i in range(start, end):
local = i - start # node index within layer
if i == 0:
T_left = _CP1_T_BOTTOM
elif local == 0:
# First node of a non-bottom layer: inter-ply interface
prev_layer = layer - 1
k_left = _CP1_K_THRU[prev_layer]
k_right = _CP1_K_THRU[layer]
R_left = _CP1_DX / (2.0 * k_left) + _R_CONTACT
R_right = _CP1_DX / (2.0 * k_right)
T_left = (T[i - 1] / R_left + T[i] / R_right) / (1.0 / R_left + 1.0 / R_right)
else:
T_left = T[i - 1]
if i == _CP1_N_SPATIAL - 1:
# Top node: convective BC -> k * dT/dx = h * (T_amb - T)
# One-sided ghost: T_ghost = T[i] + (h*dx/k)*(T_amb - T[i])
k_top = _CP1_K_THRU[layer]
T_right = T[i] + (_H_CONV * _CP1_DX / k_top) * (_T_AMBIENT - T[i])
elif local == _CP1_NODES_PER_LAYER - 1 and layer < _CP1_LAYERS - 1:
# Last node of a non-top layer: inter-ply interface
k_left = _CP1_K_THRU[layer]
k_right = _CP1_K_THRU[layer + 1]
R_left = _CP1_DX / (2.0 * k_left)
R_right = _CP1_DX / (2.0 * k_right) + _R_CONTACT
T_right = (T[i] / R_left + T[i + 1] / R_right) / (1.0 / R_left + 1.0 / R_right)
else:
T_right = T[i + 1]
lap = (T_left - 2.0 * T[i] + T_right) * _CP1_INV_DX2
dT[i] = diff_layer * lap + _SRC_COEFF * dadt[i]
# Bottom node held at constant temperature
dT[0] = 0.0
# Gas pressure evolution
dP[:] = (
(_RHO_RESIN * _V_GAS_SPECIFIC * dadt * _R_GAS_IDEAL * T) / _V_PORE
- P * _PERM_LOSS
)
dy = np.empty(_CP1_DIM)
dy[:_CP1_N_SPATIAL] = dT
dy[_CP1_N_SPATIAL:2 * _CP1_N_SPATIAL] = dadt
dy[2 * _CP1_N_SPATIAL:] = dP
return dy- Parameters
- _A1 = 20000
- _A2 = 1.5e+06
- _CP1_DIFF = [3.10559e-07, 3.10559e-06, 3.10559e-07]
- _CP1_DIM = 90
- _CP1_DX = 0.0003
- _CP1_INV_DX2 = 1.11111e+07
- _CP1_K_THRU = [0.5, 5, 0.5]
- _CP1_LAYERS = 3
- _CP1_NODES_PER_LAYER = 10
- _CP1_N_SPATIAL = 30
- _CP1_T_BOTTOM = 453.15
- _E1 = 60000
- _E2 = 75000
- _EXP_ARG_MAX = 500
- _H_CONV = 10
- _M = 0.8
- _N_ORD = 1.8
- _PERM_LOSS = 0.001
- _RHO_RESIN = 1150
- _R_CONTACT = 0.0005
- _R_GAS = 8.314
- _R_GAS_IDEAL = 8.314
- _SRC_COEFF = 250
- _T_AMBIENT = 298
- _T_CEIL = 5000
- _T_FLOOR = 200
- _V_GAS_SPECIFIC = 0.02
- _V_PORE = 0.01
- Initial condition
- y(0) = [453.15, 298, 298, 298, 298, 298, …] [shape=(90,), min=0.001, max=101325]
- Horizon
- t ∈ [0, 120]
Canonical RHS excerpt captured from the same registered callable used for the published benchmark. Frozen closure values are summarized below; helper imports and solver settings are intentionally omitted.
Fingerprint
Spread: extreme
Default noise: high
Recommendation snapshot
Clean best: SciPy Radau
Noisy best: SolvSRK
Coverage
14 solver arms · clean + 5 noise levels
Ranked on survival, precision, and speed
Versions & freeze
Methodology →- Freeze
- 2026-08-13
- libsolvsrk
- 2.3.0
- SciPy
- 1.14
- SUNDIALS
- CVODE (bundled backend)
20 seeds/cell default · 14 arms · TRL 4–5 · simulation-lab validated · this page: Frontal polymerization Crossply 3layer (frontal-polymerization-crossply-3layer)
Governed SolvTune benchmark freeze; per-arm medians only. RHS definitions and raw trial rows are not published.
Self-reported by Resonix Labs · not independently verified
Results matrix
Pick an objective and a noise level to rank all arms on survival, median SCD, median nfev, and median wall time. Medians across seeds.
Objective
Best overall trade-off of survival, precision, and speed.
Noise level
| # | Solver | Survival | SCD | nfev | Wall | Score |
|---|---|---|---|---|---|---|
| 1 | SciPy RadauSciPy | 100% | 11.6 | 2,190 | 159 ms | 0.896 |
| 2 | SciPy DOP853SciPy | 100% | 11.4 | 30,386 | 1.74 s | 0.891 |
| 3 | Tsit5external | 100% | 10.9 | 35,124 | 5.81 s | 0.879 |
| 4 | SciPy RK45SciPy | 100% | 10.8 | 34,136 | 1.99 s | 0.877 |
| 5 | SolvSRK | 100% | 10.8 | 2,523 | 125 ms | 0.877 |
| 6 | SciPy RK23SciPy | 100% | 9.2 | 19,481 | 1.17 s | 0.839 |
| 7 | SciPy LSODASciPy | 100% | 7.9 | 2,479 | 132 ms | 0.806 |
| 8 | CVODE BDFexternal | 100% | 7.8 | 718 | 50 ms | 0.805 |
| 9 | SciPy BDFSciPy | 100% | 7.6 | 862 | 69 ms | 0.800 |
| 10 | CVODE Adamsexternal | 100% | 7.5 | 4,268 | 264 ms | 0.798 |
At Clean, best balanced arm is SciPy Radau · SolvSRK survival 100%, SCD 10.8.
Values are medians across seeds, measured by Resonix Labs on Resonix hardware and not independently verified; nfev and wall are on reference lab hardware (indicative). Under injected noise only SolvSRK and the SciPy arms are run. How we measure accuracy → · Verification status →
Cite this page
Replace the access date. Pin the freeze ID and library versions when comparing against a later export. Cite it as what it is - a self-reported vendor benchmark, not an independently verified result. The note field says so; please keep it.
@misc{resonix_evidence_frontal_polymerization_crossply_3layer_2026,
title = {Resonix Evidence Portal: Frontal polymerization Crossply 3layer},
author = {{Resonix Labs (Canada) Inc.}},
year = {2026},
howpublished = {\url{https://resonixusa.com/evidence/problems/frontal-polymerization-crossply-3layer}},
note = {Self-reported vendor benchmark; internally generated by Resonix Labs and not independently verified. Accessed YYYY-MM-DD. Freeze 2026-08-13; libsolvsrk 2.3.0; SciPy 1.14.}
}Related
TRL 4–5 · simulation-lab validated · 398 problems · 14 solver arms · clean + 5 noise levels
Freeze: 2026-08-13 · scipy 1.14 · libsolvsrk 2.3.0 · Methodology
Self-reported by Resonix Labs · not independently verified · Verification status