ΕΠΕΡ

DISPLAY-2

Contents

• Basic information
• Intended field of application
• Model type and dimension
• Model description summary
• Model limitations/approximations
• Resolution
• Schemes
• Solution Technique
• Input
• Output quantities
• User interface availability
• User community
• Previous applications
• Documentation status
• Validation and Evaluation
• Frequently asked questions
• Prortability and computer requirements
• Availability
• Other references 

Basic information (top)

Model name
DISPLAY-2

Full model name
DISPLAY-2

Model version and status
Updated November 97

Latest date of revision
Updated November 97

Institutions
Environmental Research Laboratory,Institute of Nuclear Technology and Radiation Protection,National Centre for Scientific Research DEMOKRITOS,15310 Aghia Paraskevi, Athens Greece. Institute for Systems Informatics and Safety Joint Research Centre ISPRA I-

Contact person
Dr. A.G. Venetsanos

Technical support
Provided by contact person

Level of knowledge needed to operate model
Basic

Intended field of application (top)

Vapour cloud dispersion in complex terrain, including two-phase releases.

Model type and dimension (top)

Two-dimensional shallow layer, local model. Developed based on the full three dimensional model ADREA-HF and the one-dimensional shallow layer model DISPLAY-1.

Model description summary (top)

DISPLAY-2 is a two-dimensional shallow layer model, to be applied for vapour cloud dispersion in complex terrain. The working fluid is considered to be an ideal mixture of two components, the pollutant and the air The pollutant can be in two phase conditions. Thermodynamic equilibrium is assumed for the mixture components. The model solves the mass, momentum in two horizontal directions and internal energy conservation equations for the mixture and the mass conservation equation for the pollutant, in Cartesian form. The conservation equations solved are integrated in the vertical from ground to cloud top The liquid part of the pollutant is obtained using Raoult’s Law. The slip velocity between liquid droplets and air is taken into account in the conservation equations. Air/ground interaction is also taken into account, by solving the transient one-dimensional temperature equation inside the ground. Turbulence is modelled using the entrainment velocity concept. The entrainment velocities depend on the local cloud velocities and on the ambient wind field, through a simple algebraic relation. Obstacles are modelled by adding flow resistance terms in the momentum equations. The spatial discretisation of the equations is based on the control volume approach. The computational grid is terrain following. The momentum equations are solved on staggered grids. Obstacles can be added on top of the irregular ground. The obstacles crossing the control volumes/surfaces of the grid are treated using the notions of control volume porosity and area permeability. This approach, inherited from the three dimensional model ADREA-HF, is particularly attractive because an increase in ground complexity does not increase the overall model complexity The necessary geometrical data are either given by hand or can be generated using the geometrical input processor DELTA_B. Time integration is based on the fully implicit scheme. The convective terms are discretised using the Upwind scheme. The pressure is hydrostatic. The Gauss-Seidel method is used for the solution of the discretised equations.

Model limitations/approximations (top)

The pressure is assumed hydrostatic.
The working fluid is assumed an ideal mixture of the carrier gas (air) and only one pollutant, which can be in two-phase conditions.

Resolution (top)

Temporal resolution
Time step: 0.001-1 seconds.

Horizontal resolution
Grid size: 1-10m

Vertical resolution
Equations integrated in the vertical.

Schemes (top)

Advection & Convection
Upwind

Turbulence
Turbulence is modelled using the entrainment velocity concept. The entrainment velocities depend on the local cloud velocities and on the ambient wind field, through a simple algebraic relation.

Solution technique (top)

Time integration is based on the fully implicit scheme. The convective terms are discretised using the Upwind scheme. The Gauss-Seidel method is used for the solution of the discretised equations per variable. Overall solution per time step using an iterative procedure. Automatic time step selection. Maximum Courant number is 0.5. No underrelaxation.

Input (top)

Emissions
For jet releases given are the jet area, velocity, temperature, contaminant mass fraction and the void fraction (pressure is assumed equal to the environmental). Time dependent jet releases can be handled. For instantaneous releases given are the temperature, contaminant mass fraction, mixture void fraction and cloud height at specified grid cells.

Meteorology
One dimensional universal vertical profiles of temperature and wind, representing the undisturbed meteorological conditions, are used. These profiles are assumed unaffected by the presence of the cloud and by the presence of obstacles.

Topography
Ground heights are given at specified grid cells. Obstacles (3d buildings or 2d fences) can be added on top of the irregular ground. The obstacles crossing the control volumes/surfaces of the grid are treated using the notions of control volume porosity and area permeability. The necessary geometrical data are either given by hand or can be generated using the geometrical input processor DELTA_B.

Initial conditions
See meteorology

Boundary conditions
Neumann or Dirichlet or both (in case of input planes). Transient boundary conditions available

Data assimilation options
Information not available. For more details, please, refer directly to the contact person.

Other input requirements
A file with control run data is read.
Two files containing the air and contaminant physical properties.
A file containing the maximum arrays dimensions.

Output quantities (top)

Optionally cloud velocity components, temperature, pressure, contaminant mass fraction, contaminant liquid fraction, mixture void fraction at user specified planes.

User interface availability (top)

Graphical User Interface under construction, based on MOTIF.

User community (top)

The DISPLAY-2 code is being used by Universities and Research Centres. Users of DISPLAY-2 should have a sufficient background on computational fluid mechanics.

Previous applications (top)

Application description
Information not available. For more details, please, refer directly to the contact person.

Documentation status (top)

Level 2 manuals

Validation and evaluation (top)

Level 2:
Simulation of the Thorney Island 8 and 21 large scale trials. Instantaneous isothermal releases of Nitrogen/Freon mixture on flat ground, without and with a semicircular fence obstacle. Simulation of the EEC-55 large scale experiment. A transient release of flashing propane on flat ground, with and without a fence. Simulation of the Desert Tortoise 4 large scale experiment. A continuous release of flashing ammonia on flat ground without obstacle. Considerable liquid effects. Simulation of the BA-Hamburg-University wind-tunnel experiments. Instantaneous isothermal releases of SF6 on slopes. Simulation of the EMU-A1 wind tunnel experiment. A continuous release of a passive pollutant from the door of an L-shaped building.
DISPLAY-2 has been included in the work of the Model Evaluation Group (MEG), Commission of the European Community, DGXII, Contact K.E.Petersen, System Analysis Department, P.O. Box 49, DK-4000, Roskilde, Denmark.
DISPLAY-2 is currently under evaluation in the European Commission Project SMEDIS (Scientific Model Evaluation of Dense Gas Dispersion Models), co-ordinated by the UK Health and Safety Executive.

Model intercomparison
Information not available. For more details, please, refer directly to the contact person.

Input data validation
Information not available. For more details, please, refer directly to the contact person.

Portability and computer requirements (top)

Portability
Extensive use on HP and DEC workstations.

CPU time
One or a few hours, depending on the simulated period and the type of machine.

Storage
40 Mbytes RAM. 100 Mbytes disk space

Availability (top)

The model is not a public domain programme. Information on the conditions for obtaining DISPLAY-2 can be provided by the contact person.

Other references (top)

• J.G. Bartzis, ADREA-HF: A three-dimensional finite volume code for vapour cloud dispersion in complex terrain. Report EUR 13580 EN, 1991.
• J. Wuertz, A Transient One-Dimensional Shallow Layer Model for Dispersion of Denser-Than-Air Gases in Obstructed Terrains Under Non-Isothermal Conditions. Report EUR 15343 EN, 1993.
• J. Wuertz, J.G. Bartzis, A.G. Venetsanos, S. Andronopoulos, R. Nijsing, The FLADIS Project Final Report. The JRC Ispra Contribution. Report EUR 16268 EN, 1995
• A.G. Venetsanos, N. Catsaros, J. Wurtz, J Bartzis, The DELTA_B code. A computer code for the simulation of the geometry of three dimensional buildings. Code structure and users manual, Report EUR 16326 EN, 1995.
• A.G. Venetsanos, J. Wurtz, J.G. Bartzis,. J. Statharas, Modelling the effects of obstacles on dense gas dispersion in shallow layer models. Proceedings of the Third International Conference AIR-POLLUTION 95, Porto Carras, Greece, September 26-29, 1995.
• J. Wuertz, J.G. Bartzis, A.G. Venetsanos, S. Andronopoulos, R. Nijsing, The FLADIS Project Final Report. The JRC Ispra Contribution. Report EUR 16268 EN, 1995
• J. Wuertz, J.G. Bartzis, A.G. Venetsanos, S. Andronopoulos, J. Statharas, R. Nijsing. A Dense Vapour Dispersion Code Package for Applications in the Chemical and Process Industry. Journal of Hazardous Materials 46, 273-284, 1996.
• A.G. Venetsanos, J.G. Bartzis, Further development of a two-dimensional shallow layer model for dense gas dispersion in obstructed irregular terrain including two phase jets. European Commission-TMR Return Grant Final Report, November 1997.