Reproduce “Accelerating Optimal Power Flow with GPUs: SIMD Abstraction of Nonlinear Programs and Condensed-Space Interior-Point Methods” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Sungho Shin, François Pacaud, Mihai Anitescu
The authors develop a GPU solver for AC optimal power flow, choosing generator outputs to satisfy grid constraints and minimize cost. ExaModels.jl parallelizes repeated equation patterns, and MadNLP.jl reformulates the optimization step for fixed-pivot factorization. On case30000_goc, it takes 16.79 seconds versus 94.62 for CPU MadNLP; configurations differ in KKT formulation, and convergence is reliable only to 10⁻⁴.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.