Geometry Optimization with Particle Swarm and Custom UIs

Project Overview:

This project implements particle swarm optimization (PSO) for geometry and design parameters, with custom UIs for defining optimization objectives, parameter bounds, and for visualizing results. Users can drive electromagnetic or other simulations with automated parameter search instead of manual sweeps.

Features:

  • Particle swarm optimization — Multi-parameter search with configurable swarm size, iterations, and objective (e.g. minimize S11, maximize gain).
  • Custom parameter creation — Define which geometry or simulation parameters are free variables and their bounds.
  • Custom UIs — Interfaces for setting up runs, monitoring progress, and inspecting outputs (e.g. S11, far field, autoparam).
  • Integration — Fits into simulation workflows (e.g. Remcom tools) so that each particle evaluation runs a simulation and the optimizer uses the result.

Particle swarm UI Custom UI for particle swarm optimization and parameter setup.

Architecture overview Architecture linking optimizer, parameters, and simulation.

S11 result Example S11 result from an optimized design.

Autoparam Parameter and autoparam visualization.

Remcom has shared this work publicly on LinkedIn.