Distributed Optimal Active Power Control of a Wind Farm based on Fast ADMM

Abstract

This paper presents a distributed method using the Fast Alternating Direction Method of Multipliers (Fast ADMM) for wind farm optimal active power control. The optimization problem is formulated as a quadratic programming problem by using piece-wise affine wind turbine (WT) model through Model Predictive Control (MPC). With the Fast ADMM, the MPC problem can be solved in a decentralized way with superior convergence properties. The calculation burden of the central controller is reduced by a small number of decentralized iterations. A wind farm was used as the test system, which consists of twelve WTs. Case studies were conducted and analyzed, which verifies the proposed method.