Disturbance observer-based decentralised power tracking control of wind farms

Wind farm normally involves a large number of wind turbines that are interactive due to air flow influence, making it interesting yet challenging to design a decentralised control scheme for each turbine unit so that stable power generation of wind farm is achieved. In this study, a decentralised adaptive control scheme is proposed for interconnected wind power generation systems in the presence of uncertain interaction among the turbines, capturing the maximum possible wind power. Based upon the disturbance observer technique, the unknown compounded disturbance is estimated. A speed function contributing to the decentralised control solution is introduced to improve the transient behaviour of the power tracking during the main course of the system operation so that the tracking error converges to a preassigned arbitrarily small compact set with a prescribed rate of convergence in a given finite time. The effectiveness of the proposed neuroadaptive tracking control strategy is verified through numerical simulation.