1 Application of quantitative risk assessment for performance-based permitting of hydrogen fueling stations A. Christine LaFleur, Alice Muna & Katrina M. Groth Sandia National Laboratories Albuquerque, NM International Conference on Hydrogen Safety Yokohama, Japan October 21, 2015 Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL SAND C

2 Outline PBD overview & introduction Analysis to benchmark station risk

3 Project Approach: Three coordinated activitiesApply risk assessment techniques in step-out hydrogen technologies Apply R&D in RCS QRA methods, tools R&D Develop integrated algorithms for conducting QRA (Quantitative Risk Assessment) for H2 facilities and vehicles Develop and validate scientific models to accurately predict hazards and harm from liquid releases, flames, etc. H2 behavior R&D We’re conducting two main research activities that feed into an applied activities to impact codes and standards. This arrangement is in line with our program focus on identifying R&D needs, performing high-priority R&D, and impacting codes and standards. Three pieces: Developing the methods, developing scientific models, and piloting the application of the methods and models. Enabling methods, data, tools for H2 safety & RCS community

4 Separation distances in NFPA 2 for liquid hydrogen stations are often prohibitively largeA California Road Map: The Commercialization of Hydrogen Fuel Cell Vehicles, CalFCP, July 2014 Harris, SAND Of 70 stations surveyed (out of 343), none met the NFPA 2 Ch. 6 separation distance requirements. Harris, SAND

5 Key Barrier – Prescriptive LH2 Separation DistancesCurrent bulk distance values Based on historical values Present critical limitation to hydrogen infrastructure growth Science-based Code Improvements - Ongoing effort by NFPA 2 subcommittee to revise based on risk-informed science of LH2 release behavior. Best case schedule for 2019 code edition, jurisdictional adoption later Alternative Methods for Code Compliance - In the meantime, this effort is exploring a path forward for short term deviation from separation distances for LH2

6 Enabling Hydrogen Codes & Standards: Two related pathsAlternative Methods for Code Compliant Hydrogen Infrastructure Science-based Code Revisions to Address Critical Limitations to Station Implementation https://en.wikiquote.org/wiki/File:Balanced_scale_of_Justice.svg

7 PBD: An Alternative Methods for Code ComplianceNFPA 2 (Hydrogen Technologies Code) specifically allows performance-based designs for hydrogen facilities. Performance-based designs (PBD) enable alternate specifications that do not conform with the prescriptive code requirements, but ensure equivalent safety through the use of performance criteria. Performance-based design solutions have not been developed for hydrogen applications due to: Perceptions that PBD is cost prohibitive Uncertainty surrounding acceptance by (AHJs) due to lack of familiarity with risk methods Lack of validated methodology Tenability Tenability Heat Flux Heat Flux Risk Risk

8 Performance-Based Design ProcessStandard process developed by NFPA and SFPE Assumptions Only a single fire source is assumed to be present Multiple, simultaneous unauthorized releases of hazardous materials from different locations are not considered Combinations of multiple events are not considered NFPA Performance-Based Design Process

9 Outline PBD overview & introduction Analysis to benchmark station risk

10 Application of QRA to Benchmark PBD criteriaPerformance-Based Design of Refueling Station Real-World Station Representative Station: H2FIRST Reference Liquid Station Meets all Prescriptive Requirements Modifies Key Requirements Modifies One Key Requirement and Incorporates Mitigating Factors Prepare Performance-Based Design Report and Documentation Utilizing HyRAM QRA Toolkit Calculate Benchmark Performance Criteria Incorporate Mitigating Factors Follows Real-World Permitting Process Calculate Risk Risk Equivalent Performance Criteria Vet with H2 Code Industry and Stakeholders

11 System Description & Target PositionsPratt, NREL/TIP , SAND R The example system is an outdoor, publically-accessible, gaseous hydrogen station The analysis will determine the risk to a member of the public outside the NFPA-2 compliant lot line distance

12 Performance-Based Design Required Scenarios Specified in NFPA 2, Hydrogen Technologies CodeFire Explosions Pressure Vessel Burst Hydrogen Deflagration Hydrogen Detonation Hazardous Materials Unauthorized Release Exposure Fire External Factor Discharge with Protection System Failure

13 NFPA 2 Performance-Based Design: Fire & Explosion ScenariosPressure Vessel Burst Hydrogen Deflagration : Deflagration of a hydrogen-air or hydrogen-oxidant mixture within large process equipment Deflagration within the enclosure housing the compressor HyRAM peak overpressure and risk metric calculation Hydrogen Detonation : Detonation of a hydrogen-air or hydrogen-oxidant mixture within a process vessel or within piping containing hydrogen Unintended release forms localized H2/air mixture that detonates Prevention of detonation by meeting vent pipe length to diameter ratio 5.4.2: Design for life safety : Pressure vessel ruptures Scenario Description Outdoor Fueling Station Scenario Hydrogen fire resulting from a leak at the dispenser Prevention of gaseous H2 vessel rupture HyRAM jet fire risk calculation Pressure relief devices and leak –before-burst design specification Performance Criteria Approach

14 NFPA 2 Performance-Based Design: Hazardous Materials ScenariosUnauthorized Release : Unauthorized release from a single control area Accidental release of hydrogen from liquid storage tank Liquid hydrogen release model analysis Exposure Fire : Exposure fire on a location where hydrogen is being stored, used, handled or dispensed An unrelated car fire at the gasoline dispensing pump Characterization of flame radiation from vehicle fire on nearest hydrogen system components External Factor : Application of an external factor that is likely to result in a fire, explosion, toxic release or other unsafe condition Seismic Event where a pipe bursts (100% Leak Size on largest system pipe) HyRAM risk metric calculation Discharge with protection system failure : Unauthorized discharge with each protection system independently rendered ineffective An unauthorized discharge where the interlock or pressure relief valve fails Discussion of layered safety features present in the system Scenario Description Outdoor Fueling Station Scenario Performance Criteria Approach

15 HyRAM QRA used to develop baselineHyRAM toolkit is being used to calculate risk values for the performance criteria on a representative station, fully compliant with prescriptive codes This will provide a Baseline for risk values and enable easy comparison of the changes made in the second iteration The AIR value is used for comparison purposes Coordinate station with H2FIRST project

16 NFPA 2 Performance-Based Design: Fire ScenarioHyRAM Input Parameter User Input Value Compressors Cylinders Valves 7 Instruments 10 Joints Hoses 2 Pipes (length) Filters 1 Flanges Pipe OD inch (9/16) (1.43 cm) Pipe wall thickness in (0.32 cm) Internal Temperature 15 C (59 F) Internal Pressure 900 bar External Temperature External Pressure MPa Release Rate <0.125 kg/s 0.008 Release Rate kg/s 0.053 Release Rate >= 6.25 kg/s 0.23 Release Rate <0.125 0 – jet fire only Release Rate Description: Leak from gaseous side of H2 system, at dispenser. System parameters for largest pipe size, pressure and temperature were inputted into HyRAM and a baseline risk value for a code compliant station was calculated. HyRAM Input Parameter User Input Value Number of Vehicles 50 Annual fueling demands 18,000 Notional Nozzle Birch2 Flame Radiation Model Ekoto/Houf (curved flame) Deflagration Model None - Fire scenario only Thermal Probit Tsao and Perry Thermal Exposure 60 sec Overpressure Probit Population 6 people, based on 2 at H2 dispenser, 2 in the gasoline dispenser and 2 entering store. Working hours per year 6480 hrs (30 days*12 months*18 hours a day) Max Distance 120 ft. (36.6 m) distance to lot line Min Distance 1 ft. (0.3 m)

17 Baseline Result for Fire ScenarioFire Scenario Result: AIR: 1.05 x Fatalities /year Risk Values for US Gasoline Stations Member of Public (Used in NFPA 2): PLL or AIR below 2 x fatalities/station-yr Based on 2 fatalities/yr and 100,000 refueling stations in the US Workers: One order of magnitude higher than public risk 1 x 10 -4 Other Risk Statistics Average Individual Risk (CDC actuarial data 2005) = (9117,809 Deaths/Year)/296,748,000 Total U.S. Pop. = 4 x Deaths/Person-Year (~ 1/2,500 Deaths/Person-Year) In any given year, approximately 1 out of every 2,500 people in the entire U.S. population will suffer an accidental death Norwegian Petroleum Directorate guidelines use a total frequency of 5 x 10 -4/yr for all accidents for all safety functions

18 NFPA 2 PBD: Hazardous Material Scenario 3 — External FactorHyRAM Input Screen Parameter Value Components Compressors Cylinders Valves Instruments Joints Hoses Pipes (length) 10 Filters Flanges Piping Pipe OD 1.315 inch (3.34 cm) (1 inch nominal) Pipe wall thickness .179 in (0.45 cm) Internal Pressure 10 bar Pipe Leak Size for Pipe component only: Mean 0.01% 0.10% 1% 10% 100% 1 Pipe Leak Size for all components except Pipe: Mean Description A seismic event occurs that results in a 100% leak of the largest pipe in the hydrogen system due to shearing. Because explosive conditions are dealt with independently in other design scenarios, only the effects of a fire are considered in this scenario.

19 Baseline Result for External EventExternal Event Scenario Result: AIR: 1.81 x Fatalities /year, conditional upon occurrence of an earthquake Fire Scenario Result: AIR: 1.05 x Fatalities /year Next Steps: Analyze all applicable performance-based design scenarios to determine a baseline performance criteria Design various ‘trial designs’ that do not meet specific prescriptive requirements which may be infeasible given site specific conditions Evaluate trial designs against the baseline prescriptive criteria

20 Impacts Increase options for industry in siting hydrogen fueling stations Increase confidence in the performance-based approach for station design Reduce effort required by industry to use the performance-based approach by providing template and test case Provide path for international harmonization of the performance-based approach to station design Lay groundwork (establish precedence) for similar performance-based approach for other alternate fuels (i.e. CNG and LNG)

21 Thank you! Katrina Groth Sandia National Laboratories Research supported by DOE Fuel Cell Technologies Office (EERE/FCTO)

22 NFPA 2 Performance-Based Design: Scenarios which do not apply to an outdoor fueling stationEgress System Max Occupant Load with Blocked Exit : Maximum occupant load is in the assembly building and the principal exit/entrance is blocked No assembly occupancies in the vicinity and no exits to block Construction in area of building with suppression system out of service : Fire in an area of the building undergoing construction while remainder of building is occupied. The suppression system has been taken out of service. No partially-occupied buildings with suppression system out of service to analyze 5.4.2: Design for life safety affecting the egress system Scenario Description No egress system since fueling station is outdoors Discussion of Applicability