neqsim-trapped-liquid-fire-rupture

Fire Rupture Study

This skill coordinates two related fire-rupture workflows:

• blocked-in liquid-filled segments exposed to fire, where thermal expansion can overpressure the pipe/flange system; and
• blowdown pipe segments exposed to fire, where a supplied pressure profile is combined with wall heat-up, temperature-dependent material data, and a Sellars-Tegart strain-rate model to estimate time to rupture.

For both workflows, gather evidence, preserve assumptions and gaps, run the appropriate NeqSim safety.rupture calculation, hand off any rupture source term, and produce a professional report output.

For source-document-driven blowdown pipe-fire studies, use the governed handoff layer instead of calling the low-level solver directly: build SafetyEvidenceReference entries, assemble a PipeFireRuptureDataSource, run PipeFireRuptureStudyRunner, and report the returned PipeFireRuptureStudyHandoff. This preserves calculation readiness, standards readiness, deterministic uncertainty cases, and the post-rupture source-term handoff in one JSON-safe package.

When the blowdown pressure profile has not yet been generated, first use the dynamic flare/blowdown layer from neqsim-depressurization-mdmt: assemble LineEquipmentListEvidence and DynamicBlowdownFlareStudyDataSource, run DynamicBlowdownFlareStudyRunner, and use the returned source pressure profiles and dynamic_blowdown_flare_load_handoff.v1 as the governed basis for the pipe-fire rupture profile and flare/consequence handoff.

Use the workflow generically. Do not encode operator-specific criteria in public examples. If a private task has project-specific acceptance criteria, keep them inside the task folder and cite the private basis only in that task deliverable.

When to Use

• A Word report, technical note, or user request asks for a similar study on trapped liquid, blocked-in liquid, or no pressure relief on liquid-filled piping.
• A fire case can heat a liquid-filled segment isolated by closed valves or PBB.
• Thermal expansion may raise pressure above pipe, flange, gasket, or relief limits.
• The study needs PFP endurance, rupture time, source-term handoff, or an evidence matrix showing missing project data.
• A blowdown/depressurization pressure profile already exists and must be used to screen pipe wall heat-up, accumulated strain, rupture time, and release rate.
• P&ID, line-list, and piping-specification data must be assembled into pipe cases and reviewed by an engineer before calculation.

Skills to Load Together

• Document-retrieval tools for P&IDs, line lists, piping specs, material certificates, fire-zone/PFP documents, relief studies, and design basis documents.
• neqsim-technical-document-reading for structured extraction from Word, PDF, Excel, P&ID images, line lists, and material certificates.
• neqsim-process-safety for barrier/PFP demand and risk register context.
• neqsim-relief-flare-network when a thermal relief path or PSV/flare load is part of the mitigation.
• neqsim-consequence-analysis when the rupture source term must be carried into dispersion, fire, explosion, or QRA calculations.
• neqsim-professional-reporting for the final evidence-based report.

Evidence Retrieval Checklist

Before running the calculation, search the task references folder and any configured document backend for:

If any item is missing, do not invent it. Use a clearly labelled screening default only when the result can still support a preliminary decision, and put the gap in the assumptions/gaps register.

Extraction Schema

Technical document readers should return a block like this to the solver:

{
  "study_type": "trapped_liquid_fire_rupture",
  "segments": [
    {
      "segment_id": "TL-001",
      "line_numbers": ["..."],
      "isolation_boundary": {"upstream": "XV-...", "downstream": "XV-..."},
      "fluid": {"description": "...", "composition_source": "..."},
      "operating_pressure_bara": {"value": 10.0, "source": "..."},
      "operating_temperature_C": {"value": 25.0, "source": "..."},
      "pipe_internal_diameter_m": {"value": 0.10, "source": "..."},
      "wall_thickness_m": {"value": 0.003, "source": "..."},
      "exposed_length_m": {"value": 10.0, "source": "..."},
      "material_grade": {"value": "API 5L X52", "source": "..."},
      "flange_class": {"value": 900, "source": "..."},
      "fire_basis": {"type": "api521_pool_fire", "heat_flux_W_m2": null, "source": "..."},
      "pfp_requirement_s": {"value": 1800.0, "source": "..."},
      "relief": {"available": false, "set_pressure_bara": null, "source": "..."},
      "acceptance_criteria": [{"criterion": "...", "source": "..."}],
      "evidence_gaps": ["..."]
    }
  ]
}

Every numeric field should preserve original value, original unit, converted SI value, source document, page/sheet, and confidence when available.

For blowdown pipe fire rupture, technical document readers or source-document agents should return a block like this:

{
  "study_type": "blowdown_pipe_fire_rupture",
  "segment_id": "BD-001",
  "pressure_profile": {
    "time_unit": "minute",
    "pressure_unit": "bara",
    "basis": "absolute pressure profile from governed blowdown calculation",
    "points": [[0.0, 61.3], [0.083333333, 59.7053]]
  },
  "pipes": [
    {
      "pipe_id": "3DD100",
      "pipe_class": "DD100",
      "nps_in": 3.0,
      "outside_diameter_mm": 88.9,
      "wall_thickness_mm": 3.7,
      "corrosion_allowance_mm": 0.0,
      "wall_undertolerance_fraction": 0.125,
      "weld_factor": 1.0,
      "material": "22Cr duplex",
      "fluid_density_kg_m3": 23.75,
      "fluid_heat_capacity_J_kgK": 2283.35,
      "gas_molecular_weight_kg_kmol": 18.2,
      "initial_temperature_C": 20.0,
      "exposed_length_m": 1.0,
      "source": "reviewed piping specification / workbook input"
    }
  ],
  "fire_scenarios": ["Small jet fire 250 kW/m2", "Pool fire 250 kW/m2", "Large jet fire 350 kW/m2"],
  "evidence_gaps": []
}

Governed Source-Document Handoff Schema

When pipe data comes from source drawings and piping-specification rows, normalize it into a source-traceable package before NeqSim calculation:

{
  "schemaVersion": "pipe_fire_rupture_data_source.v1",
  "studyId": "BD-001",
  "input": {"segmentId": "3DD100", "evidenceReferences": []},
  "material": {"materialName": "22Cr duplex"},
  "scenario": {"name": "Large jet fire 350 kW/m2"},
  "pressureProfile": {"pressureUnit": "bara", "timeUnit": "seconds"},
  "pidTopologyEvidence": {
    "schemaVersion": "pid_topology_evidence.v1",
    "drawingId": "P-ID-001",
    "revision": "A",
    "simulationReady": false,
    "boundaryVerified": false,
    "nodes": [],
    "edges": [],
    "missingTags": []
  },
  "sourceDocumentEvidence": [],
  "pipingSpecificationEvidence": [],
  "processEvidence": [],
  "fireScenarioEvidence": [],
  "sourceDiagramsReviewed": true,
  "pidTopologyVerified": false,
  "pipingSpecificationRowsReviewed": true,
  "materialCertificateReviewed": false,
  "blowdownProfileVerified": true,
  "fireScenarioReviewed": true,
  "standardsReviewed": false,
  "humanReviewRequired": true,
  "readiness": {"verdict": "SCREENING"}
}

Readiness semantics:

• NOT_READY: missing calculation-critical input (input, material, scenario, or pressure profile). Do not run the calculation.
• SCREENING: calculation may run, but evidence gaps or unreviewed assumptions prevent design-grade use.
• DESIGN_GRADE: controlled source drawing, piping-specification, material, fire, and depressurization evidence has been reviewed and the package is ready for formal engineering review.

Quick Screening: Blocked-In Liquid Thermal Expansion (No Fire)

Before committing to a full fire-exposure rupture study, BlockedInLiquidExpansionAnalysis (neqsim.process.util.fire) provides a fast, fire-independent check of pure thermal-expansion overpressure for a blocked-in liquid segment, per API 521 §4.4.12:

SystemInterface oil = new SystemSrkEos(298.15, 10.0);
oil.addComponent("n-heptane", 100.0);
oil.setMixingRule("classic");

double[] temperaturesK = {298.15, 308.15, 318.15, 328.15, 333.15};
double[] pressuresPa =
    BlockedInLiquidExpansionAnalysis.computeIsochoricPressureProfile(oil, temperaturesK);

// Fast cross-check using the simplified API 521 beta/kappa relation
double beta = BlockedInLiquidExpansionAnalysis.estimateThermalExpansionCoefficient(oil, 1.0);
double kappa = BlockedInLiquidExpansionAnalysis.estimateIsothermalCompressibility(oil, 1.0e5);
double deltaPPa = BlockedInLiquidExpansionAnalysis.simplifiedPressureRise(beta, kappa, 35.0);

Use this when the task only needs a thermal-expansion magnitude (e.g. deciding whether a thermal relief valve is needed) without a fire scenario, material derating, or flange/pipe-stress checks. See docs/safety/blocked_in_liquid_thermal_expansion.md for the full reference.

Java Calculation Pattern

SystemInterface oil = new SystemSrkEos(298.15, 10.0);
oil.addComponent("n-heptane", 100.0);
oil.setMixingRule("classic");

InventoryResult inventory = new TrappedInventoryCalculator()
    .setFluid(oil)
    .setOperatingConditions(10.0, "bara", 25.0, "C")
    .addPipeSegment("TL-001", 0.10, 10.0, 1.0, null)
    .calculate();

TrappedLiquidFireRuptureResult result = TrappedLiquidFireRuptureStudy.builder()
    .segmentId("TL-001")
    .fluid(oil)
    .inventory(inventory)
    .pipeGeometry(0.10, "m", 3.0, "mm", 10.0, "m")
    .api5lMaterial("X52")
    .fireScenario(FireExposureScenario.api521PoolFire(3.4, 1.0))
    .flangeClass(900)
    .timeControls(1800.0, 2.0)
    .build()
    .run();

Key classes:

• MaterialStrengthCurve: ambient SMYS/SMTS plus temperature derating.
• FireExposureScenario: API 521 pool fire, fixed heat flux, or radiative fire.
• TrappedLiquidFireRuptureStudy: transient pressure, wall temperature, pipe stress, flange rating, vapor-pocket, relief-set, and rupture checks.
• TrappedLiquidFireRuptureResult: event times, time histories, JSON map, PFP demand, and source-term handoff.

For supplied-pressure-profile blowdown pipe fire rupture:

BlowdownPressureProfile profile = BlowdownPressureProfile.fromMinutesAndBara(
  new double[] {0.0, 0.083333333, 0.166666667},
  new double[] {61.3, 59.7053, 58.6349});

PipeFireRuptureInput pipe = PipeFireRuptureInput.builder("3DD100")
  .pipeClass("DD100")
  .nominalDiameterInches(3.0)
  .outsideDiameter(88.9, "mm")
  .nominalWallThickness(3.7, "mm")
  .corrosionAllowance(0.0, "mm")
  .wallThicknessUndertoleranceFraction(0.125)
  .weldFactor(1.0)
  .fluidDensityKgPerM3(23.75)
  .fluidHeatCapacityJPerKgK(2283.35)
  .gasMolecularWeightKgPerKmol(18.2)
  .initialTemperatureC(20.0)
  .exposedLength(1.0, "m")
  .build();

PipeFireRuptureResult pipeResult = PipeFireRuptureStudy
  .builder(pipe, PipeFireRuptureMaterial.fromSpreadsheetMaterialName("22Cr duplex"),
    PipeFireRuptureScenario.spreadsheetLargeJetFire(), profile)
  .timeStepSeconds(5.0)
  .maxTimeSeconds(1800.0)
  .build()
  .run();

For governed agentic studies, prefer the runner pattern:

SafetyEvidenceReference pipingSpecWall = SafetyEvidenceReference
  .builder("PIPING_SPEC", "nominal_wall_thickness_mm")
  .documentId("pipe-class=DD100;rev=D")
  .valueText("3.7")
  .unit("mm")
  .status("fetched_joined")
  .confidence(0.95)
  .build();

PipeFireRuptureInput governedPipe = pipe.toBuilder()
  .evidenceReference(pipingSpecWall)
  .build();

PipeFireRuptureDataSource dataSource = PipeFireRuptureDataSource.builder("BD-001")
  .input(governedPipe)
  .material(PipeFireRuptureMaterial.fromSpreadsheetMaterialName("22Cr duplex"))
  .scenario(PipeFireRuptureScenario.spreadsheetLargeJetFire())
  .pressureProfile(profile)
  .addPipingSpecificationEvidence(pipingSpecWall)
  .sourceDiagramsReviewed(true)
  .pidTopologyVerified(false)
  .pipingSpecificationRowsReviewed(true)
  .materialCertificateReviewed(false)
  .blowdownProfileVerified(true)
  .fireScenarioReviewed(true)
  .standardsReviewed(false)
  .build();

PipeFireRuptureStudyHandoff handoff = PipeFireRuptureStudyRunner.builder()
  .timeStepSeconds(5.0)
  .maxTimeSeconds(1800.0)
  .runUncertainty(true)
  .build()
  .run(dataSource);

Key pipe-fire classes:

• BlowdownPressureProfile: absolute pressure profile with exact tabulated-point lookup, step or linear mode, and barg conversion.
• PipeFireRuptureInput: one pipe case with geometry, wall allowance, fluid, and exposed-length data.
• PipeFireRuptureMaterial: workbook-style material curves for 22Cr duplex, SS316, CS235, CS360/API 5L-X52, superduplex, and 6Mo.
• PipeFireRuptureScenario: small jet, pool fire, large jet, and custom radiative plus convective fire exposure.
• PipeFireRuptureStudy: heat-up, thick-wall stress, Sellars-Tegart strain rate, accumulated strain, rupture event, and screening release estimate.
• PipeFireRuptureResult: time series, rupture summary, warnings, recommendations, release estimate, and JSON map.
• SafetyEvidenceReference: compact source reference for source drawings, piping specifications, process, fire, and material inputs.
• SafetyStudyReadiness: NOT_READY / SCREENING / DESIGN_GRADE verdict with findings and actions.
• PidTopologyEvidence: typed P&ID topology graph, boundary status, missing-tag register, and drawing-overlay readiness.
• PipeFireRuptureDataSource: governed data-source package binding inputs to evidence and review flags.
• PipeFireRuptureStudyRunner: readiness-gated orchestration of solver, standards check, uncertainty, and source-term handoff.
• PipeFireRuptureStudyHandoff: versioned package containing data source, readiness, result, uncertainty, and source term.
• PipeFireRuptureStandardsValidator: API 521 / ISO 23251 / NORSOK S-001 / piping-specification evidence-quality gate.
• PipeFireRuptureUncertaintyRunner: deterministic one-at-a-time perturbation screening of wall, corrosion, heat-flux, and initial-temperature assumptions.
• LineEquipmentListEvidence, DynamicBlowdownFlareStudyDataSource, DynamicBlowdownFlareStudyRunner, and DynamicBlowdownFlareStudyHandoff: governed dynamic depressurization, PSV, and flare-load setup used to create a source-traceable pressure profile before pipe-fire rupture screening.

Reusable Safety Report Template

For governed source-document pipe-fire studies, the Word/HTML report should use a repeatable evidence-first structure. At minimum include:

1. Executive verdict with NOT_READY, SCREENING, or DESIGN_GRADE, plus the human-review status.
2. Evidence matrix with source system, document id, revision, page/sheet, field, extracted value, unit, status, confidence, and notes.
3. Source drawing table with drawing id, revision, embedded-text/OCR status, topology nodes/edges count, missing tags, and overlay/annotation link.
4. Piping specification table with applicable revision, class/material references, NPS, outside diameter, wall thickness, corrosion allowance, undertolerance, and row-review status.
5. NeqSim input lineage mapping each solver input to its SafetyEvidenceReference.
6. Standards-applied table covering API 521 / ISO 23251, NORSOK S-001, piping/material basis, and consequence/source-term handoff status.
7. Assumptions and gaps register with severity, effect on result, and required action before design use.
8. Calculation results and uncertainty including rupture time, rupture pressure, wall temperature, release estimate, deterministic perturbation cases, and P10/P50/P90 where available.
9. Source-term handoff using pipe_fire_rupture_source_term_handoff.v1 when rupture is predicted.
10. Calculation lineage from document field to NeqSim input to reported result.

Before the report is considered complete, apply these hard QA gates:

• The executive summary and problem description must be populated from the current task_spec.md and results.json; no placeholder text may remain.
• The report front page must state the readiness label (NOT_READY, SCREENING, or DESIGN_GRADE) and whether design-grade use is blocked.
• The method in task_spec.md, source scripts, generated results.json, and report must agree on the pressure-profile basis. If the task moved from a reconstructed profile to a direct dynamic NeqSim profile, update all four.
• Script-backed studies are acceptable when study_config.yaml explicitly sets notebooks to not required; do not create false notebook execution warnings.
• analysis.md and neqsim_improvements.md must be filled for safety-critical workflow/code gaps, including reporting, evidence-readiness, pressure-profile export, plant-data evidence, and governed handoff gaps.
• The evidence gaps/design blockers and recommendations must appear before or alongside the conclusions, not only in appendix-style detail.

Results to Save

Save a trapped_liquid_fire_rupture section in results.json:

{
  "trapped_liquid_fire_rupture": {
    "segments": [
      {
        "segment_id": "TL-001",
        "limiting_failure_mode": "PIPE_RUPTURE",
        "time_to_pipe_rupture_s": 420.0,
        "time_to_flange_failure_s": null,
        "minimum_failure_time_s": 420.0,
        "final_pressure_bara": 145.0,
        "final_wall_temperature_C": 530.0,
        "pfp_required_endurance_s": 1800.0,
        "pfp_margin_s": -1380.0,
        "evidence_gaps": ["Material certificate not found"],
        "recommendations": ["Provide thermal relief or documented PFP upgrade"]
      }
    ],
    "standards_applied": ["API 521", "ISO 23251", "ASME B31.3", "ASME B16.5"],
    "assumptions": ["Generic API 5L X52 screening curve used pending certificate"],
    "evidence_matrix": []
  }
}

For Standard/Comprehensive studies, include:

• Segment summary table with event times and limiting mode.
• Evidence matrix with document, page/sheet, extracted value, confidence, and gap status.
• Assumptions/gaps register ranked by impact.
• Time histories for pressure, wall temperature, material allowable stress, and flange rating.
• PFP demand table from toPassiveFireProtectionDemand(...).
• Source-term handoff from createRuptureSourceTerm(...) if rupture is predicted.
• Risk register using neqsim-process-safety when consequences are material.

For blowdown pipe fire rupture, save a pipe_fire_rupture section in results.json:

{
  "pipe_fire_rupture": {
    "segment_id": "BD-001",
    "pressure_profile_basis": "absolute bara profile from governed blowdown model",
    "pipes": [
      {
        "pipe_id": "3DD100",
        "fire_scenario": "Large jet fire 350 kW/m2",
        "rupture_predicted": true,
        "time_to_rupture_s": 110.0,
        "rupture_pressure_barg": 32.03,
        "rupture_wall_temperature_C": 760.0,
        "release_estimate_kg_s": 21.9,
        "evidence_gaps": []
      }
    ],
    "standards_applied": ["API 521", "ASME B31.3"],
    "assumptions": ["Spreadsheet material curve used pending certificate review"]
  }
}

For governed studies, also persist the runner handoff:

{
  "pipe_fire_rupture_handoff": {
    "schemaVersion": "pipe_fire_rupture_study_handoff.v1",
    "calculationReadiness": {"verdict": "SCREENING"},
    "standardsReadiness": {"verdict": "SCREENING"},
    "result": {},
    "uncertainty": {"schemaVersion": "pipe_fire_rupture_uncertainty.v1"},
    "sourceTermHandoff": {"schemaVersion": "pipe_fire_rupture_source_term_handoff.v1"}
  }
}

Validation and Benchmarking

• Hand-check pressure rise using deltaP = bulk_modulus * alpha * deltaT.
• Compare API 521 heat flux or heat input against an independent spreadsheet or standard example.
• Verify material ambient strength against pipe specification or certificate.
• For pipe-fire studies, benchmark one representative case against the source workbook or an independent spreadsheet before scaling to all pipe cases.
• Verify pressure-profile absolute/gauge convention. The workbook-style stress and release calculations use barg, while the pressure profile is often supplied as bara and converted by subtracting 1 bar.
• For high-consequence segments, treat the NeqSim result as screening and recommend specialist flange/gasket assessment, FEA, or consequence modelling as needed.

Common Mistakes

Related Documentation

• docs/safety/trapped_liquid_fire_rupture.md
• docs/safety/trapped_inventory_calculator.md