Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Simulation Tools topic
No spam. Unsubscribe anytime.
Argonne’s PowerSaaS uses semi‑analytical solutions in Julia to speed reliability analysis
Summary
Wei Gao of Argonne National Laboratory demonstrated PowerSaaS, a Julia implementation of semi‑analytical solutions (SAS) and holomorphic embedding that improves convergence and runtime for power‑flow and dynamic studies; the team published source code and reported performance gains on small and medium test cases.
Wei Gao of Argonne National Laboratory presented PowerSaaS (Julia), a high‑performance transmission planning and reliability analysis tool that implements semi‑analytical solutions (SAS) and holomorphic embedding to address convergence and speed issues in power‑flow and dynamic simulations.
Gao described SAS as a high‑order Taylor‑series approach that represents differential‑algebraic solutions analytically, avoiding iterative Newton‑Raphson steps for the dynamic portion and enabling larger time steps. For power‑flow initialization, Gao used holomorphic embedding to guarantee a solvable path from an easy baseline problem to the target solution and to avoid Newton‑Raphson nonconvergence in difficult cases.
Case studies included a Polish system stability assessment and long‑duration simulations where the Julia implementation outperformed a MATLAB version on small to medium test cases by orders of magnitude; Gao noted ongoing work to optimize memory allocations on very large systems where the Julia build remains a work in progress. The team published a MATLAB version (powsas.m) and is making Julia sources available to encourage adoption and interfaces with PSSE, MATPOWER and other tools.
Gao said SAS can improve numerical robustness and execution speed for contingency, stability, and long‑horizon simulation workflows and that the team will continue to optimize the Julia implementation and expand interfaces and scenario coverage.
