ML22307A236
| ML22307A236 | |
| Person / Time | |
|---|---|
| Issue date: | 10/21/2022 |
| From: | Coyne K, Kenneth Hamburger, Mcgrattan K, Nick Melly, Markhenry Salley, Gabe Taylor NRC/RES/DRA |
| To: | |
| References | |
| Download: ML22307A236 (1) | |
Text
26thInternational Conference on Structural Mechanics in Reactor Technology (SMiRT 26) 17th International Post-Conference Seminar on FIRE SAFETY IN NUCLEAR POWER PLANTS AND INSTALLATIONS Development of Improved High Energy Arcing Fault (HEAF) Target Damage Thresholds and Zone of Influence (ZOI) Models Kevin Coyne, Gabriel Taylor, Kenneth Hamburger, Nicholas Melly, Kevin McGrattan, Mark Henry Salley
ZOI Development Comprised of three major research activities Updated Fragility CFD Simulations Confirmatory ZOI Modeling Analysis FDS development and Physical testing and application to large matrix Independent confirmation operating experience to of configurations using modified IEEE arc flash develop fragilities for model electrical cables
Cable Fragility Experiments SNL Solar Furnace Facility Heliostat and parabolic reflector, generating up to 6 MW/m2 over a 5 cm circle Varied heat flux, duration, cable material, and exposure profile Objective was to develop metrics for evaluating cable failure and quantify threshold criteria
Solar Furnace Testing Methodology Test 1-22 (7 MJ/m2): Surface damage to jacket only Test 1-32 (24 MJ/m2): Sub-jacket shielding visible Test 1-09 (206 MJ/m2): Cable insulation and bare wire exposed
Solar Furnace Testing Results Total of 38 tests conducted, and categorized by damage extent:
jacket damage, insulation exposure, conducting wire exposure Results for initial heat fluxes of >1 MW/m2 are shown (upper plot thermoplastic, lower plot thermoset)
Fragility Working Group Process for data analysis and consensus building 1 2 3 Proposal Weekly Meetings Consensus Development Proposals presented to full Technical evaluators and Two teams developed working group and integrators caucused to proposals to address the resource experts for reach agreement on path specific technical issues feedback forward for each issue
Fragility Working Group Conclusions The threshold for electrical Incident failure/damage of thermoplastic jacketed Energy cables is 15 MJ/m² and the threshold for thermoset jacketed cables is 30 MJ/m² Sustained Sustained ignition is assumed for cables Ignition within the enclosure of origin (e.g.
internal cables and components within switchgear and load centers)
Protective Features 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> (or greater) rated Electrical Fire Raceway Barrier System (ERFBS) will prevent ignition inside the enclosure of origin, and damage in the ZOI
FDS Inputs Source Term (HRR)
Radiative Volumetric HRR to Fraction represent the arc Relative convective/radiative fraction of HRR Mass of electrode vaporized during arc Particle size Electrode Mass Q distribution and extent Loss Rate of oxidation Metal Oxidation
Heat Release Rate Development Volumetric HRR Definition FDS does not include models for electrical/magnetic fields, dissociation of molecules at high temperatures, or formation of plasma Total arc energy can be accounted for using arc voltage, current, and duration A volume was defined in FDS to approximate the arcing region with the same total arc energy
Radiative Fraction & Electrode Mass Loss Data from Literature Cressault developed radiative fraction correlations for electrical arcs as a function of electrode material (aluminum, copper) and arc power Stanback developed electrode mass loss rate correlations as a function of electrode material (aluminum, copper) and arc current
Metal Oxidation Experimental Data SNL conducted experiments in a closed calorimeter with scaled current densities Experiments determined vapor fraction of lost electrode and extent of particle oxidation Nozzles defined in FDS to emit metal particles with defined oxidation rates at each conductor in the arc volume
FDS Validation 10000 ASTM PTC Full-scale experiments used for TCAP model validation; unbiased FDS 1000 predictions plotted against experimental temperatures for MV FDS Temperatur Rise (K) switchgear shown at right 100 10 1
1 10 100 1000 10000 Test Temperature Rise (K)
Processing the FDS Run Matrix Scripted Automation A matrix of FDS simulations was created to vary the important inputs: energy profile, geometric configuration, electrode material, housing material, and arc initiation location In total, 130+ simulations were run A Python script, FDS template, and parameter file were used to programmatically swap the relevant variables in and out and reduce the chance of human error
FDS Results & Conclusions FDS simulations allowed the determination of the distances where 15 MJ/m2 and 30 MJ/m2 would be exceeded.
Sample results shown for medium-voltage switchgear with arc initiated at the main bus bars and a 15 MJ/m2 1.4 fragility threshold 1.2 1.0 Dominant factor in ZOI is total arc energy.
0.8 ZOI (m)
Switchgear are sensitive to arc initiation location and geometry. Bus ducts are 0.6 sensitive to housing material (aluminum 0.4 vs. steel) 0.2 Results not sensitive to electrode 0.0 0 50 100 150 200 250 300 composition (aluminum vs. copper)
Arc Energy (MJ)
Back Left Right Top Front
Confirmatory ZOI Analysis Using a modified model based on IEEE guide 1584-2018 Current Base Enclosure Decay Model Breach Current Voltage Base model developed to Base model uses bolted Base model does not Arc energy profile estimate incident energy fault current, modified to assume a barrier between developed for generator-at various distances from use arc current. Arc source and target. fed faults during coast an arc flash for personnel voltage obtained from Modified model down following a trip safety CIGRE-602 model incorporates a time to breach before target receives energy
Modified IEEE & FDS Models Compared Modified IEEE model results are generally within 20 cm of FDS results. IEEE Max ZOI FDS Max ZOI MV Switchgear (15 MJ/m2) 1.6 m 1.3 m MV Switchgear (30 MJ/m2) 1.06 m 0.97 m NSBD Aluminum (15 MJ/m2) 1.2 m 1.41 m NSBD Aluminum (30 MJ/m2) 0.77 m 0.95 m Considering vastly different NSBD Steel (15 MJ/m2) 1.18 m 1.33 m nature of the models NSBD Steel (30 MJ/m2) 0.76 m 0.89 m (empirical vs. CFD) the ZOIs are in fairly good agreement and provide confidence in the results
THANK YOU!
https://www.nrc.gov/about-nrc/regulatory/research/fire-research/heaf-research.html