To address the challenges of wind speed representation distortion and insufficient dynamic response in ultra-short-term wind power prediction, this paper proposes a prediction method integrating high-fidelity inflow wind speed features with the TimeMixer model. First, considering the wind farm's terrain characteristics, turbine layout, and wake interference effects, a spatially corrected wind field is constructed to calculate the undisturbed inflow wind speed at hub height for each turbine. Second, using the incoming wind speed as the input feature, an improved TimeMixer model is developed, incorporating a dual-channel decomposition mechanism for trend and high-frequency fluctuation components. Through a multi-scale feature fusion strategy, the model's ability to jointly capture high-frequency fluctuations and low-frequency trends in turbine power evolution is enhanced. Finally, the proposed method is compared with the nacelle wind speed model, and validation is conducted using measured data from a mountainous wind farm in Qujing, Yunnan. The results show that the improved TimeMixer model with incoming wind speed input achieves improvements of 18.75%, 25%, 7.69% and 6.37% over the nacelle wind speed model in terms of root mean square error (E_rmse), mean absolute error (E_mae), r and Q_R respectively. Compared with the traditional temporal convolutional network (TCN) and long short-term memory (LSTM) network, the improved TimeMixer model exhibits superior performance in trend tracking and fluctuation response capture, with the correlation coefficient r improved by 2.46% and 5.04%, respectively. The proposed method effectively enhances ultra-short-term prediction accuracy and can provide technical support for the stable operation of power systems with a high penetration of renewable energy.

The aerodynamic characteristics of wind turbines operating in the wake region exhibit significant differences compared to those in free-flow conditions. In order to quantify the influence of wake on the aerodynamic characteristics of wind turbines, this paper proposes a method for calculating the aerodynamic characteristics of wind turbines based on a three-dimensional time-varying wake model (3DJGF-T) with a coupling-improved blade element momentum (BEM) model. Firstly, using sinusoidal and free-stream wind conditions as upstream turbine inputs, the 3DJGF-T wake model is employed to obtain the spatiotemporal characteristics of the wind field within the wake, and the external field experiments are conducted for validation. Secondly, utilizing the unevenly distributed flow field within the wake as the boundary condition in terms of spatial location points and time, the spatiotemporal variations of wind turbine rotor and single-blade power, torque, and axial force are investigated at different downstream longitudinal positions (x=5D, 6D, 7D, 8D) and horizontal positions (fully wake, 1/2 wake, 3/4 wake, and full-half wake) with the comparative analyses are performed. Results indicate that changes in the downstream longitudinal position will simultaneously alter the temporal and spatial characteristics of the wake, resulting in more complex fluctuations in load and power. With every increase of 1D in longitudinal distance, the rotor power increases by approximately 10%. In contrast, variations in the horizontal position of the wind turbine only affect the spatial characteristics of the wake, with rotor power losses gradually increasesing as the turbine approaches the center of the wake. Compared to the full-half wake, the relative average power losses of 1/2 wake, 3/4 wake, and full wake are 23.7%, 44.3%, and 61.2%, respectively. When the inflow is free wind, it exhibits faster variations in wind speed and stronger randomness compared to sinusoidal wind, causing greater fluctuations in the load power of the wind turbine.The findings of this study provide important reference value for wake regulation and micro-siting in wind farms.

Aiming at the problem that the current wind turbine wake model can only describe the wake distribution in the far wake region and ignores the wake characteristics in the near wake region, this paper derives a new three-dimensional wake model based on the double-Gaussian function, using the flow conservation theorem and through rotation correction. The wake model considers the influence of wind shear and is able to describe the three-dimensional wake distribution characteristics in the near wake region and the far wake region. Wind field experiments were carried out with two ground-based scanning laser radars. The experimental data shows that the distribution of near wake in the horizontal direction has the symmetrical double-Gaussian shape, and the distribution of far wake area has the symmetrical Gaussian shape, while due to the influence of wind shear in the vertical direction, the distribution of wake in the near wake area has the asymmetrical double-Gaussian shape, and the distribution of far wake area has the asymmetrical Gaussian shape. The horizontal and vertical profiles predicted by the three-dimensional wake model are compared and verified by using the measured data. The validation results show that the prediction curves of the three-dimensional wake model are in good agreement with the experimental data, and the average relative errors are mostly within 5%. The newly proposed three-dimensional wake model can better predict the spatial distribution of the whole wake area downstream of the wind turbine and can provide an optimization scheme for the layout of the wind farm.