Multiphysics and water networks: paper-to-code routes

On this page

All 60 tools · How to contribute

These are construction routes researched by agents. The execution line reports the bounded work actually performed; the comparison section defines what remains before claiming a published result. Human scientific review is pending.

elmerfem#

Elmer FEM — How does basal topography change ice-sheet surface velocity in ISMIP-HOM A010?

Route: Paper model adaptation. Execution: Not run. Source confidence: medium.

Paper for elmerfem#

O. Gagliardini; T. Zwinger; F. Gillet-Chaulet; G. Durand; L. Favier; B. de Fleurian; R. Greve; M. Malinen; C. Martín; P. Råback; J. Ruokolainen; M. Sacchettini; M. Schäfer; H. Seddik; J. Thies (2013). Capabilities and performance of Elmer/Ice, a new-generation ice sheet model. 10.5194/gmd-6-1299-2013.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Open journal landing page inspected; full numerical benchmark results not extracted.

Target in the paper: Gagliardini et al. (2013), ISMIP-HOM validation discussion; current A010 10 km sinusoidal-bed benchmark model. Exact figure/table target still to select.

Code and inputs for elmerfem#

Artifact or upstream source. Official Elmer/Ice model example linked to benchmark family discussed in the software paper; current source is newer than paper.

Source inspected: yes. Ref: devel. Commit: a4082d84aa7aaff25edbb066648428fafc3ebf75.

Files: elmerice/examples/ISMIP-HOM/A010/ismip_a.sif; elmerice/examples/ISMIP-HOM/A010/README.text

Terms: Elmer components GPL/LGPL; inspect LICENSE.md and solver notices.

Environment and compute: ElmerGrid and ElmerSolver with Elmer/Ice solvers; native Linux build or scoped runtime. CPU; Fortran FEM installation and mesh-refinement runs.

Inputs: Bundled rectangle.grd and SIF; bed geometry, viscosity, units, boundary conditions preserved.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare elmerfem#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [proposed setup] git clone https://github.com/ElmerCSC/elmerfem.git && cd elmerfem && git checkout a4082d84aa7aaff25edbb066648428fafc3ebf75
  2. [upstream README] cd elmerice/examples/ISMIP-HOM/A010 && ElmerGrid 1 2 rectangle.grd
  3. [corrected upstream filename] ElmerSolver ismip_a.sif

Expected outputs: Native mesh/result files and surface velocity field.

Comparison with the paper: Compare dimensionless surface-velocity profile to the ISMIP-HOM A 10 km intercomparison data after selecting exact source dataset and tolerance; residual convergence and mesh refinement are additional checks.

Execution result: Source and setup inspected; no new execution yet.

Execution evidence: None recorded.

Blockers: Elmer runtime not provisioned in this research task.; Exact published reference curve/data and tolerance pending.

Next action: Retrieve the source benchmark reference data; resolve upstream README typo ismisp_a.sif versus actual ismip_a.sif before running.

Gaps: Current code is not the 2013 execution environment.; No published result comparison performed.

Sources for elmerfem#

calculix#

CalculiX — How does a pressure pulse expand an elastic tube in partitioned fluid–structure simulation?

Route: Software paper example. Execution: Not run. Source confidence: medium.

Paper for calculix#

Gerasimos Chourdakis; Kyle Davis; Benjamin Rodenberg; Miriam Schulte; Frédéric Simonis; Benjamin Uekermann; Georg Abrams; Hans-Joachim Bungartz; Lucia Cheung Yau; Ishaan Desai; Konrad Eder; Richard Hertrich; Florian Lindner; Alexander Rusch; Dmytro Sashko; David Schneider; Amin Totounferoush; Dominik Volland; Peter Vollmer; Oguz Ziya Koseomur (2022). preCICE v2: A sustainable and user-friendly coupling library. 10.12688/openreseurope.14445.2.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Metadata verified; article result sections not audited.

Target in the paper: preCICE v2 software paper; current elastic-tube-3d tutorial is a related authors/maintainers example. No exact published numerical target established.

Code and inputs for calculix#

Artifact or upstream source. Official coupling-project tutorial using CalculiX; software-paper-linked current model, not an asserted paper artifact.

Source inspected: yes. Ref: develop. Commit: a2aab6b46cf556ce1a5187a40d8eb2685634f228.

Files: elastic-tube-3d/README.md; elastic-tube-3d/solid-calculix/run.sh; elastic-tube-3d/solid-calculix/tube.inp; elastic-tube-3d/fluid-openfoam/run.sh

Terms: Tutorial repository LGPL-3.0; CalculiX GPL-2.0, OpenFOAM and adapters carry own terms.

Environment and compute: Compatible pinned preCICE, CalculiX adapter/ccx_preCICE, OpenFOAM adapter and pimpleFoam. CPU; two coupled solver processes, seconds-to-hours depending runtime and mesh.

Inputs: Bundled tube mesh, pressure pulse applied for 3 ms then zero for 7 ms, precice-config.xml and solver decks.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare calculix#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [proposed setup] git clone https://github.com/precice/tutorials.git && cd tutorials && git checkout a2aab6b46cf556ce1a5187a40d8eb2685634f228
  2. [upstream route, terminal 1] cd elastic-tube-3d/fluid-openfoam && ./run.sh
  3. [upstream route, terminal 2] cd elastic-tube-3d/solid-calculix && ./run.sh

Expected outputs: CalculiX native structural output, OpenFOAM fields, preCICE midpoint displacement logs.

Comparison with the paper: Compare midpoint displacement time history to bundled reference-results/fluid-openfoam_solid-calculix.tar.gz; agree tolerance and check coupling/mesh/time-step convergence. This reference is a tutorial baseline, not yet a published target.

Execution result: Source and setup inspected; no new execution yet.

Execution evidence: None recorded.

Blockers: Compatible solver/adapter stack not installed for this task.; Exact article target and paper-era versions not established.

Next action: Pin all solver/adaptor versions and inspect the archived tutorial reference before running.

Gaps: No exact paper-result correspondence established.

Sources for calculix#

openmodelica#

OpenModelica — Can safe Bayesian tuning improve inverter current control while constraining unsafe trajectories?

Route: Paper model adaptation. Execution: Not run. Source confidence: medium.

Paper for openmodelica#

Stefan Heid; Daniel Weber; Henrik Bode; Eyke Hüllermeier; Oliver Wallscheid (2020). OMG: A Scalable and Flexible Simulation and Testing Environment Toolbox for Intelligent Microgrid Control. 10.21105/joss.02435.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: JOSS paper metadata and author README/example inspected; associated arXiv white paper 2005.04869 identified.

Target in the paper: Heid et al. (2020) OMG software paper; detailed applied example in associated Bode et al. white paper, safe Bayesian low-level controller tuning. Exact figure/curve comparison pending.

Code and inputs for openmodelica#

Artifact or upstream source. Author-associated OMG project uses OpenModelica as FMU backend and links the JOSS article and white paper.

Source inspected: yes. Ref: master. Commit: a81e522594c4048aa63ae0cfdc2ea4682ca98307.

Files: README.rst; examples/single_inverter_current_control_safe_opt.py; docs/parts/user_guide/examples/single_inverter_current_control_safe_opt.rst; setup.py

Terms: OMG GPL-3.0; OpenModelica, Modelica libraries, PyFMI and FMU components have separate terms.

Environment and compute: Legacy OMG Python environment, PyFMI, OpenModelica/FMU toolchain; use pinned project requirements and package data. CPU; FMU runtime and iterative controller tuning.

Inputs: Bundled inverter FMU, network configuration, controller parameters, SafeOpt/GP settings.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare openmodelica#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [proposed setup] git clone https://github.com/upb-lea/openmodelica-microgrid-gym.git && cd openmodelica-microgrid-gym && git checkout a81e522594c4048aa63ae0cfdc2ea4682ca98307
  2. [proposed setup after matching FMU dependencies] python -m pip install -e .
  3. [upstream example] python examples/single_inverter_current_control_safe_opt.py

Expected outputs: Controller learning histories and inverter current/control plots.

Comparison with the paper: Match source/white-paper system, random seed and tuning budget; compare control error, constraint violations and optimization trace to selected published plot. Numerical target/tolerance awaits detailed white-paper audit.

Execution result: Source and setup inspected; no new execution yet.

Execution evidence: None recorded.

Blockers: Legacy PyFMI/FMUs and Python compatibility need qualification.; No OpenModelica runtime change authorized or made.

Next action: Freeze a paper-era release/FMU and extract exact white-paper figure parameters before reproducing.

Gaps: Current repository is newer than JOSS paper.; JOSS paper itself presents software; applied results are in associated white paper.

Sources for openmodelica#

pybamm#

PyBaMM — Can we recover the published DFN constant-current discharge voltage and inspect model states?

Route: Paper model adaptation. Execution: Not run. Source confidence: high.

Paper for pybamm#

Valentin Sulzer; Scott G. Marquis; Robert Timms; Martin Robinson; S. Jon Chapman (2021). Python Battery Mathematical Modelling (PyBaMM). 10.5334/jors.309.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Open article already inspected in lab; live metadata and official DFN source refreshed.

Target in the paper: Sulzer et al. (2021), Listing 1; DFN simulation of discharge.

Code and inputs for pybamm#

Artifact or upstream source. Official current DFN example implements the model and simulation construction illustrated in Listing 1; current source includes additional options.

Source inspected: yes. Ref: main. Commit: d90fd464edefdd8de5f0de8941da84a319b699ba.

Files: examples/scripts/DFN.py

Terms: BSD-3-Clause; parameter-set and external data terms separate.

Environment and compute: Existing lab pins pybamm 26.10.1.1; exact 2021 comparison needs paper-era environment and defaults. CPU; small DFN demonstration.

Inputs: DFN model, parameter set, current protocol and t_eval preserved explicitly; use paper-era values when claiming reproduction.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare pybamm#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [existing executed adaptation] python simulation-lab/examples/battery_discharge.py --output RUN_DIR
  2. [upstream route after qualifying pinned version] python examples/scripts/DFN.py

Expected outputs: Voltage-time curve and native model solution; existing lab adaptation emits CSV/JSON.

Comparison with the paper: Recreate Listing 1 with matched PyBaMM version/default parameter set, compare voltage-time samples to article figure or archived source run with specified tolerance; current adaptation only confirms execution and model checks.

Execution result: Source and setup inspected; no new execution yet.

Execution evidence: None recorded.

Blockers: Paper-era defaults/environment not yet reconstructed.

Next action: Retrieve archived 2021 source/version and the plotted target curve; retain current checked adaptation as separate baseline.

Gaps: Four prior lab teaching runs remain adaptations, not paper-result reproductions.

Sources for pybamm#

openems#

openEMS — How do cylindrical subgrids change electromagnetic-wave propagation and efficiency?

Route: Related tutorial only. Execution: Not run. Source confidence: medium.

Paper for openems#

Thorsten Liebig; Andreas Rennings; Sebastian Held; Daniel Erni (2012). openEMS – a free and open source equivalent‐circuit (EC) FDTD simulation platform supporting cylindrical coordinates suitable for the analysis of traveling wave MRI applications. 10.1002/jnm.1875.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Publisher abstract and metadata inspected; full numeric methods/figures not audited.

Target in the paper: Liebig et al. paper on cylindrical EC-FDTD; current CylindricalWave_CC.m demonstrates the method family, not the exact traveling-wave MRI geometry.

Code and inputs for openems#

Artifact or upstream source. Official author-maintained cylindrical-grid tutorial; exact paper MRI geometry artifact not located.

Source inspected: yes. Ref: master. Commit: 92cd92a8d924c78a93453c25fa541384c2ac1d0e.

Files: matlab/Tutorials/CylindricalWave_CC.m; matlab/Tutorials/Rect_Waveguide.m

Terms: GPL-3.0; CSXCAD and other runtime component notices apply.

Environment and compute: openEMS/CSXCAD and MATLAB or Octave; current source says openEMS 0.37 and Octave 11.3 tested. CPU; FDTD mesh can be substantial.

Inputs: Bundled cylindrical mesh and excitation; preserve physical units, nested subgrids, absorbing boundary and convergence criterion.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare openems#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [proposed setup] git clone https://github.com/thliebig/openEMS.git && cd openEMS && git checkout 92cd92a8d924c78a93453c25fa541384c2ac1d0e
  2. [proposed bounded execution] run matlab/Tutorials/CylindricalWave_CC.m from Octave with installed openEMS/CSXCAD paths; stop before the final infinite phase-animation loop

Expected outputs: XML model, Et_ra VTK field dumps, Ef_ra.h5 complex fields.

Comparison with the paper: Use field continuity/refinement and analytic cylindrical-wave comparisons as tutorial checks. To reproduce the article, first obtain exact MRI geometry, dielectric properties and published field/performance targets.

Execution result: Source and setup inspected; no new execution yet.

Execution evidence: None recorded.

Blockers: Paper-specific MRI model not found.; Full paper numeric target not audited.; Source tutorial ends in while 1 animation; batch runner must omit animation and record patch.

Next action: Locate author paper archive/model before promoting to paper-result candidate.

Gaps: Current tutorial is related to the paper method, not a direct reproduction route.

Sources for openems#

ngspice#

ngspice — Can the public AutoCkt ngspice operational-amplifier task recover held-out circuit specifications?

Route: Paper result candidate. Execution: Example executed. Source confidence: high.

Paper for ngspice#

Keertana Settaluri; Ameer Haj-Ali; Qijing Huang; Kourosh Hakhamaneshi; Borivoje Nikolic (2020). AutoCkt: Deep Reinforcement Learning of Analog Circuit Designs. 10.23919/date48585.2020.9116200.

Publication type: proceedings-article. Primary metadata. Metadata verification: confirmed. Full-text access: IEEE/author repository and conference/arXiv primary records inspected; live DOI metadata verified.

Target in the paper: SectionIII-B, Figures6-8 and TableII

Code and inputs for ngspice#

Artifact or upstream source. Authors explicitly identify repository as DATE 2020 code and separate ngspice circuits from proprietary Spectre results.

Source inspected: yes. Ref: master. Commit: a6c8a61d3dffb8b433f19251e135994a5b0f6ee4.

Files: README.md; environment.yml; autockt/rollout.py; eval_engines/ngspice/ngspice_wrapper.py; eval_engines/ngspice/ngspice_inputs/yaml_files/two_stage_opamp.yaml

Terms: AutoCkt repository has no detected LICENSE; do not redistribute until reuse terms clarified. ngspice has separate BSD-based terms; transistor-model rights separate.

Environment and compute: Author environment.yml (legacy Python/Ray/TensorFlow), conda; README links ngspice release 27 (spelled 2.7 in text). CPU; README reports 8 cores, warns fewer than 2; training is not a quick smoke test.

Inputs: Bundled two-stage op-amp netlist/45nm model, generated training/test specifications; preserve random seeds and CPU-worker allocation.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare ngspice#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [upstream] conda env create -f environment.yml
  2. [upstream] source activate autockt
  3. [upstream] python eval_engines/ngspice/ngspice_inputs/correct_inputs.py
  4. [upstream paper setting] python autockt/gen_specs.py --num_specs 350
  5. [upstream IPython training] run autockt/val_autobag_ray.py
  6. [upstream validation with explicit user-set counts] run autockt/rollout.py /path/to/ray/checkpoint --run PPO --env opamp-v0 --num_val_specs NUM --traj_len LENGTH --no-render

Expected outputs: Training checkpoints and opamp_obs_reached_test/opamp_obs_nreached_test pickle files.

Comparison with the paper: Descriptive comparison of a new bounded public-source policy with articleTableII; inputs/methods are not matched, so this is not target_compared or paper reproduction. Confidence intervals quantify target-sampling uncertainty conditional on this policy and do not establish equivalence. Training-seed variation has not been characterized.

Execution result: 41/60 final held-out targets reached under author reward; 2 strictly satisfied all constraints. Training 140iterations/168000steps, criterion met=False.

Execution evidence: Author-code attempt; Run record

Blockers: Original policy, exact training/evaluation seeds and a source revision matching all article hyperparameters are unavailable.; Paper/code architecture, training-set size and reward details differ.; Declared bounded training ended before source convergence criterion.

Next action: Recover original author checkpoint and exact configuration, or explicitly agree a new independently seeded matched-method experiment with longer training budget. Do not present saved author arrays or this adaptation as a fresh reproduction of963/1000.

Gaps: Original policy, exact training/evaluation seeds and a source revision matching all article hyperparameters are unavailable.; Paper/code architecture, training-set size and reward details differ.; Declared bounded training ended before source convergence criterion.

Sources for ngspice#

cantera#

Cantera — How does a real-gas equation of state change n-dodecane shock-tube ignition delay?

Route: Paper model adaptation. Execution: Example executed. Source confidence: high.

Paper for cantera#

Gandhali Kogekar; Canan Karakaya; Gary J. Liskovich; Matthew A. Oehlschlaeger; Steven C. DeCaluwe; Robert J. Kee (2018). Impact of non-ideal behavior on ignition delay and chemical kinetics in high-pressure shock tube reactors. 10.1016/j.combustflame.2017.10.014.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Official full HTML example read, including references and tabulated printed results; primary article metadata verified.

Target in the paper: Kogekar et al. (2018), real-gas shock-tube ignition delays; official non_ideal_shock_tube.py follows method using Wang et al. reduced mechanism. It is a later implementation, not exact paper code.

Code and inputs for cantera#

Artifact or upstream source. Official Cantera v3.2.0 example explicitly cites the application paper; exact example executed. It uses a representative reduced mechanism, so article-exact reproduction remains separate.

Source inspected: yes. Ref: v3.2.0. Commit: 4a8358eb80cfeb50474386b5f9ec0b3a83519889.

Files: samples/python/reactors/non_ideal_shock_tube.py; data/nDodecane_Reitz.yaml

Terms: Cantera BSD-3-Clause; bundled mechanism source/references must be preserved.

Environment and compute: Executed Python3.12.16 with cantera3.2.0, numpy2.5.4, matplotlib3.11.2; full environment captured in requirements-tested.txt. CPU; documentation reports roughly 98 seconds full script.

Inputs: Stoichiometric n-dodecane/O2/N2 at 1000 K and 40 atm; nDodecane_Reitz.yaml phases nDodecane_RK and nDodecane_IG; OH mass-fraction peak defines ignition delay.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare cantera#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [tested isolated setup] python -m venv VENV && VENV/bin/python -m pip install cantera==3.2.0 numpy matplotlib
  2. [executed harness] From a directory containing the pinned script as upstream.py and source.json: MPLBACKEND=Agg VENV/bin/python ../run_harness.py; harness executes upstream.py unchanged, exports sweep CSV and saves existing figures.
  3. [direct source route, numerical code tested via harness] MPLBACKEND=Agg VENV/bin/python upstream.py

Expected outputs: Printed ignition delays and ideal/real-gas profile plus temperature-sweep plots; harness should serialize curves/figures because upstream only displays them.

Comparison with the paper: First recover official example baseline (3.2 docs print real-gas 4.093e-4 s at 1000 K, 40 atm). Then match actual article oxygen concentrations, mechanism and full experiment settings before comparing published shock-tube curves; tutorial air ratio may differ from article experiments.

Execution result: Official Cantera v3.2.0 script ran unchanged under a logging/plot-export harness. At 1000 K and 40 atm, real-gas ignition delay=0.0004092691037528133 s and ideal-gas delay=0.00043326392307736766 s; real-gas value matches documented 4.093e-4 s within the declared 0.2% example tolerance. Both 16-point temperature sweeps finite and positive. This checks official example output, not a paper figure.

Execution evidence: Run receipt; Numeric output

Blockers: Article-exact input/mechanism and target curve require full-text comparison.

Next action: Run official version-pinned example, retain mechanism hash and compare its own documented baseline; then derive paper-target experiment independently.

Gaps: Official example explicitly chooses a representative mechanism; it is not claimed to be authors exact 2018 artifact.

Sources for cantera#

lammps#

LAMMPS — How does a Lennard-Jones solid melt and equilibrate in a particle simulation?

Route: Software paper example. Execution: Not run. Source confidence: medium.

Paper for lammps#

Aidan P. Thompson; H. Metin Aktulga; Richard Berger; Dan S. Bolintineanu; W. Michael Brown; Paul S. Crozier; Pieter J. in 't Veld; Axel Kohlmeyer; Stan G. Moore; Trung Dac Nguyen; Ray Shan; Mark J. Stevens; Julien Tranchida; Christian Trott; Steven J. Plimpton (2022). LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales. 10.1016/j.cpc.2021.108171.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Live software-paper metadata verified; official example input and README inspected; applied result figure not established.

Target in the paper: Thompson et al. (2022) LAMMPS software overview; official examples/melt is a teaching demonstration. No exact article figure/table linkage established.

Code and inputs for lammps#

Artifact or upstream source. Official software example supplied by LAMMPS maintainers; demonstrates the software-paper model class, not an identified published application.

Source inspected: yes. Ref: develop. Commit: 713cd85974ef9697063f039155563ff722653fe7.

Files: examples/melt/in.melt; examples/README

Terms: GPL-2.0; potential/input-specific terms separate.

Environment and compute: Pinned LAMMPS CPU executable with LJ pair style; preserve build options and numeric precision. CPU; small serial LJ demonstration.

Inputs: Bundled LJ reduced-unit FCC solid, random velocity seed 87287, NVE integration, thermo output every 50 steps, 250 steps by default.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare lammps#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [proposed setup] git clone https://github.com/lammps/lammps.git && cd lammps && git checkout 713cd85974ef9697063f039155563ff722653fe7
  2. [upstream invocation form] cd examples/melt && lmp -in in.melt

Expected outputs: log.lammps thermodynamic history; optional trajectory dumps only if enabled.

Comparison with the paper: Verify recorded thermo output, energy/temperature trends and step count against same-version reference logs; for a genuine article reproduction select a published potential/model dataset and exact observable rather than claiming melt example reproduces the overview.

Execution result: Source and setup inspected; no new execution yet.

Execution evidence: None recorded.

Blockers: No applied paper-result artifact selected yet.; LAMMPS executable not installed/qualified for this task.

Next action: Use this as a runnable tutorial scaffold; select an author-associated application artifact before promoting to reproduction.

Gaps: The 2022 software citation is not evidence that this input reproduces a published scientific figure.

Sources for lammps#

dualsphysics#

DualSPHysics — Can a 2D dam-break front match published experimental tip-position data?

Route: Paper result candidate. Execution: Example executed. Source confidence: high.

Paper for dualsphysics#

A.J.C. Crespo; J.M. Domínguez; B.D. Rogers; M. Gómez-Gesteira; S. Longshaw; R. Canelas; R. Vacondio; A. Barreiro; O. García-Feal (2015). DualSPHysics: Open-source parallel CFD solver based on Smoothed Particle Hydrodynamics (SPH). 10.1016/j.cpc.2014.10.004.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Crespo software-paper metadata verified; Koshizuka/Oka benchmark metadata also verified; official wiki explains exact validation case.

Target in the paper: 1996 experimental dam-tip data as attributed and bundled in the official DualSPHysics 01_DamBreak fixture; exact original article figure, scaling and observable alignment remain unverified. Software route also associated with Crespo et al. (2015), DOI 10.1016/j.cpc.2014.10.004.

Code and inputs for dualsphysics#

Artifact or upstream source. Official dam-break validation case bundles experimental nondimensional dam-front positions and references the 1996 experiment.

Source inspected: yes. Ref: master. Commit: ef3721a861fda961f0e2f9ec4cd317b19de99086.

Files: examples/main/01_DamBreak/CaseDambreakVal2D_Def.xml; examples/main/01_DamBreak/xCaseDambreakVal2D_linux64_CPU.sh; examples/main/01_DamBreak/EXP_X-DamTipPosition_Koshizula&Oka1996.txt

Terms: LGPL-2.1 solver; verify bundled validation-data attribution and auxiliary binary terms.

Environment and compute: Complete compatible DualSPHysics release: GenCase, CPU solver, postprocessors; Linux CPU script. CPU supported; cost depends on particle resolution.

Inputs: CaseDambreakVal2D_Def.xml and experimental tip-position text; preserve particle spacing, gravity, geometry, pressure treatment and output cadence.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare dualsphysics#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [upstream CPU route] cd examples/main/01_DamBreak && bash xCaseDambreakVal2D_linux64_CPU.sh

Expected outputs: Particle output, postprocessed dam-front position versus time, comparison plot against bundled experiment.

Comparison with the paper: Numerical discrepancy against an upstream paper-associated validation fixture. No source uncertainty/acceptance threshold supplied. Fixture scaling/observable definition was not independently verified against the original1996 article; no paper-result reproduction is claimed.

Execution result: Unchanged current author model completed the original 2-second horizon with a retained timeout/restart; initial unrestarted data cover all 14 comparison points.

Execution evidence: Author-code attempt; Run record

Blockers: Original 1996 figure, coordinate scaling and front-observable alignment need the original full-text audit.; No experimental uncertainty or acceptance tolerance supplied; reported error is not a scientific pass.

Next action: Obtain and inspect the original 1996 article; confirm fixture scaling and front definition, then establish spatial/time-step convergence and an independently justified comparison rule.

Gaps: Original 1996 figure, coordinate scaling and front-observable alignment need the original full-text audit.; No experimental uncertainty or acceptance tolerance supplied; reported error is not a scientific pass.

Sources for dualsphysics#

wntr#

WNTR — How does repairing the largest earthquake-induced leaks change water-service availability?

Route: Paper model adaptation. Execution: Example executed. Source confidence: high.

Paper for wntr#

Katherine A. Klise; Michael Bynum; Dylan Moriarty; Regan Murray (2017). A software framework for assessing the resilience of drinking water systems to disasters with an example earthquake case study. 10.1016/j.envsoft.2017.06.022.

Publication type: journal-article. Primary metadata. Metadata verification: confirmed. Full-text access: Primary article abstract/metadata and full current official earthquake notebook inspected.

Target in the paper: Klise et al. (2017), example earthquake case study. Current earthquake_tutorial.ipynb implements magnitude-6.5 event, partial repair, pressure/WSA/population metrics, but is not established as exact 2017 case.

Code and inputs for wntr#

Artifact or upstream source. Official EPA WNTR example for the same earthquake-resilience method family; exact paper-era case settings not established.

Source inspected: yes. Ref: main. Commit: 39e6b34f37d4b2d074f107d822752c989bf108aa.

Files: examples/earthquake_tutorial.ipynb; examples/networks/Net3.inp; documentation/disaster_models.rst

Terms: WNTR BSD-3-Clause; EPANET and bundled network/model notices apply.

Environment and compute: Executed Python3.12.16 with wntr1.4.0, numpy2.5.4, pandas3.1.1, scipy1.18.1, matplotlib3.11.2 and six1.17.0; full environment captured. CPU; small network, two 24-hour hydraulic runs.

Inputs: Bundled Net3 network, seed12345, epicenter(32000,15000)m, depth10000m, magnitude6.5, 24h simulation, leaks start5h and 4 largest repaired15h, PDD pressure limits3.52/14.06m.

Data access: Public repository inputs; separate component and model terms still apply.

Construct and compare wntr#

The sequence below contains upstream commands and proposed setup instructions, as labeled. Only commands identified in the execution evidence were run here; inspect the pinned source and prerequisites before running.

  1. [tested isolated setup] python -m venv VENV && VENV/bin/python -m pip install wntr==1.4.0 matplotlib scipy six
  2. [executed extraction] Concatenate every code cell of examples/earthquake_tutorial.ipynb in source order without numerical edits; keep networks/Net3.inp beside the script.
  3. [executed harness] From notebook examples directory: MPLBACKEND=Agg VENV/bin/python ../run_harness.py; executes extracted upstream.py unchanged and saves existing plots.

Expected outputs: earthquake_people_impacted.csv and earthquake_people_impacted_wrepair.csv; pressure/WSA/affected-population plots.

Comparison with the paper: Compare no-repair and repair time series against same-version repository regression fixtures first; for paper reproduction recover exact 2017 network, fragility curves and realizations and compare corresponding published metric/figure.

Execution result: All current official earthquake notebook code cells executed without numerical edits. Both 25x108 population-impacted CSV arrays exactly match current upstream regression fixtures (maximum absolute difference 0). 16 damaged pipes, 4 repaired after 15 h. Missing six dependency found on first attempt and installed only in isolated venv; failed attempt retained. This reproduces current upstream fixtures, not an established 2017 paper result.

Execution evidence: Run receipt; Numeric output; Numeric output

Blockers: Paper-era input and figure mapping not yet audited.

Next action: Execute current upstream notebook, compare existing regression fixture within stated numeric tolerance, then retrieve paper-specific scenario.

Gaps: Running current earthquake tutorial does not establish reproduction of 2017 earthquake case.; Published wheel omitted six as an installed dependency in this environment; adding six resolved the import failure.

Sources for wntr#

View source on GitHub