Skip to content

New RBF-ELM Models #277

Description

@fermingut

Add rbfelm.py, implementing two new surrogate models for fluid field interpolation across non-matching meshes:

  • RBFELM: A single Radial Basis Function augmented Extreme Learning Machine that maps spatial coordinates to field quantities (e.g. pressure coefficient). Supports configurable RBF center sampling and local/global gamma estimation.

  • MultiRBFELM: An ensemble of local RBFELM models for large-scale problems where a single model is computationally unfeasible. Partitions the input domain via MiniBatchKMeans and trains one model per cluster, with overlap zones at cluster boundaries and inverse-distance blending at inference to ensure continuity across interfaces.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

enhancementNew feature or request

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions