A power smoothing strategy without energy storage system by utilizing the large rotational inertia of the wind turbine rotor to achieve power smoothing is proposed to solve the problem of output power fluctuations of modern large-scale wind turbines under rated wind speed. Firstly, the inherent characteristics of the wind turbines are employed to provide an immediate and accurate equivalent wind speed for control participation. Secondly, the upper and lower limits of the rotational speed are set for each wind speed segment based on high aerodynamic efficiency zone and wide speed-ratio design, and the entire inertia control process is given simultaneously to achieve the energy storage and release of the wind turbine rotor. Subsequently, multiple prediction models based on the slope of the wind speed pulsation trend are constructed via big data sources to realize the predictive action of the torque. Finally, the comparative simulation experiments of the power output of the wind turbine are carried out by BLADED simulation software and a certain actual wind farm. The results show that the adopted strategy stabilizes the output power effectively without the additional energy storage devices.

A load reduction method was provided by optimizing the aerodynamic configuration and structural design of the blades in this paper. The GH-Bladed computing platform was used to verify the blade performance and turbine load based on 6.25 MW wind turbine. The study finds that reducing the blade chord length has the most significant effect on reducing the blade root load, while the effect of reducing thickness on load reduction is not prominent. However, reducing blade thickness is a convenient method to improve aerodynamic performance while also weakening load. In addition, the laying thickness of the spar cap structure can change the blade flapwise stiffness to affect flapwise deformation. Expanding the blade flapwise deformation by adjusting the thickness of the blade spar cap to release concentrated stress is a novel method to promote the reduction of blade load generation.

Wind power forecasting is a crucial topic in the field of wind energy generation. With the increasing penetration of renewable energy in power systems, wind energy, as a rapidly developing renewable resource, necessitates accurate short-term forecasting for the energy industry. In this paper, a Gaussian Mixture Model-gated Recurrent Unit （GMM-GRU） based on Gaussian graph convolution is proposed. Firstly, the XGBoost algorithm optimized by Particle Swarm Optimization （PSO） is utilized to construct a feature selection network for identifying important features. Secondly, using the graph auto-encoder and cosine correlation fusion method, a network graph to effectively capture the potential long-distance correlations among sites and accurately describe the spatial correlation of multi-site features are built. Finally, the Gaussian Mixture Model （GMM） is employed to deeply extract the intrinsic relationships within the network graph, and the Gated Recurrent Unit （GRU） integrates its spatio-temporal correlations to address the power forecasting problem. The experiments on real wind farm data are carried out. The comparisons with other models demonstrate that the proposed model significantly enhances wind power forecasting performance.

This study investigates the influence of second-order wave forces on the dynamic response of a semi-submersible floating wind turbines, using the "Guoneng Sharing" as a case study. A comparative analysis of the differences between far-field and near-field methods for calculating the second-order mean wave force transfer functions （QTFs） is presented. A coupled time-domain model of aerodynamic, hydrodynamic, structural, and servo dynamics was established to analyze the dynamic responses of the floating wind turbine under four scenarios： neglecting second-order wave forces, considering second-order mean wave forces, considering second-order difference wave forces, and considering both second-order difference and sum wave forces. The results indicate that the QTFs calculated by the far-field and near-field methods are nearly identical when the angular frequency is less than 1.1 rad/s. However, as the angular frequency increases, the far-field method generally yields larger values than that of the near-field method. The second-order wave forces significantly impact the motion response of the semi-submersible platform, with the primary contributing components being the second-order mean wave forces or second-order difference wave forces, while the second-order sum wave forces exhibit negligible influence. Notably, the second-order difference frequency and sum frequency wave forces significantly affect the acceleration of the semi-submersible platform and the load at the top of the tower, whereas the second-order mean wave forces have no discernible effect on either. The high-frequency response of the tower top load is excited under the influence of second-order difference and sum frequency wave forces and must be taken into account in calculations. Furthermore, second-order wave forces markedly affect the tension and fatigue life of the mooring anchor chains, not only increasing the fatigue damage but also influencing the location of maximum fatigue damage occurrence.

To address the limitations of traditional resource allocation methods, this study analyzes the process of optimizing resource allocation strategies for offshore wind farm construction under meteorological uncertainty. A multi-objective optimization framework is proposed, incorporating four key performance indicators： minimum total working time, resource allocation optimization rate, window period utilization rate, and cost-benefit ratio. A discrete-event digital simulation-based optimization model is developed to support this framework, enabling systematic evaluation of the benefits of resource allocation strategies through statistical analysis of these indicators.The Laotian Monsoon Wind Power Project is selected as a case study for simulation. Results demonstrate that the established model can effectively simulate the entire construction process of offshore wind farms, providing real-time visualization of meteorological conditions, wind turbine unit statuses, and personnel dynamics. It also accurately quantifies critical metrics including minimum working time, resource allocation optimization rate, window period utilization rate, and cost-benefit ratio. The model also to significantly enhances resource allocation during wind farm installation, improves overall construction efficiency, and provides valuable pre-construction planning support. These outcomes collectively contribute to achieving the dual goals of cost reduction and efficiency improvement in offshore wind farm development.

A fault diagnosis method for wind turbine rolling bearings is proposed, which combines the gradient dynamic selective alignment （DSA） method, dual-channel feature fusion technique （DCFF）, and deep learning. Firstly, the acquired one-dimensional time series data are transformed into two different two-dimensional image representations using recursive mapping （RP） and Markov transfer field （MTF）. These transformed images are then input into an encoder for reconstruction into a low-dimensional representation. Next, a dual-channel feature fusion convolutional neural network （DCFFC） is constructed, embedding an improved attention mechanism within FRNet to extract in-depth feature information. A second-level feature fusion is performed to complete the feature classification. Finally, a gradient masking matrix is constructed using the data reconstruction loss of the decoder and the feature extraction network classification loss to achieve gradient dynamic self-adaptation throughout the fault diagnosis process. Experimental validation using the Paderborn dataset demonstrates that the proposed method achieves a diagnostic accuracy of 99%, effectively extracting fault-specific diagnostic information.

Wind turbine gearboxes are subjected to contact loads during operation, resulting in contact fatigue damage in the gears, which significantly affects their service life. To accurately predict the contact fatigue life of gears, this study takes a 6 MW wind turbine gearbox operating at Dabancheng Wind Farm, Xinjiang, as the research object. A gearbox dynamic model considering both internal and external excitations is established to calculate the dynamic meshing forces of each helical gear pair. The analysis incorporates local material properties and residual stresses in the gears. By applying the Dang Van multiaxial fatigue criterion combined with S-N curves, the fatigue damage distribution in the helical gears under contact loading is analyzed. Fatigue hotspots are identified, and critical operating conditions are extracted to develop a surrogate model, which quantifies the contact fatigue damage and estimates the contact fatigue life of each gear over a 20-year design period. The results indicate that the engage-in and recess points of external meshing pairs in helical gears are the primary fatigue hotspots, making them more susceptible to fatigue failure. Among all gears, the sun gears exhibit the most severe contact fatigue damage, with estimated contact fatigue life of 8 years for the high-speed sun gear and 21 years for the low-speed sun gear. Conversely, the planetary gears in both stages show relatively lower fatigue damage, with their contact fatigue life exceeding 30 years. The ring gear experiences the least fatigue damage, resulting in an exceptionally long service life.

To investigate the dynamic response of the multi-bucket jacket foundation of offshore wind turbines under storm surge load, this paper conducts a series of 1 g model tests using the motor servo random dynamic loading equipment. Then, the influence of different depth-to-diameter ratios of the suction bucket on the soil and structure is further investigated through numerical simulations. The results indicate that the extreme horizontal displacement of the structure rises as wind speed increases under storm surge conditions, and the amplification coefficient of horizontal displacement follows a trend of “increasing initially before decreasing”. The peak value of the Fourier amplitude spectrum for the nacelle position acceleration increases with the load, while the peak frequency exhibits a pattern of “decreasing first and then increasing”. As the depth-to-diameter ratio of the suction bucket increases, both the extreme horizontal displacement and the amplification of the structures coefficient corresponding decrease. A depth-to-diameter ratio of L/D= 0.5 is the least favorable, as it cannot withstand storm surge loads. The difference between depth-to-diameter ratios of 1.0 and 1.5 is negligible. Taking into account reliability and economy, a depth-to-diameter ratio of 1.0 is the optimal type. Different depth-to-diameter ratios also have the influence onthe rotation center of the SBJ foundation.

To enhance the accuracy of ultra-short-term wind power forecasting and support efficient power system dispatch, this study proposes an integrated multi-algorithm forecasting model. The variational mode decomposition （VMD） is firstly applied to suppress noise and reconstruct the original power sequence. In the modeling stage, a long short-term memory （LSTM） network is adopted to capture temporal dependencies, followed by a multi-layer convolutional neural network （CNN） to extract local features. A Self Attention mechanism is further incorporated to dynamically focus on critical time steps, resulting in a collaborative multi-module forecasting framework. To evaluate the performance of the proposed model, ablation studies, comparative experiments, and seasonal transfer tests were conducted using data from a wind farm in Shandong Province, China. The results show that, compared to baseline models, the proposed model reduces the mean absolute error （MAE） by 23.5% and the root mean square error （RMSE） by 20%, highlighting the advantages of each module in modeling complex temporal patterns. Additional validations in cross-regional （wind farms in Central and Western China） and cross-energy （photovoltaic plants） scenarios further demonstrate the model's strong generalization capability.

Under low wind speed conditions, conventional maximum power point tracking （MPPT） control strategies fail to effectively account for the dynamic influence of the pitch angle, leading to reduced wind energy utilization efficiency. Under medium and high wind speed conditions, traditional PI-based pitch angle controllers struggle to maintain stable power output. To address these challenges, this paper proposes a collaborative optimization strategy that integrates Physics-Informed Neural Networks （PINNs） with Model Predictive Control （MPC）. The PINN model accurately captures the dynamic characteristics of wind turbines, while an online training mechanism based on a stochastic weight adjustment algorithm enhances adaptability. A unified MPC strategy is developed, considering both torque and pitch angle as control variables. By dynamically adjusting the cost function, the proposed approach achieves maximum power tracking under low wind speed conditions and ensures stable power output under medium and high wind speed conditions. Experimental results validate that the proposed method significantly improves the dynamic response of the wind turbines, enhances real-time control performance, and increases wind energy utilization efficiency.