Skip to main content

Physics simulation — DLA

Diffusion-limited aggregation (DLA) models how randomly diffusing particles stick together to form branching clusters—used in physics, materials science, and computer graphics.

Problem​

Simulating DLA in a clear, repeatable way for coursework or demos: balance correctness (rules of random walk + aggregation) with performance so runs finish in reasonable time on typical hardware.

Solution​

Implement the classic algorithm on a 2D lattice (or continuous variant): seed a cluster, release particles that perform random walks from a boundary or injection ring, and attach them on first contact with the cluster. Visualize growth frame-by-frame or as a final aggregate.

Impact​

  • Makes nonlinear growth and fractal-like structures tangible for students and stakeholders.
  • Provides a baseline for comparing parameters (stick probability, lattice vs off-lattice, noise).

Lessons learned​

  • Early visualization beats staring at numeric logs: bugs in neighbor checks show up immediately.
  • Injection strategy (where new walkers start) strongly affects runtime and cluster shape.
  • For large clusters, consider optimized data structures (sparse maps, distance fields) before scaling particle count.

Tech stack​

  • Python for simulation logic and experimentation
  • JavaScript Canvas for visualization/rendering