Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
✨ 2D Neutronics feature branch (#2656)
* 2D Neutronics module (#2344) * Initial commit to repository * Create basic README.md * Editing README.md * Code styling: Black, isort and whitespace/trailing new lines (#1) * blacken and isort * whitespace and new lines * Rearrage and refactor (#2) * Adding plasma source, DCLL and WCLL. Material defintions and geometries still need tweaking. * Changing from toroidal first wall profile to match points and adding extra source geometry plots. * Adding source code for parametric plasma source * Adjusting point too close to plasma. * Minor format tidy up. Moving function. * Passing geometry_variables to get_fw_points * Update make_materials.py Changing tungsten isotopes and tidying export_materials() * Update make_materials.py W isotopes corrected. * Made most of it OOP, especially the summary, so that the data can be recalled. * Added vscode into gitignore * Finished reformatting the Summary object so now it prints the entire run properly. Not documented yet, and kwargs not passed on yet. * Finished some reformatting. Todo list is at the top. * Some work done on load_fw_points, but incomplete. TODO list updated. * Added some documentations, but found a lot of magic numbers and 'things that don't belong in this module'. I'll deal with them later. * Changed the ordering and naming of the arguments to make_geometry in make_geometry.py, and added some docstring. * Fxied the types of Blankets available using BlanketType in make_materials.py. * Fixed the naming for cells in quick_tbr_heating.py - renamed them to cells_and_cell_lists * Updated the todo list * Cleaned up create_tallies, and in the process, created MaterialsLibrary in make_materials.py so that it becomes modular enough for testing. * Formatted the 'print_df' booleans away into a decorator - can change into printing to bluemira log later. * Added documentations to the rest of the important classes. * Updated todo list * one line change to make pandas print more prettily, before merge commit * moved all relevant files into the neutronics directory. * Moved files from the 2d-neutronics repo to here. * restored comment style fix * moved the neutronics files to the correct directory. * changed the import relative path * relocated files to the correct directory * changed the import path * minor changes before complete overhaul (the latter is for adhering to bluemira code style) * moved some constants to constants.py * Finished one task: placed the relevant constants and conversion units into constants.py * updated to-do list * Further broke down the make_materials.py file into materials_definition.py file * Refactored all of materials. * updated .gitignore * Moved the results formatting functions to a separate file * removed unused file * Improved styling * Broke functions down into appropriate modules * Split into appropriate modules * Added typing information * Updated the import diagram to reflect the import relationships between modules. * Fixed all fixable formatting issues according to flake8 * Fixed all flake8 messages! * installs script * cleaned up script * added complete apt install deps * Updated install script, removed pps_api * Updated to-do list, improved result presentation workflow * .npy requires conversion to BluemiraWire * ➕ Use ninja and install libopenmpi-dev * ➕ Add openmc to conda * ✨ Start using wire as input * 🚨 Some ruff fixes * 🎨 Move ifmain to example * 🚚 Move npy * 🎨 Fix example * 🎨 Get constants from base * 🎨 Fix imports * 🚨 Some ruff fixes * added openmc version to install script, added pps-isotropic package import to quick tbr * removed my path from config * 🎨 Cleanup printing * 🎨 Unit fix * 🎨 Unit fix2 * 🎨 Unit fix3 * 🎨 Use openmc enum * Preparing for unit changes * quick commit before rebase, units half-done * 🙈 Remove noise * 🔥 Remove so file * Removed the not-sensible default constants and placed thme into neutronics.ex.py instead. * Added TODO comment * 🎨 Cell filter cleanup * 🎨 Cleanup tallies * Changed all units * minor fix of import statement * Monkey patched successfully * monkeypatched successfully, and separated plasma geometry variables from tokamak geometry variables. This is the commit BEFORE applying the DataclassUnitConverter pattern. * Minor update to add comments. * Removed the duplicate codes inside params.py. This is the commit AFTER applying the DataclassUnitConverter pattern. * Bug fix - missing 'self,' argument fixed. * Updated todos * 🎨 Change raw_uc percent use and runtime var * 🚧 Make geometry dataclass refactor * 🚧 Slight cleanup of params * 🚨 Ignores * 📝 Docs * 📝 Fixes and docs * 📝 Fix documentation build * Update neutronics.ex.py * Update pyproject.toml * 🐛 Fix volume calculation * 🎨 Cleanup vol calc * 🚨 Fix black * 🎨 Small fixes * 🚚 Move quick tbr to neutronics_axisymmetric * 🎨 Add stochastic volume to output results * Added comments on why only two quantities are excluded in the conversion * pps-isotropic's API changed, making appropriate changes here to maintain usability * Added markdown docs on neutronics.ex.py * 💩 Hack for old version of openmc * 👥 Add contributors --------- Co-authored-by: jamesnha <[email protected]> Co-authored-by: je-cook <[email protected]> Co-authored-by: Oliver Funk <[email protected]> * ✨ OpenMC data downloader (#2618) * ✨ OpenMC data downloader * 🎨 Parse known args * 🎨 Ifmain * 🎨 Rename file * 🎨 Write them all then delete all * 🐛 Overindent * 🚨 Cleanup and linting * 🐛 No output * 🐛 Fix neutronics json location * 🐛 Hack to force data into a given directory * 🎨 Spelling * 🏷️ Typing fix * 🚨 Ruff fixes * 🙈 Add to gitignore * Minor error fixes to the outputted tables (#2689) * Fixed unit conversion error in photon heat flux; and turned off Bremsstrahlung heating by default * Undo accidental indent * Kept all statepoint files' units as openmc as cgs, so that the conversion to SI is done purely in python. Fixed misinterpretation of photon flux. * added comments to explain tally score choice. * Minor bug fix to pass code quality checks * 🚨 Ruff fixes * first move to BM materials, lots of refactoring needed but the example runs * 🎨 Basic cleanup * 🎨 Dataclassify * 🐛 Undo Be12Ti density change * 💡 Minor fixes * 📄 Copyright fixes after rebase * ✨ Allow '*' to extract all XS data * ⬆️ Openmc 0.13.3 * temperature docstring. * removed rounding * calculate the DT energy more accurately using plasma physics. * Simplified code to use n_DT_reactions instead. * 1. extracted (major_r, minor_r, elong, triang) into a different class than TokamakGeometry, and named this class PlasmaGeometry. 2. Removed the 'CGS' and 'PPS' from the 3 classes in params.py, and instead appended their baseclasses with 'Base', so that it becomes clearer that these dataclasses store SI units unless the .csg or .plasma_physics_units property is accessed. * (minor bug fix) * Updated the Shafranov shift description. * 🚨 Fix linter * Minor updates to docstrings * 🚨 Ruff & python3.10 fixes * Increase CSG speed without compromising on flexibility (#3079) * :Changed params.py according to Matti's suggestion * Made temporary script * Added json data to be loaded as test case bluemirawires * Added post_init checks. * Created basic script to translate from existing dataclass TokamakGeometry (make_csg) * WIP: making pre-cell. * WIP: making pre-cell. * Bodging serializing issue - fix later. * Deleted failed files. * Test script in attempt to replace the magic numbers/magic procedure to something more generalizable * quick commit * Script to load the divertor. * Added capability to automatically slice up the blanket according to the first wall panels * Updated script for creating neutronics csg * extended the functionality of signed distance to check points as well. * Added function based on signed_distance to check for overlaps. * Add tolerance to account for low sampling rate/ float errors. * Adding module to optimally slice the neutronics model. * Updated module that chooses better surfaces for faster neutronics runs. * Made the volume optimization converge. * Made a better approximation of the reactor. Added tools necessary to make such approximation * Added better documentations. Formatting. * Made a very rigorous structure to create openmc.Cells. * removed unused code * Updated test case * Half-working state * Finally finished the blanket bug fix. * Updated docstrings * Updated equations * Updated comment * Interited from collections.abc.Sequence to avoid having to make def __iter__(). * Cleaned up excess code * minor changes to naming and docstring * Updated the run modes into context managers. * Enforced self.setup() to be ran before self.run() in the contextmanagers. * Added ability to slice divertor wire * Variable renaming * Removed redundant checks * Improved typing (Iterables vs Sequence), and added classes to prepare for tallying. * Added tallying functionality * Developing divertor cells automatic conversion into pre-cells... * Added type hints for return types. * Fixed import * Successfully exported into DivertorPreCellArray. * Updated script correspondingly * split up WireInfo etc. into its own file. * Made make_divertor_pre_cell_array more robust (able to straighten out the cell walls better before passing onto DivertorPreCell). * Typing and docstring improvement * Added function to flip a direction vector if it is pointing the wrong way; Moved plane functionalities to geometry/plane.py * Removed plane creation functionality (migrated to bluemira.geometry.plane) * Minor change to type hinting in make_csg.py * Minor changes to type hints in radial_wall.py * Slight change to variable name * Added plan to make the divertor slicing more robust. * Moved the warning to an appropriate place * Improved the interior_wire_offset method * Fixed minor mistake in direction * Working version: produces pre-cell arrays for both divertor and blanket * Pointed the chord in the correct direction again * Removed unused data loading files. * Improved file structure so each class has separate responsibility. Surfaces that aren't shared among different BlanketCell stacks won't be created in BlanketCellArray. * Minor bug fixes related to types and WireInfo * Moved constants into constants.py * Docstring updates * Simplified boolean expression; made plasma cell. * Deleted unused statement. * updated test script to plot better. * Named file more sensibly. * Fixed some ID control issues * docstring rearrangement * Fixed up the get_all_hollow_merged_cells method so the test script won't have to write something as long to plot the overall-cells. * Fixed the thickness issue - now instead we have absolute thickness. * Got as far as getting some segfaults out of openmc. * Finally working. Tallies need fixing * Fixed floating point issue * Added missing conversion function * Changed thickness. * Modified to allow for a variable number of batches. * Delete some of the obsolete classes * Updated TODO tags in test script. * Updated TODOs * Added missing method back in * Updated documentations slightly to make it more readable. * Updated docstring to conform usage of anti- vs counter- * Ensured backward (neutronics.ex.py) compatibility * Calculated volumes without relying on openmc. * Added function to support volume calculation of the openmc.Cell's .volumes. * Optimized volume calculation (commented them out where they're not needed) * Working version. * Added TODO on reducing the number of negated regions. * ruff fixes * Some updates to how surfaces are get * fixed ONE TODO tag. * Updated zip strict = True almost everywhere, pairwise where applicable. * Cleaned up code slightly at make_csg * We can now control the thickness of the central solenoid from parameter frame directly. * Attempting to use vacuum vessel wire as well. * Further attempt at incorporating the vacuum vessel interior curve into the pre-cells * VV now successfully sandwhiched between the two curvesgit add . * Minor formatting updates * Finished porting over the materials library * Cleaned out all low-lying fruit TODO tags * Made note about convexity. * Neutronics Solver interface (#3231) * 🚧 OpenMC designer * 🚧 Working designer * ✨ Openmc solver * ✨ Fixup from upstream * 🚚 Move stuff * Neutronics old code removal and architectural adjustments (#3232) * 🌐 Bye Z * ♻️ Initial cleanup * ♻️ More cleanup * ♻️ More cleanup * ♻️ More cleanup * 🔥 Destroying Vertices pt1 * 🔥 Destroying Vertices pt2 * ♻️ More cleanup * 🔥 Remove Sequence inheritance * 🎨 Types * 🚨 Remove ruff ignores * ♻️ More cleanup * ♻️ Remove global hangar variable * ♻️ More cleanup * ♻️ More cleanup * 🚨 More cleanup * 📝 Docs cleanup * 📝 Docs cleanup * 🐛 Fix double counting of divertor material * 🐛 Fix 0 volumetric heating * 🎨 Convert SingleNullTokamak to a function * 🏷️ Typing and general cleanup * ♻️ Cleanup dimensions * 🎨 Order functions * ✨ Split of Neutroics pre cell csg and openmc running (#3243) * 🎨 Neutronics PreCell / openmc separation * 🚚 Move stuff about * 🎨 Reduce surface that requires openmc * 🚚 Move materials about * 🎨 Initial material rearrangement * 🌐 More internationalisation * ♻️ Neutronics material movement * 🎨 More materials stuff * ♻️ Cleanup * 🎨 Materials cleanup * Neutronics EUDEMO integration (#3239) * inital go * Saving * hard part done * pass vals to Solver * Updated WireInfoList so it doesn't use horrible mixed types. * expected div wire * Fixed the issue of 'sometimes misordering wires', and used recursion to shorted the break_wire_into_convex_chunks method for better readability with no loss in speed. * 🚧 Tmp get it working * Fixed the .reverse() method, which didn't flip the sign of the tangent. * Improved plotting in plot_2d and plot_surfaces (mainly used for debugging) * 🐛 Fix inner panel point * 🔥 Reduce interface * ♻️ More cleanup * 🎨 More cleanup * 🎨 Remove while loop from break wire into convex chunks * 🎨 Reduce interface * 🎨 More cleanup * 🔥 Remove readme * 🐛 Typing and divertor start point fix * 🎨 Fix extra error and cleanup some reactor mechanics * 🎨 Improve file save location * 🎨 Only pass in ivcshapes not divertor * 🎨 Feed in correct blanket cut point * 🎨 Cleanup runmode * Fixed leaking plasma void. * 🎨 Cleanup inputs * Added the the half_bounding_box function to get the minimum positive r. * 🔥 Remove old data files * ♻️ Run through cleanup * Fixed the final overlap issue at the bottom of the divertor * 🚚 Move some geometry functions to geometry * 🐛 Fix tfcoil size * 🐛 Fix blanket test * 🚨 Linter fix * 🚚 Move some things about * 🚚 Move openmc to codes * 🎨 Isolate eudemo from openmc * 🚚 Move neutronics under radiation_transport * Added volume tag * Added checks to ensure convexity. * Forced the pair of first and last pre-cells in the blanket to mate with the first and last pre-cell pair of divertor. * Fixed the convexity proble; but a different issue arose: floating point precision of the exterior points are all wrong. * Removed debug statement * Added tolerances back in * Floating point precision issue FIXEDgit add make_pre_cell.py openmc/make_csg.py Now we can use tighter tolerances. * 🔥 Remove old examples * integrated rad shield into the neutronics run * move set_volumes to CellStage --------- Co-authored-by: ocean <[email protected]> Co-authored-by: james <[email protected]> * 🎨 Radiation shield initial material * 📝 Remove neutronics example reference * 📝 Add notice for use and update environment * Improve tallies (#3275) * 🎨 Allow custom filter for tallies * 🐛 Fix tally function for eudemo run * 🐛 Fix remove reindex to avoid nans * Changed tallies to get the required data (need to multiply by appropriate constants in output.py to finish the job) * minor bug fix (missed a variable) * 🐛 Fix filter creation * 🐛 Fix output for temporary output --------- Co-authored-by: ocean <[email protected]> * ⬆️ Upgrade openmc --------- Co-authored-by: Ocean <[email protected]> Co-authored-by: jamesnha <[email protected]> Co-authored-by: Oliver Funk <[email protected]> Co-authored-by: Oliver Funk <[email protected]>
- Loading branch information
5 people
authored
May 14, 2024
1 parent
66e83e1
commit 6d64073