1. Data for the Use in Quantitative Risk Analysis of Hydrogen Refueling Stations
J. LaChance, J. Brown, B. Middleton, and D. Robinson
Sandia National Laboratories
Presented at the NHA Annual Hydrogen Conference 2008
Sacramento, CA
31 March 2008
11/30/2018

2. Outline Background Component leak frequencies
Traditional statistical analysis
Bayesian analysis
Generic data
Generic results updated with hydrogen data
Application of resulting data
11/30/2018

3. Project Background
Work performed under U.S. DOE Hydrogen, Fuel Cells & Infrastructure Technologies Program, Multi-Year Research, Development and Demonstration Plan
Hydrogen Safety, Codes & Standard R&D
Sandia National Laboratories is developing the scientific basis for assessing credible safety scenarios and providing the technical data for use in the development of codes and standards
Includes experimentation and modeling to understand behavior of hydrogen for different release scenarios
Use of Quantitative Risk Assessment (QRA) methods to help establish separation distances at hydrogen facilities and to identify accident prevention and mitigation strategies for key risk drivers
11/30/2018

4. The Problem
QRAs require component leak frequencies as a function of leak size and pressure
Data is not always available as a function of these parameters
There is little hydrogen-specific data that is available for use in QRA
So what data do you use?
Traditionally, representative values are selected from available sources from other industries
Problems with this approach:
Data is not necessarily reflective of hydrogen components and environments
Parameter uncertainty distribution is not characterized
11/30/2018

5. Possible Solutions
Use traditional statistical approaches to data analysis
Use Bayesian approaches to generate data
Used to combine multiple sources of generic data
Can give equal weight to all sources
Can exclude some sources (e.g., nuclear data) or specific data (e.g., outliers)
Can give variable weight to sources
Update results (prior distribution) with hydrogen-specific data (posterior distribution)
Hierarchical Bayesian approach used in our work allows one to attach different “layers” of significance to all the data that are used in the modeling process
11/30/2018

6. Traditional Statistics
11/30/2018

7. Traditional Statistical Equations for Accident Initiators
11/30/2018

8. Component Leakage Data
Generic leakage data is available from multiple sources covering different industries
Some data is provided as a function of leak size (i.e., small leaks, large leaks, and ruptures)
Actual data from offshore oil industry substantiates that leak frequency is a power function of leak size
Data is not generally differentiated based on operating pressure
Some limited hydrogen-specific data was obtained for this analysis
More hydrogen data is needed
11/30/2018

9. Hydrogen Leak Size Definitions
Very small - Leak area is <0.1% of total flow area
Minor – Leak area is 0.1% of total flow area
Medium – Leak area is 1.0% of total flow area
Major – Leak area is 10% of total flow area
Rupture – Leak area is 100% of total flow area
11/30/2018

10. Hydrogen Leak Rates – Traditional Statistics
11/30/2018

11. Bayes Theorem
11/30/2018

12. Example Sources for Generic Leakage Data
Center for Chemical Process Safety of the American Institute of Chemical Engineers, “Guidelines for Process Equipment Reliability Data with Data Tables,” 1989.
Cox, A.W., Lees, F.P., Ang, M.L., “Classifications of Hazardous Locations,” Institution of Chemical Engineers, 2003.
CPR 18E ed. 1, “Guidelines for Quantitative Risk Assessment: The Purple Book,” 1999.
Eide, S.A, Khericha, S.T., Calley, M.B., Johnson, D.A., Marteeny, M.L., “Component External Leakage and Rupture Frequency Estimates,” EGG-SSRE-9639, Nov 1991.
EIGA, “Determination of Safety Distances,” IGC Doc 75/01/E/rev, 2001.
NUREG/CR-6928, “Industry-Average Performance for Components and Initiating Events at U.S. Commercial Nuclear Power Plants,” February 2007.
NUREG-75/014, “Reactor Safety Study: An Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants,” WASH-1400, Oct 1975.
Rijnmond, Openbaar Lichaam; “Risk Analysis of Six Potentially Hazardous Industrial Objects in the Rijnmond Area, A Pilot Study,” COVO, 1982.
Savannah River Site, “Generic Data Base Development,” WSRC-TR , June 1993
11/30/2018

13. Hierarchical Bayesian Leak Rate Model
11/30/2018

14. Bayesian Results - Pipes
No hydrogen failures in very large operating history.
11/30/2018

15. Pipe Leak Results Minor Leak (0.1%A)
Generic prior - red
Hydrogen posterior- blue
11/30/2018

16. Pipe Leak Results Mayor Leak (10%A)
Generic prior - red
Hydrogen posterior- blue
11/30/2018

17. Bayesian Results - Joints
Significant number of hydrogen joint failures in large operating history.
11/30/2018

18. Bayesian Results -Cylinders
No hydrogen failures in very large operating history.
11/30/2018

19. Bayesian Results - Compressors
Significant number of hydrogen failures in a short operating history.
11/30/2018

20. Sensitivity Analysis
Impact of using all generic data versus just data from compressed gas sources.
11/30/2018

21. Sensitivity Analysis
Impact of using all generic data versus all data excluding nuclear sources.
11/30/2018

22. Application of Data
Data analysis can identify major contributors to leakage
Cumulative probabilities identify leak sizes most important to address in establishing separation distances in NFPA 2 and 55
Leakage frequencies are being used in hydrogen refueling station QRAs
11/30/2018

23. Summary
Component leak frequencies, component failure probabilities, and hydrogen ignition probabilities are required for QRA
Little hydrogen-specific data is currently available for traditional statistical analysis
Bayesian methods can utilize this limited data to obtain the parameters required for QRA
Additional hydrogen data will result in more realistic parameters
Data generated in this effort is being used to risk-inform NFPA 2 separation distances
11/30/2018

24. Reference
C.L. Atwood, J.L. LaChance, H.F. Martz, D.J. Anderson, M. Englehardt, D. Whitehead, T. Wheeler, “Handbook of Parameter Estimation for Probabilistic Risk Assessment,” NUREG/CR-6823, U.S. Nuclear Regulatory Commission, Washington, D.C. (2003).
11/30/2018

25. Backup Slides
11/30/2018

26. Comparison of Generic Results to EIGA Suggested Data
11/30/2018

27. Sources for Ignition Data
Canadian Hydrogen Safety Program, “Quantitative Risk Comparison of Hydrogen and CNG Refueling Options,” Presentation, IEA Task 19 Meeting, 2006.
HYSAFE, “An Ignition Probability Model Methodology for
Guidelines for quantitative risk assessment. "Purple Book" CPR 18E, ed. 1, 1999.
Matthijsen, A.J.C.M., Kooi, E.S., "Safety Distances for Hydrogen Filling Stations," Journal of Loss Prevention in the Process Industries 19, pp , 2006.
Melchers, Robert E., Feutrill, William R., "Risk Assessment of LPG Automotive Refueling Facilities," Reliability Engineering and System Safety, 74, 2001.
TNO: LPG - A study - A comparative analysis of the risks inherent in the storage, transshipment, transport, and use of LPG and motor spirit, 1983.
Cadwallader, C.J. and Herring, J.S., "Safety Issues with hydrogen as a vehicle fuel," 1999.
Cox, A.W., Lees, F.P., Ang, M.L., "Classifications of Hazardous Locations," Institution of Chemical Engineers, 2003
EIGA, "Determination of Safety Distances," IGC Doc 75/01/E/rev, 2001
F.P. Lees, “Loss Prevention in the Process Industries.”
11/30/2018

28. Ignition Probability Results
11/30/2018

29. Immediate Ignition Probability
11/30/2018

30. Bayesian Leak Rate Results
Component
Leak size
Generic Leak Frequencies (Phase 1)
Hydrogen Leak Frequencies (Phase 2)
Mean
5.0%
Median
95.0%
Compressors
Minor
2.3E-01
1.5E-02
1.1E-01
7.1E-01
2.2E-02
7.8E-03
2.0E-02
4.4E-02
Medium
1.2E-02
7.9E-04
5.2E-03
3.4E-02
7.9E-03
1.4E-03
5.9E-03
2.1E-02
Major
3.9E-04
6.6E-05
2.5E-04
1.0E-03
2.1E-04
3.5E-05
1.4E-04
5.7E-04
Rupture
9.7E-05
1.2E-06
1.2E-05
1.3E-04
3.4E-05
1.3E-06
1.1E-04
Cylinders
4.3E-02
2.3E-03
1.3E-01
9.8E-07
1.9E-07
8.3E-07
2.3E-06
9.5E-04
1.2E-04
6.3E-04
2.6E-03
6.7E-07
1.5E-07
5.6E-07
1.6E-06
2.7E-05
5.3E-06
1.8E-05
7.1E-05
3.9E-07
9.0E-08
3.2E-07
9.0E-07
8.4E-07
6.1E-07
2.1E-06
2.1E-07
4.8E-08
1.7E-07
5.0E-07
Filters
2.3E-02
4.1E-04
5.1E-03
6.1E-02 NA
4.2E-02
4.8E-03
5.5E-02
7.7E-03
1.1E-03
4.6E-03
5.4E-02
9.1E-04
4.4E-03
Flanges
5.3E-03
2.8E-04
2.2E-03
1.7E-02
6.3E-06
2.4E-04
9.0E-03
4.1E-05
6.8E-06
2.6E-05
1.0E-04
2.5E-05
1.4E-07
2.9E-06
5.9E-05
11/30/2018

31. Bayesian Leak Rate Results
Component
Leak size
Generic Leak Frequencies (Phase 1)
Hydrogen Leak Frequencies (Phase 2)
Mean
5.0%
Median
95.0%
Hoses
Minor
2.7E+00
2.1E-01
1.4E+00
8.1E+00
2.0E-04
3.7E-05
1.8E-04
4.4E-04
Medium
2.4E-01
3.4E-02
1.6E-01
6.4E-01
1.7E-04
3.9E-05
1.5E-04
3.8E-04
Major
2.4E-02
5.0E-03
1.7E-02
6.3E-02
1.6E-04
3.8E-05
1.4E-04
3.4E-04
Rupture
8.7E-03
2.0E-03
7.3E-05
6.2E-06
5.2E-05
2.1E-04
Joints
1.6E-02
1.0E-01
6.7E-01
3.4E-06
2.0E-07
2.7E-06
9.3E-06
4.1E-02
3.4E-03
1.8E-02
1.2E-01
7.6E-06
2.4E-06
7.0E-06
1.5E-05
4.3E-03
1.2E-03
3.6E-03
9.7E-03
6.8E-06
1.6E-06
6.0E-06
1.4E-05
9.2E-04
6.3E-04
2.3E-03
6.1E-06
1.3E-06
5.3E-06
1.3E-05
Pipes
1.0E-04
6.2E-05
2.7E-04
4.5E-06
8.6E-07
3.6E-06
1.1E-05
4.0E-05
8.2E-07
1.7E-06
9.1E-08
9.5E-07
5.4E-06
1.8E-06
1.8E-05
8.9E-07
5.2E-08
4.7E-07
3.1E-06
8.3E-09
3.2E-07
1.2E-05
5.6E-07
4.8E-09
1.5E-07
2.5E-06
Valves
6.4E-03
4.1E-04
1.9E-03
8.8E-03
7.4E-04
3.1E-04
6.9E-04
1.3E-03
1.4E-03
2.2E-05
4.4E-03
9.6E-05
6.3E-05
3.0E-04
7.2E-05
4.1E-05
9.7E-06
3.3E-05
3.0E-05
7.1E-07
8.4E-06
5.8E-07
5.5E-05
11/30/2018

32. Delayed Ignition Probability
11/30/2018