neqsim-standards-lookup

# NeqSim Standards Lookup Reference for identifying, applying, and documenting industry standards compliance in every engineering task. All tasks — Quick, Standard, or Comprehensive — must identify applicable standards proportional to task depth. ## Standards Identification (MANDATORY First Step) Before any simulation or analysis, identify applicable standards: | Task Scale | Standards Requirement | |------------|----------------------| | **Quick** | 1-line note: "Per \[STANDARD\]" or "N/A — property lookup" | | **Standard** | Table of applicable standards with scope and status | | **Comprehensive** | Full table with clause numbers, design values, and compliance evidence | ## Search the Local Standards Library First (MANDATORY) Before paraphrasing a standard from memory, citing it second-hand, or declaring it unobtainable, search the user's configured document root. It holds full PDFs and is more authoritative than recall: ```bash neqsim --show-document-root # may be unset - then there is no library neqsim documents "API 521" # recursive; every word must appear in the path ``` Words may be in any order and need not be adjacent (`"API 521"` finds `API STD 521 ...`), but a standard filed under another body's number (for example `DNV-ST-F101`) is only found by that exact designation. When several editions are present (for example two `TR3500` versions), cite the edition you actually read. Every task folder also carries the listing at `step1_scope_and_research/references/document_root_index.md` — read it, since a subagent may have no CLI access. Refresh it with `neqsim documents --index .`. - **Cite clause numbers from the PDF**, not from memory. A remembered limit that turns out to be the previous edition is a silent compliance error. - **File names are indicative only.** A document can contain the standard you need without naming it (company guidelines often embed API/NORSOK requirements), so open likely candidates before concluding it is absent. - Copy the standards actually used into `references/literature/` (or `manual/` for company documents) so the task folder stays self-contained, and cite the document number in `standards_applied`. - Only after this search comes up empty may a standard be recorded as a `data_gap`. When no document root is configured, say so explicitly in the gap entry rather than implying the standard does not exist. ## Equipment → Standards Mapping NeqSim's standards database is in `src/main/resources/designdata/standards/`. The index file `standards_index.csv` maps equipment types to applicable standards: | Equipment Type | Primary Standards | NeqSim CSV File | |----------------|-------------------|-----------------| | Separator, ThreePhaseSeparator, GasScrubber | NORSOK P-001, API 12J, ASME VIII | `norsok_standards.csv`, `api_standards.csv`, `asme_standards.csv` | | Compressor | API 617, NORSOK P-002 | `api_standards.csv`, `norsok_standards.csv` | | Pump | API 610 | `api_standards.csv` | | Pipeline, AdiabaticPipe, MultiphasePipe | NORSOK L-001, ASME B31.3/B31.4/B31.8, DNV-ST-F101 | `norsok_standards.csv`, `asme_standards.csv`, `dnv_iso_en_standards.csv` | | Pipeline, FlexiblePipe, Cable, Umbilical on seabed | DNV-RP-F109 | Typed kernel `DnvRpF109OnBottomStabilityKernel`; project values remain explicit inputs | | HeatExchanger, Heater, Cooler | API 660/661, TEMA | `api_standards.csv` | | Tank | API 650/620, API 2000 | Use `Api2000TankVentingScreeningKernel` for current 7th-edition caller-controlled normal/emergency demand and rated-capacity screening; `api_standards.csv` covers catalog data | | Valve | ASME B31.3 | `asme_standards.csv` | | Subsea equipment | NORSOK U-001, DNV-ST-F101 | `norsok_standards.csv`, `subsea_standards.csv` | | Well casing/tubing | API 5CT, API TR 5C3, NORSOK D-010 | `api_standards.csv`, `norsok_standards.csv` | | Flange | ASME B16.5 | `asme_standards.csv` | | Process flow diagram, block flow diagram, graphical symbols (documentation) | ISO 10628-1:2014, ISO 10628-2:2012 (collective application of ISO 14617); NORSOK P-002:2023+AC:2024 sets no diagram requirement | Not in the CSV tables; texts are paywalled, so claim "conventions applied, conformance not claimed". Edition evidence and checks: `process-study-report` skill, `references/standards_verification.md` | | Orifice plate / differential-pressure metering | ISO 5167-1/-2, AGA 3 / API MPMS 14.3 | Use `Iso5167OrificeMeteringKernel` for strict ISO 5167-2:2022 screening; keep `Standard_AGA3` for an AGA/API basis and `Orifice` for process simulation | | CO2 corrosion / materials selection | NORSOK M-506, ISO 15156 / NACE MR0175, NORSOK M-001 | Use `NorsokM506CorrosionDesignKernel` for strict M-506 screening; use `NorsokM506ElectrolyteBridge` for a rigorous brine pH/FeCO3 basis and keep `NorsokM506CorrosionRate` for legacy sweeps | | Offshore steel fatigue | DNV-RP-C203 | Use `DnvRpC203FatigueDesignKernel` with a verified project-controlled S-N curve and stress spectrum; do not report legacy pipeline/riser shortcuts as exact-edition C203 evidence | | Submarine-pipeline free spans | DNV-RP-F105 | Use `DnvRpF105FreeSpanScreeningKernel` for the current 2025-12 first-mode/dimensionless screen; project response triggers are not DNV acceptance criteria and the legacy allowable-span calculator is not F105 evidence | | Inspected pipeline metal loss | DNV-RP-F101 | Use `DnvRpF101CorrodedPipelineScreeningKernel` for the current isolated longitudinal defect/internal-pressure screen; measured geometry and caller-controlled factors are evidence, not values inferred from M-506, and ST-F101 design checks remain separate | | CO2 pipeline design and operation | DNV-RP-F104 | Use `DnvRpF104Co2PipelineEnvelopeScreeningKernel` for the current caller-controlled composition and operating-envelope margin screen; use its requirement pack only for bounded capability discovery and keep ST-F101, fracture, materials/corrosion, construction, operation, safety, and requalification separate | | Submarine-pipeline global buckling | DNV-RP-F110 | Use `DnvRpF110GlobalBucklingResponseScreeningKernel` for the current caller-controlled external-analysis force/strain/displacement/feed-in response envelope; keep structural response, soil springs, critical buckling, imperfections/triggers, local capacity, and F109/F114/F105/ST-F101 acceptance separate | | Submarine-pipeline pipe-soil interaction | DNV-RP-F114 | Use `DnvRpF114PipeSoilInteractionScreeningKernel` for the current caller-controlled vertical/axial/lateral demand-resistance envelope; keep geotechnical model derivation and F109/F110/F105/ST-F101 acceptance separate | | Fixed-roof tank venting | API 2000 | Use `Api2000TankVentingScreeningKernel` for the current 7th-edition caller-controlled demand/capacity screen; do not infer licensed demand factors or report it as device sizing/conformity | | Mineral scale / produced water | (industry practice; Davies + Ksp(T)) | `ElectrolyteScaleCalculator` / `ScaleKinetics` / `BrineMixingScaleEvaluator` (`process.chemistry.scale`) | ### ISO 18453 / GERG-water execution rule For a natural-gas **water dew point** request with a named **ISO 18453:2004** or **GERG-water** basis, use this setup and retain its model basis: 1. Prefer `neqsim.standards.gasquality.Standard_ISO18453` for a water dew point at a specified pressure. Its constructor sets up `SystemGERGwaterEos` and `setMixingRule(8)`; explicitly call `setPressure(pressureBara)` before `calculate()`, then read `getValue("dewPointTemperature", "C")`. Do not assume the feed pressure replaces the wrapper's reference pressure. 2. For a direct calculation, construct `SystemGERGwaterEos(temperatureK, pressureBara)`, add the composition, call `setMixingRule(8)`, initialize, then create `ThermodynamicOperations` for that fluid and call `waterDewPointTemperatureFlash()`. Use a clone when the original process fluid must be preserved. Do not substitute GERG-2008, ordinary PR with `"classic"`, or CPA for an ISO 18453 model basis. 3. Require positive finite water mole fraction and positive finite absolute pressure. State whether moisture is on a wet-gas or dry-gas basis. For wet-gas ppmv, `y_water = ppmv * 1e-6`; scale dry-component fractions by `1 - y_water` and verify the complete wet composition sums to one. Convert dry-basis moisture before using it as a wet-gas mole fraction. For mass per volume, retain the specified reference temperature, pressure, and dry/real-gas basis. 4. Below **273.16 K**, ISO 18453 uses an **ice reference**, not supercooled liquid water. CPA liquid-water and hydrate boundaries are different calculations. Reject non-finite output and report convergence failures. 5. Report pressure, water basis, composition, NeqSim version, model and mixing rule. Check the standard's composition limits: the validated range is **−15 to +5 °C and 5–100 bara**; the extended range is **−50 to +40 °C and 1–300 bara**, with unknown correlation uncertainty. A converged result or a class name does not establish full ISO compliance. Use the [maintained executable example and validation boundaries](../../../docs/standards/dew_point_standards.md) and route the task to the [gas-quality agent](../../agents/gas-quality.agent.md). `Draft_ISO18453` remains the legacy contract path; new direct calculations use `Standard_ISO18453`. Keep hydrocarbon dew point and cricondentherm separate. ### DNV-RP-F109 implementation status `EquipmentDesignKernelRegistry.lookup(StandardType.DNV_RP_F109)` exposes a `SCREENING` kernel for the exact catalogued edition `2021-05+AMD 2025-09`. It calculates vertical equilibrium and a transparent absolute-static lateral screen, or checks displacement supplied by an externally validated generalized or dynamic response model. It intentionally excludes licensed generalized-design tables, response generation, environmental-statistics derivation, soil-model qualification, and conformity assessment. Do not report a passing kernel result as DNV certification or clause-complete compliance; report the implemented check scope and retain `engineeringApprovalRequired=true`. ### DNV-ST-F101 pipeline screening For current DNV-ST-F101 requests, use `neqsim.process.engineering.calculation.DnvStF101PipelineDesignKernel` with a complete `DnvStF101PipelineDesignInput`. It preserves operating, incidental, and test pressure; collapse; propagation buckling; local-buckling load interaction; fatigue; temperature/material de-rating; safety class; ovality; fabrication route; and installation strain as distinct checks. Do not route DNV-ST-F101 to `PipeMechanicalDesignCalculator.DNV_OS_F101`. That constant is the legacy DNV-OS-F101 screen. Missing structural inputs or unsupported editions must remain blocked, and a calculated result must retain `CALCULATED_REVIEW_REQUIRED`. Never describe a passing screen as certification or code compliance; require a licensed project copy and independent review. See `docs/process/dnv_st_f101_pipeline_screening.md` and the `neqsim-capability-map` skill. ## TR/NORSOK Integration Classes ### NORSOK M-506 execution rule For an explicit NORSOK M-506 calculation, prefer `neqsim.process.engineering.calculation.NorsokM506CorrosionDesignKernel`. It implements only the unamended 2017 edition, checks `Pipeline` / `AdiabaticPipe` / `Pipe` applicability, retains raw unit-explicit input values, and blocks unsupported or out-of-range cases before the mutable legacy calculator runs. Treat `NorsokM506CorrosionAssessment.getProjectedUniformWallLossMm()` as rate multiplied by exposure time, not as a specified corrosion allowance or acceptance decision. The kernel is `SCREENING`: verify the purchased standard, wetting and water-chemistry basis, localized corrosion, sour service, inhibitor availability, materials selection, and project criteria independently. When `feCO3SaturationRatio` is enabled, report it as a NeqSim film-factor extension and retain the source chemistry evidence. Standards Norway's May 2026 systematic-review notice is the lifecycle source for the catalogued 2017 edition; do not silently apply this kernel to a later revision. ### ISO 5167 execution rule For an explicit ISO orifice-metering calculation, use `neqsim.process.engineering.calculation.Iso5167OrificeMeteringKernel`. The registered method supports only unamended ISO 5167-2:2022, paired with ISO 5167-1:2022. It requires an `Orifice` equipment basis, explicit liquid or gas/vapour service, a supported tapping arrangement, and affirmative single-phase/full-pipe/subsonic/non-pulsating and installation evidence before calculation. Do not treat `geometryAndInstallationVerified(true)` as evidence created by NeqSim; retain the inspection and installation record. Keep uncertainty, calibration, straight lengths, plate condition, custody-transfer acceptance, and project metering procedure outside the kernel. Use `Standard_AGA3` under an AGA 3/API MPMS 14.3 basis and do not relabel that result as ISO 5167. ### DNV-RP-C203 execution rule For an explicit DNV-RP-C203 basis, use `neqsim.process.engineering.calculation.DnvRpC203FatigueDesignKernel`. It supports the catalogued `2024-10` edition including amendment `2025-10`, rejects additional project amendments, and is a `SCREENING` S-N/Palmgren-Miner calculation. The caller must supply and verify a controlled single-slope or continuous bi-linear curve, stress spectrum, SCF, thickness factor, other stress-range factor, design fatigue factor, damage limit, and exposure. Do not copy licensed curve tables into NeqSim and do not infer that a verification Boolean is the evidence itself. Retain curve/detail selection, environment, fabrication, thickness, weld and SCF basis, structural stress derivation, load combinations, rainflow counting, inspection plan, and accountable approval externally. The older pipeline/riser fatigue methods have inconsistent embedded intercepts and remain legacy estimates, not exact-edition C203 calculations. ### DNV-RP-F105 execution rule For an explicit DNV-RP-F105 basis, use `neqsim.process.engineering.calculation.DnvRpF105FreeSpanScreeningKernel`. It supports only the catalogued unamended `2025-12` edition for `Pipeline` and `AdiabaticPipe`. The kernel evaluates a simply supported Euler-Bernoulli first mode with caller-supplied effective modal mass and axial force, then reports current/wave frequency ratios, reduced velocities, and Keulegan-Carpenter number. Steel outside diameter and hydrodynamic diameter are separate inputs. Treat the geometry, structural-model, environmental, and project-trigger verification Booleans as attestations whose evidence must be retained externally. Strouhal number, frequency-ratio band, and reduced-velocity limits are caller-controlled escalation triggers, not embedded DNV requirements or acceptance decisions. Keep soil/shoulder stiffness, interacting spans, detailed in-line/cross-flow VIV and direct-wave response, ULS/FLS, fatigue, monitoring, intervention, and conformity open. `PipeMechanicalDesignCalculator.calculateAllowableSpanLength(...)` is a legacy fixed-assumption estimate with fallback and cap behavior. Never report that output as current-edition F105 evidence. ### DNV-RP-F101 execution rule For an explicit DNV-RP-F101 basis, use `neqsim.process.engineering.calculation.DnvRpF101CorrodedPipelineScreeningKernel`. It supports the catalogued `2019-09+AMD:2025-09` edition and only calculates the deterministic isolated longitudinal metal-loss equation under internal pressure. Require externally verified assessment wall thickness, measured defect depth and length, caller-controlled depth allowance, characteristic ultimate tensile strength, internal/external pressures, caller-controlled pressure factor, and isolated-defect applicability. Treat every verification Boolean as an attestation, not the evidence itself. Keep inspection accuracy and growth derivation, factor selection, interacting/complex defects, combined longitudinal compression, probabilistic methods, crack/dent/gouge/blister or weld damage, repair, fitness-for-service acceptance, and accountable approval external. Do not turn `NorsokM506CorrosionAssessment.getProjectedUniformWallLossMm()` into RP-F101 defect dimensions. RP-F101 remaining-strength screening is not DNV-ST-F101 original design. It does not replace pressure containment, collapse, propagation buckling, local buckling, load interaction, fatigue, incidental/test pressure, de-rating, safety class, ovality, fabrication route, or installation strain checks. ### DNV-RP-F104 execution rule For an explicit current `DNV-RP-F104 2021-02+AMD:2021-09` basis, use `DnvRpF104Co2PipelineEnvelopeScreeningKernel`. Require verified project CO2/water limits, other-impurity status, composition/EOS basis, the minimum-pressure interpretation and uncertainty of each supplied single-phase boundary, an ordered operating profile, MAOP, design temperatures, and external integrity/lifecycle review evidence. Negative margins remain calculated findings; missing evidence blocks execution. Use `StandardRegistry.requireRequirementPack(StandardSelection.strictRequirements( StandardType.DNV_RP_F104))` to discover related thermodynamic, hydraulic, corrosion, mechanical, monitoring, and consequence capabilities. The pack does not prove clause coverage. Do not substitute the pure-CO2 critical point, `CO2FlowCorrections.isDensePhase(...)`, or embedded `DensePhaseCO2Corrosion` values for the project basis. F104 does not replace DNV-ST-F101 structural design or the external fracture, materials, corrosion, construction, safety, operation, and requalification assessments. ### DNV-RP-F114 execution rule For an explicit current `DNV-RP-F114 2021-05` basis, use `DnvRpF114PipeSoilInteractionScreeningKernel`. Supply positive pipe diameter and submerged weight, then named route/design-situation cases with externally established non-negative vertical, axial, and lateral demand magnitudes and positive resistance magnitudes on a consistent N/m basis. Require external evidence for applicability, site investigation, soil interpretation, pipe/interface configuration, installation history, cyclic/drainage/rate/consolidation effects, load-displacement and resistance models, uncertainty/spatial variability, design actions and acceptance criteria, and lifecycle interfaces. A negative margin remains a calculated finding. Never derive F114 resistance from NeqSim soil thermal conductivity, burial heat-transfer inputs, a generic friction coefficient, or submerged weight alone. Keep F109 on-bottom stability, F110 global buckling, F105 free spans, ST-F101 structural design, and conformity external. ### DNV-RP-F110 execution rule For an explicit current `DNV-RP-F110 2019-09+AMD:2021-09` basis, use `DnvRpF110GlobalBucklingResponseScreeningKernel`. Supply positive pipe diameter and structural wall thickness, then named route/design-situation cases containing externally analysed effective force, peak longitudinal strain, peak global displacement, and required feed-in length together with caller-controlled allowable or available values. Require external evidence for applicability, operating envelope/effective force, pipe properties and as-laid geometry, pipe-soil interaction, imperfections/triggers/design strategy, global structural model, design situations/load combinations, local capacity/strain criteria, uncertainty/sensitivity/buckle sharing, and lifecycle actions. A negative margin remains calculated. Never interpret the force limit as a NeqSim-derived critical-buckling or initiation criterion. Keep F109/F114/F105, every DNV-ST-F101 check, conformity, and accountable approval external. ### API 2000 execution rule For an explicit API 2000 basis, use `neqsim.process.engineering.calculation.Api2000TankVentingScreeningKernel`. It supports only the catalogued unamended `7th Ed` for caller-verified non-refrigerated fixed-roof `Tank` or `SimpleTankFiller` service. Supply maximum filling/withdrawal rates, caller-controlled movement ratios, externally established thermal/other normal demands, total emergency demand, rated normal/emergency capacities, their rated pressure/vacuum conditions, tank limits, and one common gas-volume reference state. Treat the evidence Booleans as attestations, not proof. Keep API demand tables/equations, scenario derivation, vent area and device selection, manufacturer curves, pressure losses, flame arresters, blanketing, external floating roofs, refrigerated storage, installation/testing, and conformity external. An adequate caller-controlled constraint result is not API compliance. Use these Java classes when a task references Equinor technical requirements, STS0131, TR1965, TR2237, or NORSOK P-002: | Scope | Class | Use | |-------|-------|-----| | Gas scrubber conformance | `neqsim.process.mechanicaldesign.separator.conformity.ConformityRuleSet.create("TR1965")` | Checks TR1965 K-factor, gas/liquid margins, entrainment, and scrubber layout metadata configured on `GasScrubberMechanicalDesign`. | | Blowdown fire acceptance | `neqsim.process.safety.depressurization.STS0131AcceptanceCriteria` | Evaluates `DepressurizationResult` against time-to-escape/time-to-rupture pressure, inventory, and escalated fire-rate limits. | | Piping/line sizing | `neqsim.process.mechanicaldesign.pipeline.NorsokP002LineSizingValidator` | Screens `PipeLineInterface` velocity, pressure gradient, and erosional velocity using NORSOK P-002 style limits. | | Leak detection sensitivity | `neqsim.process.safety.leakdetection.MassBalanceLeakDetector` | Estimates minimum detectable leak rate from flow, pressure, temperature, and linepack uncertainty. | | Performance standards | `neqsim.process.safety.barrier.TR2237Templates` | Creates starter barrier registers with TR2237-style performance standards and NORSOK S-001 topic mappings. | | Standards review | `neqsim.process.safety.compliance.StandardsDesignReview` | Converts supported calculated checks from a `ProcessSystem` into `StandardsComplianceReport`. | | Overpressure LOPA targets | `LOPAResult.getSTS0131OverpressureTargetFrequency(...)` | Selects target event frequency from STS0131 pressure severity bands. | | LEL endpoint policy | `GasDispersionAnalyzer.builder().sts0131IntegralEndpoint()` | Uses 20% LFL for integral dispersion tools; `sts0131CfdEndpoint()` uses 50% LFL. | ### TR1965 Gas Scrubber Pattern