Benefits of an Advanced AC OPF Model in the Croatian Transmission Network

Modern transmission systems operation and planing problems involve computationally difficult optimizations. Transmission system operators forecast, monitor and manage power flows to ensure that the system always operates within permitted limits. This paper develops an AC Optimal Power Flow (AC OPF) algorithm intended to enhance numerical efficiency for commonly occurring cases in real-world transmission system networks. These systems include discrete devices, e.g. controllable transformer tap, generator automatic voltage control and variable shunt reactor (VSR), whose operational models are too computationally demanding for general nonlinear optimization. The developed algorithm divides the computational burden in two parts. The first part solves AC OPF with computationally demanding binary variables fixed to the initial power flow state, while the second part tries to find a better solution for fixed variables using more tractable AC OPF approximation. The efficiency of the approach is demonstrated on the Croatian transmission network and real operating data. The optimization succeeds to satisfy the initially exceeded operating constraints. Results also indicate that the direct approach that solves for all variables simultaneously using the exact nonlinear AC OPF model is computationally intractable for the considered cases.