Skip to content

feat: implement Maximum Variance Analysis on Electric Field (MVAE)#6

Merged
Beforerr merged 1 commit intomainfrom
claude/implement-mvae-algorithm-3lZeg
Feb 13, 2026
Merged

feat: implement Maximum Variance Analysis on Electric Field (MVAE)#6
Beforerr merged 1 commit intomainfrom
claude/implement-mvae-algorithm-3lZeg

Conversation

@Beforerr
Copy link
Copy Markdown
Member

Add MVAE algorithm following Sonnerup & Scheible (1998). For a 1D current layer, the normal direction has maximum electric field variance (unlike MVAB where the normal has minimum magnetic field variance).

New public API:

  • mvae_eigen(E): eigendecomposition with max eigenvalue = normal direction
  • mvae(V, E): transform timeseries to MVAE coordinate system
  • check_mvae_eigen(F): quality check using λ₁/λ₂ ratio
  • convection_efield(v, B): compute E = -v×B proxy electric field

Also adds E_x1_error for MVAE error estimation, DimensionalData extension support, and comprehensive tests.

https://claude.ai/code/session_01NJD1MCzbYteJBHoFu6K3vC

Add MVAE algorithm following Sonnerup & Scheible (1998). For a 1D current
layer, the normal direction has maximum electric field variance (unlike
MVAB where the normal has minimum magnetic field variance).

New public API:
- mvae_eigen(E): eigendecomposition with max eigenvalue = normal direction
- mvae(V, E): transform timeseries to MVAE coordinate system
- check_mvae_eigen(F): quality check using λ₁/λ₂ ratio
- convection_efield(v, B): compute E = -v×B proxy electric field

Also adds E_x1_error for MVAE error estimation, DimensionalData extension
support, and comprehensive tests.

https://claude.ai/code/session_01NJD1MCzbYteJBHoFu6K3vC
@Beforerr Beforerr merged commit 26b4e73 into main Feb 13, 2026
1 of 2 checks passed
@Beforerr Beforerr deleted the claude/implement-mvae-algorithm-3lZeg branch February 13, 2026 15:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants