Capability & Workflow Guide

This application is organized around the question you are trying to answer — not around the architecture of the math. Use the simplest valid model that answers your question, and escalate only when the question requires it. The guide explains why each escalation is necessary.

What are you trying to determine?

Pick your objective — the guide shows the minimum workflow required. You are never forced to start at the bottom of the modeling hierarchy.

Metric Finder

Search for the metric you need. The finder shows the primary tool, required inputs, the optional experimental route, and whether a full simulation is required.

Filter EvaluationFull simulation requiredCalibration
Evaluate a filter dynamically against a scenario using the canonical containment engine (generation → inventory → pressure → filter sink).
Required inputs: Scenario (generation source + container), Filter definition
Outputs: Pressure vs time, Peak pressure, Species inventory, Filter removal rate, No-filter vs filter comparison, Calculation Trace, Warnings
Data: A scenario (generation source + container) and a filter definition.
Filter Design & EvaluationFull simulation requiredCalibration
A guided workflow that connects filter geometry, measured flow/ΔP data, particulate retention, gas transport, structural requirements, and scenario evaluation into one cohesive process. Enter at any step if you already have data.
Required inputs: Filter geometry (nominal/active diameter, thickness), Application requirements (flow, ΔP, particulate, transport)
Outputs: Filter characterization status, Design-point PASS/NEAR/EXCEEDS, Transport properties, Scenario evaluation, Engineering report
Data:
RadiolysisFull simulation required
Model radiolytic hydrogen generation from absorbed dose / activity / G-values and route it through containment.
Required inputs: Radiation source (activity / dose rate), G-value, Container free volume
Outputs: Radiolytic H2 generation rate, Pressure vs time, Peak pressure, Calculation Trace, Report
Data: Radiation source (activity or dose rate) and G-value.
Battery Incident ModelingFull simulation required
Model lithium-ion battery thermal-runaway / post-fire gas generation across multi-phase chemistry and containment.
Required inputs: Battery chemistry, Incident profile, Container
Outputs: Gas generation rate, Aqueous generation, Phase inventory, Pressure vs time, Peak pressure, Battery report
Data: Battery chemistry, incident profile, and container.
Membrane Pressure-Decay CalibrationFull simulation requiredCalibration
Fit a membrane transport property (permeance / permeability) from upstream pressure-decay experimental data.
Required inputs: Membrane geometry, Pressure-decay time series, Temperature
Outputs: Fitted permeance / permeability, Fit quality (R², RMSE), Identifiability, CalibratedTransportProfile, Calculation Trace
Experimental route: import data → calibrate/validate
Data: Pressure-decay time series + membrane geometry.
Time-Lag Diffusivity & SolubilityFull simulation requiredCalibration
Extract time lag θ from cumulative permeation data and derive diffusivity D = L²/(6θ) and solubility S = P/D.
Required inputs: Cumulative permeation Q(t) data, Membrane thickness, Temperature
Outputs: Time lag θ, Diffusivity D, Solubility S, Steady-state flux, Permeance / permeability, Uncertainty, Calculation Trace
Experimental route: import data → calibrate/validate
Data: Cumulative permeation Q(t) data + membrane thickness.
Multicomponent Mixture PermeationFull simulation requiredCalibration
Simulate species-specific permeation of a gas mixture using independent-component partial-pressure driving force.
Required inputs: Gas composition (mole fractions), Membrane geometry, Temperature, Initial pressure
Outputs: Per-species flux, Composition evolution, Pressure decay, Mass-balance check, Identifiability, Calculation Trace
Experimental route: import data → calibrate/validate
Data: Gas composition + per-species permeance.
Surrogate Gas Translation (He → H₂)Direct calculationCalibration
Translate helium experimental transport data to hydrogen (experimental-ratio or solution-diffusion method).
Required inputs: He experimental profile, Translation method
Outputs: Translated H2 permeance / permeability, Translation ratio, Comparability status, Propagated uncertainty, MODEL_TRANSLATED evidence, Calculation Trace
Experimental route: import data → calibrate/validate
Data: He experimental profile (+ H2 or D/S profiles depending on method).
Design OptimizationDirect calculation
Search a design space under constraints and objectives to identify the best design and Pareto frontier.
Required inputs: Design space, Objective, Constraints
Outputs: Optimal configuration, Pareto frontier, Constraint margins, Optimization record
Data: Design space + objective + constraints.
Study Library & ReproductionDirect calculation
Browse, reproduce, and verify saved studies and systems with their decision-quality and uncertainty metadata.
Required inputs: A saved study or system
Outputs: Study detail, Reproduction verification, Decision-quality metadata, Uncertainty metadata
Data: A saved study or system.
Gas GenerationFull simulation required
Use this to define how hydrogen is produced — stoichiometric, rate-based, or electrochemical (Faradaic).
Required inputs: Reaction type & equation or current
Outputs: H₂ production rate, Total moles generated, Faradaic yield
Data: Optional — timeseries current/temperature can be imported.
Peak Pressure / Pressure vs TimeFull simulation requiredCalibration
Use this to find peak pressure, time to peak, pressure at a specific time, or the pressure response to a generation profile.
Required inputs: Gas generation profile, System volume, Temperature, Initial conditions
Outputs: Peak pressure, Time of peak, Pressure vs time, Gas inventory at peak, Membrane state at peak
Experimental route: import data → calibrate/validate
Data: Optional — experimental pressure curves can validate.

How much model do I need?

Answer a few questions. The assistant recommends the simplest valid model — and explains why if a more sophisticated model becomes necessary.

Are conditions changing over time?
e.g. gas is being generated while pressure rises.
Is there more than one gas species?
Composition gradients or competing transport.
Is adsorption by the membrane important?
Gas retention affects inventory & pressure.
Is transport through the membrane important?
Gas moves across the membrane.
Do you have experimental data?
Measured flux, pressure, or isotherm data.
Are you trying to estimate a parameter from data?
Calibration rather than a direct calculation.
Do you need peak conditions?
Peak pressure or time to peak.
Do you need a reportable / reproducible result?
A technical report or provenance record.
RecommendationSimple
Surface Area & Geometry
Use this to compute geometric area, effective area, area per mass, or area per volume from membrane dimensions.
Open
→A geometric/structural calculation answers this directly.

The Full Model Workflow

The complete pipeline — but you do not have to run all of it. Click any stage to see what it needs and whether it can be used independently.

Membrane— Geometry / Surface Area

Defines the physical membrane: dimensions, density, porosity, tortuosity, pore size.

Usable independently
Inputs: Dimensions, Density, Porosity, Tortuosity
Outputs: Geometric & effective area, Mass, Volume
Depends on: None — foundational
Experimental: Optional — measured porosity/density improves accuracy.

Feature Matrix

Generated from the live capability registry — it stays accurate as features are added. Click a row to open the tool.

MetricPrimary ToolOptional ModelsFull Sim?CalibrationExperimental
filter evaluationFilter EvaluationTransport property resolverYesYes—
filter designFilter Design & EvaluationContainment engine, Transport property resolverYesYes—
radiolytic h2RadiolysisFilter transportYes——
batteryBattery Incident ModelingFilter transport, Battery reportingYes——
permeanceMembrane Pressure-Decay CalibrationTransport property resolverYesYesYes
time lagTime-Lag Diffusivity & SolubilityUncertainty propagationYesYesYes
mixtureMulticomponent Mixture PermeationInverse fitYesYesYes
surrogateSurrogate Gas Translation (He → H₂)Uncertainty propagationNoYesYes
optimizeDesign OptimizationConstraint engineNo——
reproduceStudy Library & ReproductionDecision evidenceNo——
generation rateGas GenerationElectrochemistry (Faraday)Yes——
peak pressurePeak Pressure / Pressure vs TimeAdsorption, Transport, DegradationYesYesYes
gas inventoryGas Inventory & AccumulationAdsorption, TransportYes—Yes
filter countFilter SizingTransient modelNo——
compare filtersFilter ComparisonSizing modelNo——
compare membranesMembrane / Filter ComparisonPerformance modelNo——
scenario comparisonScenario ComparisonTransient modelNo——
optimal configurationOptimizationSweep engineNo——
sensitivity to parameterSensitivity AnalysisCoupled modelsNo——
diffusion coefficientDiffusion / Transport CoefficientPorous-media correctionNoYesYes
multicomponent fluxMulticomponent / Competing TransportAdsorption couplingNoYesYes
geometric areaSurface Area & GeometryPorosity correctionNo——
equivalent areaSurface-Area EquivalencyDensity modelNo——
porosity effectsMembrane Structure EffectsPorosity/tortuosity correctionNo——
adsorption capacityAdsorption Capacity (Static)Isotherm fitNoYesYes
adsorption rateAdsorption vs TimeTransport couplingYesYesYes
experimental fluxFilter Flux & Sizing Lab (Experimental)Transient mass balanceYesYesYes
calibrated parameterModel CalibrationOptimizationNoYesYes
experiment vs modelExperimental ValidationSimulationYes—Yes
technical reportTechnical ReportSnapshotNo——
reference packageReference Packages—No——
design spaceDesign Space ExplorationConstraint engineNo——
verificationCalculation Verification Center—No——
uncertaintyDecision Quality & UncertaintyUncertainty registryNo——

Capabilities

Every feature with its 'why use', 'when not to use', required/optional inputs, outputs, minimum path, and what to do next. Expand any card.

Membrane Geometry
Membrane Structure
Adsorption
Transport
Gas / Pressure
Performance
Experimental Analysis
Reporting
Cross-Cutting

Provenance Before Calibration

Establish the Data Profile and provenance record BEFORE experimental data is used for calibration. Calibration transforms observations into parameters — both states must be preserved.

Before → After calibration
Before
Diffusion coefficient: 0.034 cm²/s
Source: User-defined estimate
↓ calibration
After (new version)
Diffusion coefficient: 0.0278 cm²/s
Source: Calibrated from Diffusion Test #007

The original value is never overwritten. A new model version is created. The system can answer "What data produced this coefficient?" with: Data Profile ID, original dataset, test conditions, model used, calibration method, original & calibrated parameters, fit metrics, version, and timestamp.

Prevent Circular Validation

Data used for calibration must not be presented as independent validation data.

Data Profile #007 was used during calibration. Agreement with this dataset does not constitute independent validation.

This warning is shown both in the UI and in generated reports whenever a calibration dataset is reused as validation data.

Calibration Dataset≠Independent Validation Dataset

User Modes

The same underlying calculation engine is used in every mode. Modes only change the amount of complexity exposed to you.

Quick Calculationmode

For users who need one metric. Opens the direct tool for the question — no coupled models, no simulation.

Guided Analysismode

For users who want help selecting the appropriate workflow. Uses the objective browser and decision assistant.

Full Modelingmode

For users who want the complete coupled time-dependent simulation (generation → adsorption → transport → pressure).

Experimental / Calibrationmode

For users working with measured data — import, calibrate, validate, with provenance enforced.

Expertmode

For users who want direct control over model selection and all parameters.

Model Selection Guidance

Higher model complexity is not automatically better. Pick the tier that matches your problem.

Simpletier
When: One gas · static conditions · no time dependence
Use: Simplified diffusion / adsorption / geometric model.
Intermediatetier
When: One or more gases · changing pressure · porous membrane · time-dependent
Use: Coupled transport + time integration.
Advancedtier
When: Multiple gases · competing diffusion · adsorption/desorption · experimental calibration
Use: Maxwell-Stefan + sorption-coupled simulation.
Governing Principle

The user should never have to understand the architecture of the application to use it. The architecture exists to support the user's question. The question determines the workflow. The workflow determines the required model complexity. The model produces the result. The provenance system explains where the result came from. A simple answer stays simple; a complex question has access to the full engine.

QuestionDirect calculationTargeted modelCoupled modelTime-dependent simulationExperimental calibrationAdvanced analysis