Distributed Extremum-Seeking for Wind Farm Power Maximization Using Sliding Mode Control

Abstract

This paper introduces a sliding-mode-based extremum-seeking algorithm aimed at generating optimal set-points of wind turbines in wind farms. A distributed extremum-seeking control is directed to fully utilize the captured wind energy by taking into consideration the wake and aerodynamic properties between wind turbines. The proposed approach is a model-free algorithm. Namely, it is independent of the model selection of the wake interaction between the wind turbines. The proposed distributed scheme consists of two parts. A dynamic consensus algorithm and an extremum-seeking controller based on sliding-mode theory. The distributed consensus algorithm is exploited to estimate the value of the total power produced by a wind farm. Subsequently, sliding-mode extremum-seeking controllers are intended to cooperatively produce optimal set-points for wind turbines within the farm. Scheme performance is tested via extensive simulations under both steady and varying wind speed and directions. The presented distributed scheme is compared with a centralized approach, in which the problem can be seen as a multivariable optimization. The results show that the employed scheme is able to successfully maximize power production in wind farms.

Turbulence and Control of Wind Farms

“…Early wind farm control designs focused on power output maximization. Continuing work in this area employs approaches ranging from simple lookup tables (13,25) to closed-loop extremum seeking and proportional-integral (PI) control designs (see, e.g., [26][27][28][29][30]. Yaw control to maximize power output is another growing area of research (13,25,31,32).…”

A Robust Time-Delay Approach to Extremum Seeking via ISS Analysis of the Averaged System

“…Under the premise of the existence of extremum value, the ES control can search the extremum value without relying on the prior knowledge of the input-output mapping relationship. Because of its advantages of simple principle, low computational complexity and model free, ES control is used in many fields including anti-lock braking system [29], [31], aircraft formation flight [3], maximum power point tracking of new energy generation such as solar [14], wind [24] and fuel cells [32].…”