Optimisation of dynamic reactive power sources using mesh adaptive direct search

Abstract

Dynamic reactive power sources can be used to effectively mitigate the fault‐induced delayed voltage recovery and transient voltage instability issues. When many var sources need to be installed at planned locations, optimisation of their sizes is a complicated non‐linear optimisation problem due to its non‐convexity and the dependence of the constraint on time‐series trajectories of post‐fault voltages. Solving this optimal sizing problem needs to utilise both a non‐linear optimisation solver and a power system differential‐algebraic equation solver. This study proposes a new approach for solving this problem under both a single contingency and multiple contingencies by using the mesh adaptive direct search algorithm interfaced with a power system simulator. The proposed approach is validated by case studies on a North American Eastern Interconnection model to optimise the sizes of planned STATCOMs against critical contingencies.

Keywords

• dynamic reactive power sources
• mesh adaptive direct search algorithm
• fault‐induced delayed voltage recovery
• transient voltage instability issues
• nonlinear optimisation problem
• nonconvexity