Non-centralized hierarchical model predictive control strategy of floating offshore wind farms for fatigue load reduction

Abstract

Wind power is increasing rapidly and specially offshore wind. Offshore wind offers certain advantages as more constant wind and additional space without large restriction. However, deep-waters require floating technologies. A key drawback of offshore wind is the reduced windows for operation and maintenance. Therefore, the use of optimal control algorithms that ensure the correct operation of the wind farm is essential. Offshore wind farms are usually oriented towards a defined direction of wind flow, so upstream turbines tend to provide more active power, carrying higher fatigue load, which results in uneven distribution of fatigue across the wind farms. Loads must be distributed among the members of the farm, to extend the farm and turbines life and reduce possible maintenance or breakage costs. Taking into account wake effects as well as hydrodynamic impacts which add additional motion and stress to the system, this paper presents a wind farm Model Predictive Controller (MPC) in order to optimize the loads of each wind turbine for life-extension. The results of the control show how the power generation is met and the load distribution are better balanced reducing system stress.