Step-by-step guide

1. Follow the installation guide to install OpenTerrace.
2. Browse through the tutorials to get an idea of how to set up simulations.
3. Then come back to this section to get more details on the individual steps.

Import OpenTerrace and define global parameters

import openterrace
ot = openterrace.Simulate(t_end=3600*6, dt=0.025, n_fluid=50, n_bed=5)

Line 1: OpenTerrace is imported so we can access its functions

Line 2: Create an instance ot and initialise with parameters that control our simulation. They include:

• t_end (required): End time of the simulation in seconds

• dt (required): Time step size in seconds

• n_fluid (optional): Number of discretisations for the fluid phase (omit if only solving the bed phase)

• n_bed (optional): Number of discretisations for the bed phase (omit if only solving the fluid phase)

Note, either n_fluid or n_bed should be specified. If either is omitted that phase wont be simulated.

ot.fluid.select_substance(substance='air')

Line 3: One of the predefined predefined substances is selected as the fluid substance. By using the predefined substances that come with OpenTerrace all properties such as density, viscosity, thermal conductivity will be temperature dependent:

• substance (required): Name of substance for the phase we are defining

Alternatively, we may define a custom substance on-the-fly with a constant density, specific heat capacity and thermal conductivity by:

ot.fluid.select_substance_on_the_fly(rho=1.2, cp=1000, k=0.06)

• rho (required): Density in kg/m^3

• cp (required): Specific heat capacity in J/(kg*K)

• k (required): Thermal conductivity in W/(m*K)

2. Select a domain shape

ot.fluid.select_domain_shape(domain='cylinder_1d', D=0.5, H=2)

• domain (required): Domain used for the phase

We choose a domain shape using on of the primivite shapes that come built into OpenTerrace. See this list for avialable shapes. Note, each domain will have additional parameters required, e.g. a sphere requires a radius to be defined. You will be prompted to add these if you are missing some.

3. Select a porosity (optional)

ot.fluid.select_porosity(phi=0.4)

• phi (required): Porosity m^3/m^3 (e.g. set to 0.4 and the fluid only occupies 40% of the volume)

The domain may be only partially filled with fluid as a bed phase occupies some space. This command may be omitted in which case the fluid occupies the whole domain.

4. Select discretisation schemes

ot.fluid.select_schemes(diff='central_difference_1d', conv='upwind_1d')

• diff (required): Discretisation scheme for the diffusion term

• conv (required): Discretisation scheme for the convective term

We choose how to discretise our diffusion and convective terms in our governing equations. A list of avialable schemes is avialable here diffusion schemes and convective schemes.

5. Select intial conditions

ot.fluid.select_initial_conditions(T=273.15+50, mdot=0.1)

• T (required): Temperature in K

• mdot (optional): mass flow rate in kg/s

We choose initial conditions for our simulation in terms of temperature and mass flow rate.

6. Select boundary conditions

ot.fluid.select_bc(bc_type='dirichlet', parameter='T', position=(slice(None, None, None), 0), value=273.15+600)
ot.fluid.select_bc(bc_type='neumann', parameter='T', position=(slice(None, None, None), -1))

• bc_type (required): Name of the boundary condition type

• parameter (required): Parameter for which you are specifying boundary condition

• position (required): Can be either (slice(None, None, None), 0) for lower bc (e.g. 1d cylinder) or centre boundary (e.g. 1d sphere), or (slice(None, None, None), -1) for upper bc (e.g. 1d cylinder) or surface boundary (e.g. 1d sphere).

• value (required): Specifies the value for dirichlet or dirichlet_timevarying type bcs.

Setup the bed phase

The bed phase is setup in a similar fashion to the fluid phase. An example is given below:

ot.bed.select_substance_on_the_fly(rho=5150, cp=1130, k=1.9)
ot.bed.select_domain_shape(domain='sphere_1d', R=0.01)
ot.bed.select_schemes(diff='central_difference_1d')
ot.bed.select_initial_conditions(T=273.15+50)
ot.bed.select_bc(bc_type='neumann', parameter='T', position=(slice(None, None, None), 0))
ot.bed.select_bc(bc_type='neumann', parameter='T', position=(slice(None, None, None), -1))

Note we use neumann type bcs for both boundaries. The coupling between the phases is a source term, which will be added using the coupling keyword next.

Define phase interactions and post-processing

ot.select_coupling(h_coeff='constant', h_value=20)

• h_coeff (required): Correlation for the heat transfer coefficient (note currently only constant is available)

• h_value (required): Heat transfer coefficient in W/(m^2*K)

ot.output_animation(save_int=6000, file_name='mySimulations')

• save_int (required): Save interval (number of time steps between two successive write outs)

• file_name (optional): Specfies a filename for the animation

Finally, we define how to visualise the results. Here, we save results every 6000 timestep (dt=0.025 s) e.g. every 150 s.

ot.run_simulation()

Finally, we execute the simulation and wait for the results to be generated.