Skip to content

Releases: fsepy/SFEPRAPY

0.8.1 (#17)

23 Oct 15:04
cf1804c
Compare
Choose a tag to compare
0.8.1 (#17) Pre-release
Pre-release

This release removed quite a few modules and migrated them to another open source project fsetools. The purpose of sfeprapy is to aid the sophisticated performance based approach of probability reliability assessment (PRA) in fire safety & structural engineering. However, many helper tools which are not so relevant to PRA had been built as parts of sfeprapy during research and development. These helper tools lost their purpose as part of sfeprapy when features are improved or new technics are discovered and implemented. Thus, to provide accommodation to these helper tools as they may still be useful other than PRA, fsetools has been created to serve the purpose of collecting all fire safety engineering related open source codes.

  • New:

    • New sfeprapy.func.mcs module, a base class for all Monte Carlo Simulation classes.
    • New sfeprapy.mcs2 module behaves like sfeprapy.mcs0 but takes different input parameters.
  • Improved:

    • Removed probability_weight. Only p1, p2, p3, p4 and general_room_floor_area are required.
    • Multiprocessing instances in sfeprapy.func.mcs are being reused to improve efficiency.
    • One way data to hard drive jobs in sfeprapy.func.mcs are carried out in non-blocking threads to improve its performance.
  • Depreciated:

    • Removed all GUI components.
    • Removed unnecessary output from sfeprapy.mcs0, e.g. fire_time, fire_temperature etc.
    • Removed figure plotting features, i.e. all features relevant to plotly and matplotlib.

0.6.9

28 Oct 10:37
80a3fc8
Compare
Choose a tag to compare
0.6.9 Pre-release
Pre-release
  • New: repository feature, codecov integration.
  • New: repository feature, travis integration.
  • Improved: CLI commands simplified and updated.
  • Improved: updated and some new test functions in \test directory.

0.6.8

22 Oct 23:15
Compare
Choose a tag to compare
0.6.8 Pre-release
Pre-release
  • New: use sfeprapy to trigger recently refreshed CLI. Previously this is python -m sfeprapy.mcs0.
  • Improved: CLI, added figure and save template input file features. Use sfeprapy -h to find more information.
  • Depreciated: sfeprapy.mcs0 module can no longer be triggered directly as python -m sfeprapy.mcs0 (i.e. codes are removed after if __name__ == '__main__' .
  • Depreciated: sfeprapy.mcs0 GUI to select input file, use CLI instead.

0.6.7

10 Oct 07:04
Compare
Choose a tag to compare
0.6.7 Pre-release
Pre-release
  • New: sfeprapy.mcs0 added exposure time dependent timber charring rate.
  • New: sfeprapy.func.mcs_gen added ramp input variable type for time dependent variables. the fill value should be csv formatted string without headers, consists of two columns, 1st column is the time and 2nd column is the value at the specific time.
  • New: sfeprapy.mcs0 added plotting feature, to run simulation with this feature activated python -m sfeprapy.mcs0 {fp} mp2 fig.
  • Depreciated sfeprapy.mc0. This module is imported/combined into sfeprapy.mcs0 at version 0.6.
  • Depreciated sfeprapy.mcs2. This module is imported/combined into sfeprapy.mcs0 at version 0.6.5.

0.6.6

12 Sep 08:20
Compare
Choose a tag to compare
0.6.6 Pre-release
Pre-release
  • New: system arguments added to python -m sfeprapy.mcs0 command line call. For example, calling python -m sfeprapy.mcs0 example_input.csv mp4 will run the problem definition file example_input.csv with 4 processes.

0.6.5

28 Aug 07:48
Compare
Choose a tag to compare
0.6.5 Pre-release
Pre-release
  • New: phi_teq is added in sfeprapy.mcs0 to include Model Uncertainty factor, see README for details.
  • New: sfeprapy.mcs0 implemented timber fuel contribution.
  • New: timber_exposed_area is added in sfeprapy.mcs0, see README for details.
  • New: timber_charring_rate is added in sfeprapy.mcs0, see README for details.
  • New: timber_density is added in sfeprapy.mcs0, see README for details.
  • New: timber_hc is added in sfeprapy.mcs0, see README for details.
  • New: timber_solver_ilim is added in sfeprapy.mcs0, see README for details.
  • New: timber_solver_tol is added in sfeprapy.mcs0, see README for details.
  • New: sfeprapy.mcs0.test automated tests for sfeprapy.mcs0.
  • Improved: sfeprapy.mcs0 changed criteria for parametric fire when fire_mode is set to 3. opening_factor should be within 0.01 and 0.2 (instead 0.02 and 0.2) to compliant to UK NA to Eurocode 1991-1-2.

0.6.4

19 Aug 07:56
269dddb
Compare
Choose a tag to compare
0.6.4 Pre-release
Pre-release

0.6.4

  • Fixed: window_open_fraction_permanent in sfeprapy.mcs0.
  • Fixed: added long_description_content_type to setup.py to change description type to markdown.
  • Improved: README.md updated to reflect the new sfeprapy.mcs0.

0.6.1

  • New: sfeprapy.func.mcs_gen general purpose stochastic variable generator.
  • New: sfeprapy.func_mcs_obj general purpose Monte Carlo Simulation object.
  • New: sfeprapy.mcs0 to implement mcs_gen and mcs_obj into the time equivalence calculation.
  • New: sfeprapy.mcs0 when beam_loc is removed, the most onerous location will be calculated and used based on specific fire curve.
  • Improved: sfeprapy.mc1 MCS routine is converted to an object, MonteCarloCase and MonteCarlo classes are provided to substitute existing factions.
  • Fixed: sfeprapy.mc1 convergence of protection thickness maybe failed to find.
  • Fixed: sfeprapy.dat.ec_3_1_2kyT units fixed, used degree K but should be degree C.

0.5

23 Jul 16:58
Compare
Choose a tag to compare
0.5 Pre-release
Pre-release
  • New: sfeprapy.pd6688.annex_b_equivalent_time_of_fire_exposure PD 6688 equivalent time exposure calculation. Manual can be found in its docstring annex_b_equivalent_time_of_fire_exposure.__doc__.
  • New: (WIP) sfeprapy.mc1 new equivalent time exposure procedure.
  • Improved: sfeprapy.mc optimised temperature dependent steel heat capacity routine, resulted in 65% less simulation time. Tested case shows 32.8 seconds reduced to 16.7 seconds for 1000 simulations on i7-7660U with 2 threads.
  • Fixed: sfeprapy.mc.mc_inputs_generator.py:mc_inputs_generator eliminated nan values in sampled stochastic variables - it is discovered negative or positive values are sampled even with predefined boundary limits, these extreme values (i.e. ±inf) are replaced with prescribed (user defined) limits (i.e. lbound and ubound).