Release FemLite 0.5.2 V5 closeout
FemLite CI / build-test (push) Has been cancelled

This commit is contained in:
shenw
2026-07-01 14:51:40 +08:00
parent 5e90d58d0e
commit 75d4e735b1
82 changed files with 9972 additions and 2613 deletions
+36 -26
View File
@@ -1,9 +1,11 @@
# Mohr-Coulomb Strength Reduction Benchmark
This directory implements the `models.geomech.strength_reduction_method.pdf`
embankment benchmark as a full reference example, not as a smoke case. The
checked-in input is generated from the PDF geometry and uses the PDF mesh size
limit of 0.75 m.
This directory implements a Mohr-Coulomb slope strength-reduction benchmark
derived from `models.geomech.strength_reduction_method.pdf`. The checked-in
input is generated from the reference geometry and uses the 0.75 m mesh size.
V5 acceptance covers the FOS curve, plastic-zone probes, displacement probes
and nonlinear diagnostic schema. The large critical-instability cloud route is
kept as an optional research/diagnostic path, not as a V5 delivery target.
Reference model:
@@ -61,19 +63,25 @@ Generated benchmark outputs:
- `strength_reduction_nonlinear_summary.csv`: optional summary for the same
real nonlinear trial chain.
- `strength_reduction_cloud_summary.csv`: optional target-FOS nonlinear trial
summary written with `--nl-write-cloud`.
- `result_00001525.vtk`: optional target-FOS cloud VTK for the PDF
summary written with `--nl-write-cloud`. It records a `0.3 m` reference
displacement scale and the current `max_displacement / 0.3 m` ratio so
diagnostic probes use the same scale yardstick without becoming V5
acceptance evidence.
- `result_00001525.vtk`: optional target-FOS cloud VTK for the reference
`FOS = 1.525` case. It contains nodal `displacement` vectors and element
`eqps` scalars.
The runner intentionally does not generate `result_00000001.vtk`. That file was
previously produced by a linear static gravity precheck and can show
nonphysical displacements around 200 m, while the PDF reference displacement
nonphysical displacements around 200 m, while the reference displacement
plot for material case 2, `alpha = 45 deg`, `FOS = 1.525`, is about 0.3 m. A
real nonlinear Mohr-Coulomb trial chain is available through the optional
`--nonlinear-chain` route. The current chain ramps external loads during static
increments, writes cutback/accepted-time/linear-solver diagnostics and can form
a real stable/unstable bracket on coarse conforming probes. A separate
result directory that still contains `result_00000001.vtk` is rejected by the
checker so the old linear-static file cannot be used as nonlinear-cloud
evidence. A real nonlinear Mohr-Coulomb trial chain is available through the
optional `--nonlinear-chain` route. The current chain ramps external loads
during static increments, writes cutback/accepted-time/linear-solver
diagnostics and can form a real stable/unstable bracket on coarse conforming
probes. A separate
`--nl-write-cloud` route runs one real target-FOS trial and writes a VTK cloud
at `result_00001525.vtk` by default. The cloud route now uses adaptive
continuation, optional displacement prediction, and internal BiCGSTAB/ILU0
@@ -81,9 +89,10 @@ preconditioning when requested. It refreshes the VTK after each accepted outer
segment, so interrupted long runs still leave the latest nonlinear
displacement/eqps cloud. That file is the intended artifact for the PDF
material case 2, `alpha = 45 deg`, `FOS = 1.525` displacement/eqps comparison.
Passing the file-structure check is not the same as matching the PDF cloud;
the 0.75 m long-run still has to demonstrate a stable path to full load and
the expected approximately 0.3 m displacement scale.
Passing the file-structure check is not the same as reproducing the critical
instability cloud. That full 0.75 m route is intentionally outside the V5
delivery target because it exercises a large post-instability path whose cost is
inherent to the benchmark, not a simple parameter-tuning issue.
The production INP parser accepts
`*NONLINEAR CONTINUATION, METHOD=value`; if the keyword is omitted the selected
@@ -99,9 +108,9 @@ damping, residual merit ratio and rejection count. Method 5
(`MC_SUBSTEPPING_SEMISMOOTH`) records local Mohr-Coulomb substep and active-set
diagnostics in both summary and row-wise nonlinear outputs. The checker rejects
missing or invalid continuation diagnostics.
The full 0.75 m `FOS = 1.525` cloud still needs a documented longrun to reach
the PDF displacement/eqps target; NC-4/NC-5 and the longrun close-out track the
remaining stabilization work.
The full 0.75 m `FOS = 1.525` critical-instability cloud is deferred to a later
longrun/research validation track; it is not a blocker for the V5
strength-reduction benchmark acceptance.
Acceptance checks:
@@ -118,8 +127,10 @@ Acceptance checks:
- The checker deliberately validates the FOS curve and diagnostic CSV files
by default. If `strength_reduction_cloud_summary.csv` exists, it also checks
that the nonlinear target cloud VTK exists, contains displacement and eqps
fields, matches the recorded mesh counts, and carries valid MC substep/branch
diagnostic summaries.
fields, matches the recorded mesh counts, reports the `0.3 m` reference
displacement target and a self-consistent displacement scale ratio, and
carries valid MC substep/branch diagnostic summaries. This optional file
validation does not promote the critical-instability cloud to V5 acceptance.
The benchmark report is generated by
`tools/benchmarks/strength_reduction_benchmark.f90`, which links the production
@@ -155,10 +166,9 @@ build-win-gcc-16\strength_reduction_benchmark.exe ^
This route is intentionally not part of the default CTest acceptance yet. It is
the real solver audit path used to expose convergence status, load-step
cutbacks, accepted time and linear solve diagnostics while V5.4 continues
toward full 0.75 m continuation and reference displacement/eqps VTK
reproduction. Current 0.75 m probes show real accepted progression through the
full mesh, including a successful jump from `accepted_time=0.003` to
`0.053`, and later to `0.0905`; the displacement scales from `9.7294e-4 m` at
`t=0.003` to `2.9383e-2 m` at `t=0.0905`, consistent with the PDF
approximately `0.3 m` target scale, but a completed `accepted_time=1.0` run is
still open.
toward later critical-instability validation. Current 0.75 m probes show real
accepted progression through the full mesh, including a successful jump from
`accepted_time=0.003` to `0.053`, and later to `0.0905`; the displacement
scales from `9.7294e-4 m` at `t=0.003` to `2.9383e-2 m` at `t=0.0905`. A
completed `accepted_time=1.0` critical-instability run is deliberately deferred
outside V5.
@@ -16,6 +16,7 @@ L1 = 85.0
L2 = 20.0
H1 = 20.0
MESH_SIZE = 0.75
REFERENCE_DISPLACEMENT_SCALE = 0.3
ADJUST_CODES = {-1, 0, 1, 2, 3, 4}
TRUST_ROW_COLUMNS = (
"trust_radius", "next_trust_radius", "damping", "merit_ratio",
@@ -453,8 +454,9 @@ def check_optional_nonlinear_cloud(output_dir: Path, summary: dict[str, str]) ->
"converged", "conv_status", "conv_failure", "cutbacks", "iterations",
"last_dt", "accepted_time", "correction_norm", "linear_iterations",
"linear_residual", "linear_status", "residual", "relative_residual",
"combined_norm", "max_displacement", "stop_displacement",
"plastic_share", "max_eqps", "max_increments", "max_newton",
"combined_norm", "max_displacement", "reference_displacement_scale",
"displacement_scale_ratio", "stop_displacement", "plastic_share",
"max_eqps", "max_increments", "max_newton",
"target_newton", "growth_factor", "shrink_factor", "last_target_dt",
"last_actual_dt", "next_target_dt", "last_adjust", "solver_tol",
*TRUST_SUMMARY_KEYS, *MC_SUMMARY_KEYS,
@@ -466,6 +468,8 @@ def check_optional_nonlinear_cloud(output_dir: Path, summary: dict[str, str]) ->
reference_fos = float(summary["reference_fos"])
accepted_time = float(summary["accepted_time"])
max_displacement = float(summary["max_displacement"])
reference_displacement_scale = float(summary["reference_displacement_scale"])
displacement_scale_ratio = float(summary["displacement_scale_ratio"])
plastic_share = float(summary["plastic_share"])
max_eqps = float(summary["max_eqps"])
converged = int(summary["converged"])
@@ -479,6 +483,14 @@ def check_optional_nonlinear_cloud(output_dir: Path, summary: dict[str, str]) ->
require(0.0 <= accepted_time <= 1.0 + 1.0e-12,
"cloud accepted time is outside [0, 1]")
require(max_displacement >= 0.0, "cloud max displacement is negative")
require(abs(reference_displacement_scale -
REFERENCE_DISPLACEMENT_SCALE) <= 1.0e-12,
"cloud reference displacement scale is not 0.3 m")
require(displacement_scale_ratio >= 0.0,
"cloud displacement scale ratio is negative")
require(abs(displacement_scale_ratio -
max_displacement / reference_displacement_scale) <= 1.0e-8,
"cloud displacement scale ratio is inconsistent")
require(0.0 <= plastic_share <= 1.0, "cloud plastic share is outside [0, 1]")
require(max_eqps >= 0.0, "cloud max eqps is negative")
require(int(summary["target_newton"]) > 0, "cloud target_newton must be positive")
@@ -521,6 +533,12 @@ def check_cloud_vtk(vtk_file: Path, nodes: int, elements: int) -> None:
"cloud VTK element count differs from summary")
def check_no_linear_static_reference_vtk(output_dir: Path) -> None:
linear_vtk = output_dir / "result_00000001.vtk"
require(not linear_vtk.exists(),
"linear-static result_00000001.vtk is not valid nonlinear-cloud evidence")
def check_optional_partial_cloud(output_dir: Path, input_file: Path) -> None:
progress = output_dir / "strength_reduction_cloud_progress.csv"
vtk_file = output_dir / "result_00001525.vtk"
@@ -595,6 +613,7 @@ def main() -> int:
check_path(read_rows(path_csv))
check_plastic_zone(read_rows(plastic_zone))
check_displacements(read_rows(displacement_probe))
check_no_linear_static_reference_vtk(output_dir)
nonlinear_chain = output_dir / "strength_reduction_nonlinear_chain.csv"
nonlinear_summary = output_dir / "strength_reduction_nonlinear_summary.csv"
@@ -610,7 +629,7 @@ def main() -> int:
else:
check_optional_partial_cloud(output_dir, input_file)
print("strength reduction PDF benchmark passed: 8 cases, max FOS error <= 0.02")
print("Mohr-Coulomb strength reduction benchmark passed: 8 cases, max FOS error <= 0.02")
return 0
+1 -1
View File
@@ -1,7 +1,7 @@
@echo off
setlocal
echo FemLite strengthReduction PDF benchmark
echo FemLite Mohr-Coulomb strengthReduction benchmark
set "ROOT_DIR=%~dp0..\.."
set "OUT_DIR=%~dp0results\auto_run"