SFEPRAPY Processor¶

Input File¶

An input file should be a spreadsheet file in csv or xlsx format. Below shows an example.

\[\begin{split}\begin{matrix} \text{case_name} & \text{case_name_a} & \text{case_name_b} & ... \\ \text{n_simulations} & 2500 & 2500 & ... \\ \text{fire_time_step} & 10 & 10 & ... \\ \text{fire_time_duration} & 18000 & 18000 & ... \\ \text{fire_hrr_density:dist} & \text{uniform_} & \text{uniform_} & ... \\ \text{fire_hrr_density:lbound} & 0.249 & 0.249 & ... \\ \text{fire_hrr_density:ubound} & 0.251 & 0.251 & ... \\ \text{fire_load_density:mean} & \text{gumbel_r_} & \text{gumbel_r_} & ... \\ \text{fire_load_density:sd} & 10 & 10 & ... \\ ... & ... & ... & ... \end{matrix}\end{split}\]

The first column is the required input parameters by the time equivalence Monte Carlo simulation routine. The other columns are input parameters of specific simulation cases. There are no limits on the number of simulation cases.

Input parameters, their description, data type and dimensions are documented below.

case_name: str¶: A unique name for a simulation case. It should be unique among all simulation cases. This will be used as a file name to save MCS outputs, thus it should be file name safe on the operating system.

fire_mode: int¶: Should be an integer from 0 to 4, inclusive. To define what design fires to use:

0 - Only to use Eurocode parametric fire 1.

1 - Only to use travelling fire only.

2 - Only to use Eurocode parametric fire, German Annex 2.

3 - Use 0 and 1 as above based on certain conditions.

4 - Use 2 and 1 as above based on certain conditions.

n_simulations: int¶: The number of simulations that will be running. A sensitivity analysis should be carried out to determine the appropriate number of simulations. This should be a positive integer greater or equal to one.

room_breadth: float¶: [\(m\)] Breadth of room (the shorter dimension).

room_depth: float¶: [\(m\)] Depth of room (the greater dimension).

room_height: float¶: [\(m\)] Height of room (floor slab to ceiling slab).

room_wall_thermal_inertia: float¶: [\(J/m²K√s\)] Compartment lining thermal inertia. Thermal inertia is the tendency of a material to resist changes in temperature, i.e. to differentiate between thermal conductivity and heat capacity.

window_width: float¶: [\(m\)] Total width of all opening areas for a compartment.

window_height: float¶: [\(m\)] Weighted height of all opening areas.

beam_position_vertical: float¶: [\(m\)] Height of test structure element within the compartment for TFM. This can be altered to assess the influence of height in tall compartments. Need to assess the worst-case height for columns.

beam_position_horizontal: float¶: [\(m\)] Minimum beam location relative to compartment length for TFM - Linear distribution.

window_open_fraction: float¶: [\(1\)]. Glazing fall-out fraction.

window_open_fraction_permanent: float¶: [\(1\)]. Use this to force a ratio of open windows. If there is a vent to the outside this can be included here.

fire_tlim: float¶: [\(hour\)] Time for maximum gas temperature in case of fuel-controlled fire, value options can be found in Annex A EN 1991-1-2 1:

Slow: 25/60

Medium: 20/60

Fast: 15/60

fire_time_duration: float¶: [\(s\)] End of simulation. This should be set so that output data is produced allowing the target reliability to be determined. Normally set it to 4 hours and a longer period of time for greater room length in order for travelling fire to propagate the entire room.

fire_load_density: float¶: [\(MJ/m²\)] Fire load density. This should be selected based on occupancy characteristics. See literature for typical values for different occupancies 1 3.

fire_hrr_density: float¶: [\(MW/m²\)] Heat release rate. This should be selected based on the fuel. See literature for typical values for different occupancies 1 3.

fire_spread_speed: float¶: [\(m/s\)] Min spread rate for travelling fire.

fire_nft_limit: float¶: [\(K\)] TFM near field temperature.

fire_combustion_efficiency: float¶: [\(1\)]. Combustion efficiency (0.8 to 1.0 1 3).

fire_gamma_fi_q: float¶: [\(1\)]. The partial factor for EC fire (German Annex).

fire_t_alpha: float¶: [\(s\)] The fire growth factor.

beam_cross_section_area: float¶: [\(m²\)] Cross-section area of the structural member.

beam_rho: float¶: [\(kg/m³\)] Density of the structural member.

protection_protected_perimeter: float¶: [\(m\)] Heated perimeter.

beam_protection_thickness: float¶: [\(m\)] Thickness of protection.

protection_k: float¶: [\(W/m/K\)] Protection conductivity.

protection_rho: float¶: [\(kg/m³\)] Density of protection to beam.

protection_c: float¶: [\(J/kg/K\)] Specific heat of protection

solver_temperature_goal: float¶: [\(K\)] The temperature to be solved for. This is the critical temperature of the beam structural element, i.e. 550 or 620 °C.

solver_max_iter: float¶: [\(1\)]. The maximum iteration for the solver to find convergence. Suggest 20 as most (if not all) cases converge in less than 20 iterations.

solver_thickness_lbound: float¶: [\(m\)] The smallest protection thickness. This is used to solve the maximum steel temperature at solver_temperature_goal.

solver_thickness_ubound: float¶: [\(m\)] The greatest protection thickness. This is used to solve the maximum steel temperature at solver_temperature_goal.

solver_tol: float¶: [\(K\)] Tolerance of the temperature (in Kelvin) to be solved for. Set to 1 means convergence will be sought when the solved steel temperature is within solver_temperature_goal \(\pm 1\).

phi_teq: float¶: [\(1\)]. Model uncertainty factor multiplied with the evaluated characteristic time equivalence value to get the design time equivalence value.

timber_exposed_area: float¶: [\(m²\)] Exposed timber surface within the compartment, including CLT slab, glulam columns and glulam beams. Set timber_exposed_area to \(0\) to omit timber involvement.

timber_charring_rate: float¶: [\(mm/min\)] Timber constant charring rate. This is currently independent of temperature or heat flux.

timber_hc: float¶: [\(MJ/kg\)] Heat of combustion of timber.

timber_density: float¶: [\(kg/m³\)] Density of timber.

timber_solver_ilim: float¶: [\(1\)]. The maximum number of iterations that the solver can run. timber_solver_iter in the output file should be inspected to determine appropriate value for timber_solver_ilim. Consider to increase timber_solver_ilim (or increase timber_solver_tol) if many solved values have timber_solver_iter == timber_solver_ilim.

timber_solver_tol: float¶: [\(s\)] Tolerance of the solver. Convergence is sought if the change in time equivalence is less than timber_solver_tol.

Output Files¶

MCS results are saved in .\mcs.out, where .\ is the directory containing the input file. Below shows an example directory tree including input and output files:

.
├── input.xlsx
└── mcs.out
    ├── case_a.csv
    ├── case_b.csv
    ├── case_c.csv
    ...

Where .\mcs.out\case_name.csv contains the results of each simulation case (as per case_name) and this output file is produced upon completion of the simulation case. Below shows what an output file looks like.

\[\begin{split}\begin{matrix} \text{index} & \text{beam_position_horizontal} & \text{fire_combustion_efficiency} & ... \\ 318 & 27.35 & 0.96 & ... \\ 1065 & 25.04 & 0.83 & ... \\ 1244 & 20.22 & 0.92 & ... \\ 814 & 19.62 & 0.94 & ... \\ 1276 & 18.75 & 0.82 & ... \\ ... & ... & ... & ... \end{matrix}\end{split}\]

Each row of the output (as above) contains the deterministic parameters that are used per iteration.

index: int: An unique number associated with a MCS iteration.

beam_position_horizontal: float: See beam_position_horizontal.

fire_combustion_efficiency: float: See fire_combustion_efficiency.

fire_hrr_density: float: See fire_hrr_density.

fire_nft_limit: float: See fire_nft_limit.

fire_spread_speed: float: See fire_spread_speed.

window_open_fraction: float: See window_open_fraction.

fire_load_density: float: See fire_load_density in inputs.

fire_type: float¶: The type of design fire being selected for the iteration. See fire_mode.

0: Parametric fire

1: Travelling fire

2: Parametric fire (DIN)

t1: float¶: New in version 0.1.3.

[\(s\)] Fire growth phase end time.

For fire_type 0 this is currently not determined.

For fire_type 1 determiend as the time that the fire back face starting to move forward.

For fire_type 2 determined as the variable \(t_1\) in J. Zehfuss and D. Hosser 4.

t2: float¶: New in version 0.1.3.

[\(s\)] Fire decay phase start time.

For fire_type 0 this is currently not determiend.

For fire_type 1 determiend as the time that the fire front face reaching the room end.

For fire_type 2 determined as the variable \(t_{x,2}\) in J. Zehfuss and D. Hosser 4.

t3: float¶: New in version 0.1.3.

[\(s\)] Fire decay phase start time.

For fire_type 0 this is currently not determiend.

For fire_type 1 determiend as the time the fire back face reaching the room end.

For fire_type 2 determined as the variable \(t_{x,3}\) in J. Zehfuss and D. Hosser 4.

solver_steel_temperature_solved: float¶: The actual steel temperature from the last solver’s iteration. if solver_time_equivalence_solved is \(\text{True}\) then this should be in the range of solver_temperature_goal \(\pm\) solver_tol.

solver_time_critical_temp_solved: float¶: The exposure time for the structural element to reach solver_steel_temperature_solved when exposed to the selected design fire.

solver_protection_thickness: float¶: Solved protection thickness.

solver_iter_count: float¶: The number of iterations took to solve the time equivalence.

solver_time_equivalence_solved: float¶: The solved time equivalence value. This is the exposure time for the structural element to reach solver_steel_temperature_solved when exposed to the standard fire 5.

Reference¶

1(1,2,3,4,5): BSI, BS EN 1991-1-2:2002 Eurocode 1. Actions on structures. General actions. Actions on structures exposed to fire, British Standards Institution, London, 2002.
2: DIN, Eurocode 1: Actions on structures - Part 1-2: General actions - Actions on structures exposed to fire; German version EN 1991-1-2:2002 + AC:2009. DIN Deutsches Institut für Normung e. V., Sep. 2015.
3(1,2,3): BSI, PD 6688-1-2:2007 Background paper to the UK National Annex to BS EN 1991-1-2, BSI, London, 2007.
4(1,2,3): J. Zehfuss and D. Hosser, A parametric natural fire model for the structural fire design of multi-storey buildings, Fire Safety Journal, vol. 42, no. 2, pp. 115–126, Mar. 2007.
5: BSI, BS ISO 834-1:1999 Fire resistance tests. Elements of building construction. General requirements., BSI, London, 1991.