Slide_template_blank.ppt

1
SAFETY OF LABORATORIES FOR NEW HYDROGEN
TECHNIQUES Heitsch, M., Baraldi, D., Moretto,
P., Wilkening, H. Institute for Energy
(IE) Petten, The Netherlands
2
JRC Locations
3
Outline

• Motivation
• Problem Set-up
• Geometry
• Initial and Boundary Conditions
• Selected Results
• Conclusions and Continuation of Work

4
Motivation

• Assessment of the risk involved in accidental
  hydrogen releases,
• Creation of a decision basis for the placement of
  hydrogen sensors,
• Simulation of hydrogen release in the laboratory
• Assessment of flammability of clouds by size and
  duration,
• Investigation of several hydrogen release
  scenarios
• Small break low pressure release,
• Full cross section release.

5
Problem Set-up Geometry

• Laboratory has dimensions of about 18mx8.5mx3.5m
  with an approximate volume of 285 m3,
• Several test stands to investigate hydrogen
  sorption in materials,
• Lab is ventilated above each test stand fume
  hoods are installed, hydrogen sensors underneath
  all hoods,
• Hydrogen system
• 200 bar pressure, total hydrogen mass available
  is about 1,2 kg,
• Pipes and joints distributed at the walls.

6
Problem Set-up Geometry

• Laboratory and CFD model

7
Problem Set-up Geometry

• CFD Model consists of hexahedrals in most
  locations but also regions of pyramids and
  tetrahedrals to include hydrogen pipes with 4 mm
  and 10 mm diameter.

• Mesh about 774250 cells with 518660 hexahedrals.

8
Problem Set-up Initial and Boundary Cond.

• Analysis software used is CFX 10.
• Accident happens under normal working conditions
  in the lab
• Ventilation is working
• Flow speeds at inlets and outlets are used in the
  code as measured,
• A run without hydrogen release to get converged
  flow conditions as initial guess for the
  accident,
• The total amount of hydrogen available (1,2 kg)
  is assumed to eject into the lab. The release is
  idealised pressure drop in the hydrogen bottle
  is not considered -gt constant flow rate over time.

9
Problem Set-up Initial and Boundary Cond.

• The hydrogen system itself is not modelled,
  simulation boundaries are located at the open end
  of the ruptured pipe
• The low pressure scenario assumes a leak at a
  pipe connector with high pressure loss
  (sub-critical flow with 100 m/s and 500 m/s),
• The high pressure scenario assumes a full cross
  section break (4 mm diameter) with 50 bar and 100
  bar pressure at the pipe outlet. The outlet
  pressures chosen are based on rough estimates of
  the internal pressure drop in the hydrogen system
  at critical outflow conditions (parametric
  studies).
• Heat transfer to walls in modelled.

10
Problem Set-up Initial and Boundary Cond.

• Validation of CFX for relevant phenomena
• Gas mixing in closed space, e.g. HYJET, ThAI,
  LSGMF from AECL and OECD-SETH tests with helium
  instead of hydrogen and SBEP-V3 investigating
  hydrogen,
• Critical outflow some work by vendor, but needs
  more data, e.g. paper 1.1.125 (FZK and
  ProScience) from this conference.

11
Problem Set-up Initial and Boundary Cond.

• Leak outflow histories for critical cases (50 and
  100 bar)

12
Selected Results Sub-critical cases

• Very low amount of flammable mass

13
Selected Results Critical cases

• Hydrogen release for the 100 bar case

14
Selected Results Critical cases

• Flammable mass similar for both pressures

15
Selected Results Critical cases

• Influence of mesh resolution on flammable mass

16
Conclusions and Continuation of Work

• Simulations of accidental hydrogen release with
  CFX show that the existing ventilation system in
  the lab removes almost all hydrogen after about
  80 s. Most of the flammable mass is restricted to
  concentrations below 10.
• CFX proved to be robust enough to include very
  different length scales (from 4 mm to several
  metres) in combination with a range of time
  scales (from milliseconds to minutes) in a single
  run.

17
Conclusions and Continuation of Work

• Continuation of work is focused on
• the inclusion of the hydrogen storage system,
• high mesh resolution in the outflow region,
• allow for hydrogen deflagration in combination
  with the flame acceleration criterion,
• more comprehensive application of Best Practice
  Guidelines.