Skip to content

ACTINV v1.2.0

Choose a tag to compare

View all tags
@Connoravila1 Connoravila1 released this 22 Sep 01:45
· 829 commits to master since this release
v1.2.0
9808219

ACTINV v1.2.0 — release notes

ACTINV 1.2.0 is a minor release: it adds time-varying irradiation spectra, multi-spectrum uncertainty
propagation, a decay-photon source export for downstream transport, and a set of isomer-resolution and
decay-data improvements that move the identical-data benchmark score. The solver, schemas and data
catalog remain backward-compatible with 1.1.x; the embedded catalog stays data-v1.1.0.

Upgrade the Python package with:

python -m pip install --upgrade actinv

Rust CLI users can install this exact version with:

cargo install --locked --force actinv-cli --version 1.2.0

What changes

  • Per-step spectra. Each schedule step may declare its own spectrum; the step's flux multiplier
    scales that spectrum's total. Runs with step spectra use the groupwise library path and are
    validated against the library's group count. This enables activation under time-varying fields
    (pulsed operation, shifting blankets, schedule-driven spectral drift) without manual per-step runs.
  • Multi-spectrum uncertainty. MF=33 covariance propagation collapses jointly over (spectrum, row)
    parameters, so a run with per-step spectra still reports propagated cross-section, fission-yield and
    decay-constant uncertainty. Identical overrides reduce to the bitwise-identical single-spectrum
    path.
  • activity.total uncertainty response. Total activity is propagated through the full covariance
    as a first-class response rather than a per-nuclide RSS combination.
  • actinv export-openmc-mesh. Exports decay-photon sources for downstream photon transport: one
    IndependentSource per mesh cell, box-sampled inside recorded bounds, with the cell's discrete
    spectrum and absolute strengths.
  • Isomer resolution. Emitted-state products resolve to decay-library isomer levels through the
    MF=1 state table with rank-compressed LFS mapping; a LIS label fallback plus a --decay-fallback
    second sublibrary recover channels that previously decayed to artificial homes. On the identical-data
    FNS corpus this moves the within-30% experiment count from 69 (parity with the frozen FISPACT-4
    reference) to 71, with median |log C/E| 0.103 vs 0.105.
  • Robustness and numerics. Perturbation channels draw mean-preserving lognormal factors from an
    eigen-clipped PSD covariance; CRAM applies a selective refinement gate and exempts subnormal-scale
    rows from convergence forcing. The zero-spectrum collapse fallback fix from the unreleased queue is
    included.
  • Data-side disclosure. A benchmark-corpus defect screen identified a spurious high-energy capture
    bump in the TENDL-2017 In-115 evaluation (capture exceeding the nonelastic bound at >10 MeV, fixed in
    TENDL-2025). A ledgered repair exists for the research corpus and is described in ledger.md entry
    53; it is not part of the shipped data-v1.1.0 catalog.

Qualification boundary

Unchanged — docs/DATA_LIMITATIONS.md and docs/QUALIFICATION.md apply as before. New capabilities
extend the envelope (time-varying spectra, photon-source handoff, multi-spectrum UQ); they do not
alter the validation status of the underlying solver or data.

Assets 7
Loading