No abstract

As wind energy becomes one of the fastest growing renewable energy resources, the control of large-scale wind turbines remains a challenging task due to its system model nonlinearities and high external uncertainties. In this paper, an adaptive neural pitch angle control strategy is proposed for the variable-speed wind turbines (VSWT) operating in pitch control region. The control objective is to maintain the rotor speed and generator power at the prescribed reference values in the presence of external disturbance, without the need of the information of system parameters and aerodynamics. First, the order of the system dynamics is increased by defining a filtered regulation error. By this means, the non-affine characteristics of the VSWT model is transformed into a simple affine control problem and thus the feedback linearization technique can be employed. The continuousness of control signal is also guaranteed to relax the requirement on the bandwidth of actuators, and the mechanical load on pitching systems is reduced. Subsequently, an online learning approximator (OLA) is utilized to estimate the unknown nonlinear aerodynamics of the wind turbine and extend the practicability of the proposed adaptive parameter-free controller. In addition, a high-gain observer is implemented to obtain an estimation of rotor acceleration, which rejects the need of additional sensors. Rigid theoretical analysis guarantees the tracking of rotor speed/generator power and the boundedness of all other signals of the closed-loop system. Finally, the effectiveness of the proposed scheme is testified via the Wind Turbine Blockset simulation package in Matlab/Simulink environment. Moreover, comparison results reveal that the introduced solution is able to provide better regulation performance than the conventional PI counterpart.

No abstract

No abstract

No abstract

No abstract

As wind energy becomes a larger part of the world's energy portfolio, the control of wind turbines is still confronted with challenges including wind speed randomness and high system uncertainties. In this study, a novel pitch angle controller based on effective wind speed estimation (EWSE) and uncertainty and disturbance estimator (UDE) is proposed for wind turbine systems (WTS) operating in above-rated wind speed region. The controller task is to maintain the WTS's generator power and rotor speed at their prescribed references, without measuring the wind speed information and accurate system model. This attempt also aims to bring a systematic solution to deal with different system characteristics over wide working range, including extreme and dynamic environmental conditions. First, support vector machine (SVR) based EWSE model is developed to estimate the effective wind speed in an online manner. Second, by integrating an UDE and EWSE model into the controller, highly turbulent and unpredictable dynamics introduced by wind speed and internal uncertainties is compensated. Rigid theoretical analysis guarantees the stability of the overall system. Finally, the performance of the novel pitch control scheme is testified via the professional Garrad Hassan (GH) bladed simulation platform with various working scenarios. The results reveal that the proposed approach achieves better performance in contrast to traditional L1 adaptive and proportional-integral (PI) pitch angle controllers.

Wind speed prediction is very important in the field of wind power generation technology. It is helpful for increasing the quantity and quality of generated wind power from wind farms. By using univariate wind speed time series, this paper proposes a hybrid wind speed prediction model based on Autoregressive Moving Average-Support Vector Regression (ARMA-SVR) and error compensation. First, to explore the balance between the computation cost and the sufficiency of the input features, the characteristics of ARMA are employed to determine the number of historical wind speeds for the prediction model. According to the selected number of input features, the original data are divided into multiple groups that can be used to train the SVR-based wind speed prediction model. Furthermore, in order to compensate for the time lag introduced by the frequent and sharp fluctuations in natural wind speed, a novel Extreme Learning Machine (ELM)-based error correction technique is developed to decrease the deviations between the predicted wind speed and its real values. By this means, more accurate wind speed prediction results can be obtained. Finally, verification studies were conducted by using real data collected from actual wind farms. Comparison results demonstrate that the proposed method can achieve better prediction results than traditional approaches.