gurobi.optimods, ACOPF formulation with Branch Switching

Answered

Vincenzo Di Vito Francesco

February 05, 2026 17:01 Edited

• 

Hello everyone,

I am trying to run AC OPF with branch switching using Optimods on test cases beyond the provided example (e.g., IEEE-57, IEEE-118, etc.). While a “switching” version exists for the 9-bus case (case9-switching), similar switching-ready variants are not available for the larger test cases.

 

I would like to understand how such switching cases are typically constructed in practice. In particular, I tried augmenting the original networks by adding candidate lines (e.g., generator–generator connections) to enable branch switching, similarly to the ‘case9-switching.mat’. However, the optimization tends to converge to trivial topologies (e.g., all branches active).

 

Is there a standard procedure to obtain “switching” versions of larger test cases? For example, how candidate lines should be selected and parameterized? 

Any guidance on how to systematically build and use switching-enabled versions of larger IEEE test systems would be very helpful.

 

Thanks in advance,

Vincenzo

Related to

• Nonconvex
• Optimal Power Flow

0

• 

  Hassan Hijazi

  • Gurobi Staff

  February 27, 2026 19:31

  Hi Vincenzo, I recommend having a look here: https://data.openei.org/submissions/6197. The Grid Optimization Competition offers a number of large datasets where branch switching was allowed and sometimes led to better optimal solutions.

  0

Please sign in to leave a comment.