Fatigue damage of turbine components is typically computed by running a rain-flow counting algorithm on the load signals of the components. This process is not linear and time consuming, thus, it is non-trivial for an application of wind farm control design and optimisation. To compensate this limitation, this paper will develop and compare different types of surrogate models that can predict the short term damage equivalent loads and electrical power of wind turbines, with respect to various wind conditions and down regulation set-points, in a wind farm. More specifically, Linear Regression, Artificial Neural Network and Gaussian Process Regression are the types of the developed surrogate models in this work. The results showed that Gaussian Process Regression outperforms the other types of surrogate models and can effectively estimate the aforementioned target variables.

An open-source wind turbine controller, DTU Wind Energy Controller (DTUWEC), developed by Technical University of Denmark (DTU), is presented. The DTUWEC includes many advanced features and it is publicly available to the wind energy community. It is envisioned that this research work converts the old version of Basic DTU Wind Energy Controller into a generic, powerful and robust open-source wind turbine controller with a large number of developers and users in both research institutions and industries. DTUWEC is implemented in Fortran and offers the DNV-GL Bladed/OpenFAST style interface. Because of its open-source and modular characters, researchers are able to couple DTUWEC with different aero-elastic simulation codes to investigate the effect of different controllers on the wind turbine structural loads and power output, and to collaborate for future potential improvements.

The wake position is crucial feedback information for closed loop wind farm control. This paper presents a wake center position tracking approach based on turbine rotor loads, which is feasible to track the wake position while the wake is impinging on the downstream wind turbine. The rotor equivalent wind characteristics are estimated based on the blade root bending moments. Furthermore, a wake-wind model is developed to describe the relationship between rotor averaged wind characteristics and wake position, in which the velocity deficit profile of the wake is defined by the Gaussian model. An adaptive extended Kalman filter is proposed to estimate the lateral and vertical wake position states, in which wake dynamics is described by the dynamic wake meandering model. The wake meandering is driven by a colored noise with a covariance matrix that is gain scheduled according to the rotor averaged wind characteristic. Uncertainties of wake position detection are discussed in various combinations of the ambient atmospheric condition. Simulation results show that the proposed tracking method achieves good estimates in low ambient turbulence intensity, while, for a wind condition with high turbulence intensity, relatively longer time averaging needs to be applied to smooth out short-term fluctuations and highlight longer-term wake trends.