Search Results (105)

Short-term wind power forecasting is a key enabling technology for wind farm operation optimization, power grid dispatch, and electricity market decision-making. However, existing studies often lack unified standards in data partitioning, input feature construction, and hyperparameter tuning, making fair and reproducible comparisons across models difficult to achieve. To address this issue, this study focuses on day-ahead wind power forecasting for a single wind farm and establishes a benchmarking framework with strict chronological splitting, a shared feature information set, and a consistent hyperparameter tuning budget. Within this framework, six representative models, namely Ridge, XGBoost, LightGBM, DLinear, Transformer, and PatchTST, are systematically evaluated. A two-level evaluation protocol combining a fixed hold-out split and expanding-window rolling validation is adopted to compare model performance from multiple perspectives, including overall accuracy, sensitivity to hyperparameter tuning, robustness across rolling windows, and performance under typical operating scenarios. The results show that model rankings are not fully consistent between the hold-out evaluation and the rolling-validation setting. Under the fixed hold-out split, LightGBM achieved the lowest NRMSE of 10.2326%, while Transformer obtained the lowest NMAE of 6.9944%. In contrast, under the 8-fold expanding-window rolling validation, Transformer achieved the lowest average NRMSE of 8.1684%, followed by LightGBM with 8.7344%. These results indicate that the best performance on a single test split does not necessarily imply the strongest robustness across multiple time windows. In addition, strong tree-based models remain highly competitive in this single-wind-farm forecasting task, whereas more complex deep temporal models do not always deliver stable advantages. Meanwhile, the performance gains brought by hyperparameter optimization vary substantially across models, suggesting that conclusions drawn from default-parameter comparisons are of limited reliability. These findings demonstrate that systematic benchmarking under strict temporal constraints and fair tuning conditions is essential for clarifying the comparative performance, robustness, and engineering applicability of different model families. The study can therefore provide practical guidance for model selection and deployment in short-term wind power forecasting for single wind farms. Full article

Integrating renewable energy into modern grids while reducing carbon emissions represents a critical challenge for achieving “dual carbon” objectives. This paper proposes a two-stage stochastic optimization scheduling model for integrated energy systems (IES) that accounts for dynamic carbon emission factors and spatiotemporal uncertainty in wind power. First, a dynamic carbon emission factor model is developed to reflect real-time grid operational status and marginal power generation characteristics, replacing the conventional fixed-factor approach and enabling precise guidance for low-carbon electricity procurement strategies. Second, a Copula-based joint probability distribution model is established to capture complex temporal and spatial correlations in multi-wind-farm clusters, from which representative scenarios are generated and reduced through advanced pruning techniques. The scheduling model minimizes total operating costs and tiered carbon trading costs via mixed-integer quadratic programming (MIQP) and Benders decomposition. Case studies demonstrate that the proposed approach reduces daily operating costs by 6.4% (from 2.069 to 1.936 million yuan) and total carbon emissions by 8.4% (from 1051.8 to 963.2 tonnes) compared to conventional static-factor methods. Further, by accurately characterizing wind power uncertainty, the model achieves wind power absorption rates exceeding 90%, reducing curtailment from 272 kWh to 75 kWh and improving renewable energy utilization from 57.5% to 92%. The results validate that dynamic carbon factors and spatiotemporal correlation modelling effectively enhance both low-carbon performance and economic efficiency in IES dispatch, offering theoretical and practical guidance for achieving carbon-neutral energy system operations. Full article

With the large-scale integration of offshore wind farms (OWFs), harmonic issues caused by the interaction between high-frequency switching of converters and complex network impedances pose severe challenges to power quality. Traditional harmonic monitoring heavily relies on post-event fixed-threshold alarm mechanisms, which struggle to achieve early warning during the low-distortion sub-health operation stage and lack the capability for multi-dimensional tracing of harmonic degradation sources. To address these limitations, this paper proposes a deep warning network for grid harmonics combining multi-step regression and multi-dimensional tracing within a unified multi-task learning (MTL) architecture. First, a deep shared feature encoder, integrating a bi-directional long short-term memory (Bi-LSTM) network with a multi-head self-attention (MHSA) mechanism, is utilized to extract high-order temporal coupling features between meteorological evolution and multi-node electrical states. Subsequently, the main task branch executes a k-step-ahead multivariate time-series regression to accurately predict the evolution trend of total harmonic distortion (THD) at both the point of common coupling (PCC) and the turbine terminal. Simultaneously, the auxiliary task branch performs multi-label micro-state classification based on relative degradation thresholds, achieving fine-grained multi-dimensional tracing covering spatial nodes, electrical attributes, and their joint micro-states. Experimental results on real-world OWF operational data demonstrate that through the joint optimization of regression and tracing tasks, the proposed MultiDimKStepMTL model significantly improves time-series prediction accuracy, achieving a $10.3\%$ relative improvement over single-task baselines, while substantially reducing computational overhead. This research successfully advances grid harmonic monitoring from passive response to proactive micro-state early warning, providing a solid, highly interpretable data-driven foundation for active filter control of offshore wind clusters. Full article

Wind energy is a cornerstone of the global transition to renewable and sustainable energy systems. However, the same meteorological processes that generate this clean energy can also produce extreme wind events that threaten the structural integrity and operational reliability of wind turbines and power grids. Therefore, accurately predicting extreme wind speeds is a critical link between promoting clean energy and ensuring infrastructure resilience. Traditional models often struggle to capture the multimodal characteristics of extreme wind speeds under complex meteorological conditions due to fixed distribution assumptions or unstable training of mixture models, leading to estimation biases that undermine planning reliability and may result in catastrophic turbine failures or overly conservative designs. To address these issues—particularly weight imbalance and overfitting–this study proposes an enhanced regularized extreme value mixture model (ERDC-EVMM). This method integrates elastic network regularization and Kullback–Leibler divergence constraints within a Mixture of Experts framework, and employs K-means initialization and momentum-based training to enhance convergence stability. Validated using daily extreme wind speed sequences from coastal and inland wind farms, the model outperforms standard GEV and mixture models in terms of goodness-of-fit, percentile accuracy, and return period estimates, while achieving a convergence speed that is more than 30% faster (82 iterations). By balancing accuracy and training stability, the ERDC-EVMM model provides a reliable statistical tool for extreme wind speed forecasting, supporting the safe expansion of wind energy infrastructure and the design of climate-resilient communities. Full article

Wind energy stands as one of the most technologically mature renewable sources, playing a pivotal role in the mitigation of greenhouse gas emissions. However, wind farms and associated infrastructures increase collision risk for flying organisms. Implementing higher cut-in speeds is a proven mitigation strategy to significantly decrease wildlife mortality rates, particularly for bat species, by preventing turbine operation during low-wind periods of high activity. The suggested, non-standard, increased cut-in speed for wind turbines is generally 5.0 m/s. To test the effectiveness of cut-in speed increase, bat activity was monitored at three wind farms in northern Portugal (Gevancas, Azinheira, and Lagoa de Dom João e Feirão), to characterize spatial and temporal activity patterns and assess the potential associated risk. Ultrasonic acoustic detection was carried out at fixed stations, at heights of 55 m above ground level from March to October. Wind speed data were recorded concurrently using anemometers mounted on meteorological towers. Contradicting recommendations, the results show that significant bat activity might occur at wind speeds above the current curtailment values. Since turbine operation coincides with peak bat activity, it is imperative to implement site-specific mitigation strategies, such as optimized cut-in speeds, to minimize mortality risk. Full article

The rapid growth of renewable energy and the inherent volatility of wind power grid integration have imposed stringent requirements on power system security and economic operation. To address this challenge, energy storage systems (ESSs) are widely adopted as flexible regulation tools; however, their high capital costs make the shared energy storage model a more efficient and viable solution. This paper proposes an optimal configuration model for wind farms participating in shared energy storage (SES) based on cooperative game theory. First, integrating wind power output forecasting data and market electricity price information, a wind-storage combined optimization model accounting for wind power uncertainty is first established. Subsequently, a core pricing strategy integrating the core allocation rule with the Vickrey–Clarke–Groves (VCG) auction mechanism is proposed to realize the fair allocation of energy storage resources and effective revenue incentives. Finally, comparative experiments between the proposed core pricing mechanism and the fixed pricing mechanism verify its superiority in terms of social welfare, budget balance, and allocation fairness. The results demonstrate that the proposed mechanism not only enhances the overall social benefits of the wind-storage system but also effectively ensures the incentive compatibility of all participants and the stability of the alliance, providing feasible theoretical and methodological support for the economic dispatch of wind-farm-shared energy storage. Full article

Offshore wind power currently shows the trend of a larger wind turbine capacity and deep-sea wind farm sites. Traditional wind farms with single-type wind turbines can hardly accommodate the dual requirements of high-efficiency power generation and wake effect mitigation of wind farms. Moreover, the wind shear of fixed wind turbines and the platform motion of floating wind turbines result in insufficient adaptability to a hybrid wind farm with multi-type wind turbines. To address that issue, this paper takes offshore wind farms with a multi-type hybrid layout for wake optimization. Firstly, based on the wind shear model, the influence of hub height difference for fixed wind turbines is analyzed, and the platform motion of semi-submersible floating wind turbines is evaluated through MoorDyn. On this basis, the wake optimization strategy for maximizing the total power generation of a wind farm is proposed based on the Gaussian Curl Hybrid model, which realizes three-dimensional wake control by considering the hub height difference and floating platform motion of multi-type wind turbines. The case study demonstrates that the multi-type hybrid layout itself has inherent wake suppression and optimization potential. The fixed wind farm with a row–column hybrid layout achieves an average power generation efficiency of 65.25%, which is superior to the single-type layout. For the floating wind farm with an inner–outer hybrid layout, the displacement misalignment effect is significant, with a maximum offset of 21.66 m in surge and 10.32 m in sway, and the total power is increased by 6.87 MW. And a hierarchical wake control mode matching multi-type wind turbines is formed. It provides a novel wake regulation mechanism for the design and operation of hybrid offshore wind farms. Full article

Floating wind turbines are becoming increasingly mainstream. Floating wind turbines have strengths in terms of cost-efficiency compared to conventional wind turbines. Comparisons between floating and conventional wind turbines can be easily conducted through simulations. In this paper, we estimate the CAPEX and LCOE of fixed and floating offshore wind farms following the INNU cost model with LIR rotor technology and hybrid floater technology. Full article

The offshore wind energy sector faces significant logistics costs due to complex maritime environments. This study addresses the multi-period Crew Transfer Vessel routing within offshore wind farms and scheduling problems through a novel mixed-integer linear programming framework. The model integrates personnel inventory management with dynamic service times. It determines optimal routing and scheduling plans to minimise total operational costs. Numerical experiments demonstrate the effectiveness of the approach. The results indicate that increasing vessel capacity from 8 to 20 reduces total expenses by approximately 80%. Moreover, shifting from single-trip to multi-trip operations decreases fixed charter costs by 30%. The computational performance is efficient, and the solver achieves optimal solutions within an average of 5.67 s. This framework provides operators with precise decision support for complex offshore wind farm maintenance scenarios. Full article

This study compares the performance of two wind farm sites located in Northern Europe: an onshore site and an offshore area in the eastern Baltic Sea region. This study investigates the optimisation of wind farm performance within a fixed project area by maximising annual energy production (AEP) and increasing energy density. Three wake-loss scenarios (≤10%, ≤15%, and ≤20%) were examined to assess the sensitivity of layout optimisation to aerodynamic interaction constraints. Several layout configurations were analysed to reduce wake losses and enhance overall energy output. Wind conditions were assessed using NORA3 reanalysis data, and wake interactions were modelled using the Jensen wake model to estimate AEP. Both wind farms were further compared across key criteria, including cost, power generation efficiency, installation and maintenance requirements, and site availability. Offshore wind farms achieve 1.5–1.7 times higher energy density under similar spatial conditions. However, offshore levelised cost of energy (LCOE) remains roughly 25% higher due to higher capital and infrastructure costs, while onshore LCOE demonstrates better economic performance, driven by lower CAPEX and O&M expenses. The findings highlight the trade-offs between cost efficiency and wake-driven energy performance for onshore and offshore wind development in the eastern Baltic Sea region. Full article

This work focuses on evaluating the fatigue characteristics of a 66 kV wet-type submarine cable. The analysis incorporates realistic offshore environmental conditions and considers installation through a flexible pull-in system. The target system was deployed at the Southwest Offshore Wind Farm in Korea, where the structural response of the cable is governed by combined wave- and wind-induced loading. To determine the ultimate limit state (ULS) behavior of the cable arranged in the flexible pull-in installation configuration, a global dynamic analysis was initially performed, from which time-dependent displacement histories were extracted under site-specific metocean conditions. The results indicate that lateral wave-induced loading dominates the global response owing to the vertically suspended configuration of the flexible pull-in installation system. A detailed local fatigue analysis was performed based on the critical loading scenarios identified from the global analysis. Prior to the fatigue evaluation, a numerical modeling approach for the local analysis of the 66 kV wet-type submarine cable was developed and validated through comparisons with experimental tensile test results. The developed numerical model successfully captured the axial stiffness characteristics and deformation response of the cable, demonstrating a discrepancy of less than 1% between the numerical predictions and experimental measurements. Fatigue analyses were performed for 120 displacement-based time-history loading cases derived from the global dynamic response. The results show that the fatigue damage is predominantly concentrated in the steel-wire armor, particularly near the fixed boundary region, identifying it as the governing fatigue-critical component. In addition, a parametric study of the wire-armor helix angle revealed that increasing the helix angle significantly improved the fatigue life by enhancing the bending flexibility and reducing the stress concentration under transverse loading. These findings provide practical insights into the fatigue-resistant design and optimization of wet-type submarine cables installed using flexible pull-in installation systems in floating offshore wind applications. Full article

Wind turbine (WT) misalignment fault diagnosis is constrained by critical signal processing challenges: weak fault features, intense background noise, and poor generalization. This study proposes a lightweight method for high-precision fault diagnosis. A fixed-threshold wavelet denoising method with the scene-specific pre-optimized parameter a (0 < a ≤ 1.3) is proposed: the parameter a is determined via offline grid search using the feature retention rate (FRR) as the objective function for typical wind farm operating scenarios. A multi-scale depthwise separable CNN (MS-DSCNN) captures multi-scale spatial features via 3 × 1 and 5 × 1 kernels, reducing computational complexity by 73.4% versus standard CNNs. An attention-based minimal peephole LSTM (AttMPLSTM) enhances temporal feature measurement, using minimal peephole connections for long-term dependencies and channel attention to weight fault-relevant signals. Joint L1–L2 regularization mitigates overfitting and environmental interference, improving model robustness. Validated on a WT test bench, the Adams simulation dataset, and the CWRU benchmark, the model achieves a 90.2 ± 1.4% feature retention rate (FRR) in signal processing, an over 98% F1-score for fault classification, and over 99% accuracy. With 2.5 s single-epoch training and a 12.8 ± 0.5 ms single-sample inference time, the reduced parameters enable real-time deployment in embedded systems, advancing signal processing for rotating machinery fault diagnosis. Full article

Submarine power cables for offshore wind farms experience continuous cyclic loading from environmental forces and floating-platform motions, making fatigue performance a critical design factor. This study combined global and local analyses to investigate the fatigue behavior of a 66 kV dry-type submarine cable installed using a flexible pull-in installation system. A global dynamic analysis using site-specific meteorological and oceanographic data provided time-series displacement responses that were used to evaluate the fatigue damage to the metallic components of the cable. The results indicated that the minimum fatigue life of 8.71 × 10⁴ cycles occurred at the upper metallic sheath near the fixed end, with a corresponding cumulative damage of 1.147 × 10⁻⁵. Fatigue accumulation was predominantly governed by lateral (y-direction) displacement, while axial and vertical displacement components contributed minimally. Furthermore, the predicted fatigue life of the metallic sheath varied by a factor of up to 3.6 depending on the selected curve, comparing the cyclic stress amplitude and number of cycles to failure (S–N curve), highlighting the importance of accurate material fatigue data. These findings emphasize the need for careful evaluation of the environmental loading and sheath fatigue properties in flexible pull-in installation system-based submarine cable system designs. Full article

Aiming at the nonlinear and non-stationary characteristics of wind speed series, this study proposes a hybrid forecasting framework that integrates Variational Mode Decomposition (VMD), Sparrow Search Algorithm (SSA), and Long Short-Term Memory (LSTM) networks. First, VMD is employed to adaptively decompose the original wind speed series into multiple Intrinsic Mode Functions (IMFs) with distinct frequency features, thereby reducing the non-stationarity of the original sequence. Second, SSA is utilized to adaptively optimize key parameters of the LSTM network, including the number of hidden units, learning rate, and dropout rate, to enhance the model’s capability in capturing complex temporal patterns. Finally, the predictions from all modal components are integrated to generate the final wind speed forecast. Experimental results based on 10-min resolution measured data from a coastal wind farm in southeastern China in June 2020 show that the model achieves a Root Mean Square Error (RMSE) of 0.208, a Mean Absolute Error (MAE) of 0.161, and a Mean Absolute Percentage Error (MAPE) of 3.284% on the test set, with its comprehensive performance significantly surpassing benchmark models such as LSTM, VMD-LSTM, MLP, XGBoost, and ARIMA. The limitations of this study mainly include the use of only one month of data for validation, the lack of segmented performance analysis across different wind speed regimes, and a fixed prediction horizon of 10 min. The results indicate that the proposed hybrid forecasting framework provides an effective approach with practical engineering potential for ultra-short-term wind power prediction. Full article

As the proportion of renewable energy generation in the power grid continues to rise, the operational state of the power system changes frequently with fluctuations in renewable power output. However, the traditional fixed-weight multi-objective reactive power optimization method lacks the necessary flexibility and adaptability, as it is unable to dynamically adjust the priority levels of different objectives based on real-time operating conditions (such as load fluctuations and changes in network structure). As a result, its optimization decisions may deviate from the system’s most urgent economic or security needs. To address this issue, this paper proposes an adaptive multi-objective reactive power optimization control method. The proposed approach formulates the objective function as the weighted sum of system active power loss and voltage deviation at the grid connection point, with weight coefficients adaptively adjusted based on the voltage deviation at the grid connection point. First, the relationship between voltage fluctuations at the offshore wind farm grid connection point and active/reactive power output is analyzed, and a corresponding reactive power allocation model is established. Second, taking into account the input–output characteristics of wind turbine generators and static var compensators, a reactive power control model is constructed. Third, considering offshore operational constraints such as power and voltage limits, a weighted variation particle swarm optimization algorithm (WVPSO) is developed to solve for the reactive power control strategy. Finally, the proposed method is validated through tests using a practical offshore wind farm as a case study. The test results demonstrate that, compared with the traditional fixed-weight multi-objective reactive power optimization approach, the proposed method can rapidly adjust the priority of each optimization objective according to the real-time grid conditions, achieving effective coordinated optimization of both active power loss and voltage at the grid connection point, and the voltage deviation is kept within 5%, even with power system fluctuations. In addition, compared with the traditional PSO algorithm, for various test situations, WVPSO exhibits above 15% improvement in solution speed and enhanced solution accuracy. Full article