Search

The increasing size and clustering of wind turbines have amplified wake effects, reducing wind farm power generation. For this reason, a multi-priority control strategy based on axial induction control was proposed to enhance the total power output of the wind farm in this paper. Firstly, a wind speed time-delay prediction model based on LiDAR for tandem turbines was constructed, by employing a time-delay processing algorithm to refine the wind speed engineering model and integrating a neural network model for the axial induction factor at the rotor. After that, a multi-priority control strategy, defining turbine priorities based on wake effects and adjusting power distribution via the axial induction factor, was proposed to maximize power capture for wind farms. Simulink and FAST co-simulations shown that, under steady-state wind input conditions, the multi-priority control strategy increased power output of the wind farm by 6.62 %, compared to the maximum power point tracking strategy. Finally, preliminary hardware-in-the-loop experiments demonstrated that the control strategy did not have a negative impact in a semi-physical environment, providing theoretical support for subsequent ground-based wind turbine experiments. • A wind speed time-delay prediction model for tandem turbines is constructed. • A multi-priority control strategy for wind farms based on axial induction control is proposed. • Simulink and FAST co-simulations are carried out to validate the performance of the control strategy. • The feasibility of the multi-priority control strategy is preliminarily verified through hardware-in-the-loop experiments. [ABSTRACT FROM AUTHOR]

This study explores how renewable energy communities (RECs), leveraging regional resources and community involvement, can contribute to advancing sustainable energy transitions. The proposal is an integrated framework that combines an optimization model with the principles outlined in the EU Renewable Energy Directive (RED II). The model optimizes the allocation of photovoltaic (PV) and wind capacity while facilitating power sharing and addressing investment and operational strategies within RECs. The framework accounts for factors such as geographic location, regulatory constraints, and financial incentives, aiming to maximize resource utilization and promote energy self-sufficiency. Sensitivity analyses and case studies demonstrate the model's effectiveness, providing insights into optimal revenue generation and investment strategies. By linking theory with practice, this study advances the understanding of REC initiatives and highlights their role in accelerating the adoption of renewable energy. Our findings offer valuable guidance for policymakers and REC stakeholders, emphasizing the importance of integrated approaches in energy transition planning. • Proposes an integrated framework optimizing photovoltaic and wind energy allocation for Renewable Energy Communities (RECs). • Combines optimization modeling with EU Renewable Energy Directive (RED II) principles. • Addresses investment, operational strategies, power sharing, and regulatory constraints. • Provides actionable insights for policymakers and stakeholders to enhance energy self-sufficiency and resource utilization. [ABSTRACT FROM AUTHOR]

This study presents an integrated modeling framework for the design and optimization of a sustainable energy hub tailored to coastal urban areas, using a case study. The model addresses critical energy demands-electricity, heating, cooling, and water supply-while integrating renewable energy sources (RESs) such as wind turbines, photovoltaic thermal (PV/T) systems, and microturbines. A novel two-level optimization approach is adopted, combining "design" and "environmental design" levels to evaluate both technical performance and environmental impacts. The water cycle algorithm (WCA) is employed to enhance optimization efficiency, ensuring precise utilization of RESs while minimizing costs and emissions. Results demonstrate the system's capability to dynamically adapt to environmental conditions, such as wind speed and solar radiation, optimizing energy production across seasons. The optimized energy hub achieves a total energy generation of 14810 kW for electricity, 25210 kW for heating, and 53748 kW for cooling, with an overall 28 % reduction in operational costs compared to traditional setups. Detailed economic analysis highlights the PV/T system's significant role, constituting nearly 60 % of the initial investment while ensuring sustainable and efficient operation. Moreover, seasonal wind energy potential contributes substantially to electricity supply during peak consumption periods. This work underscores the feasibility and benefits of integrating RESs in energy hubs for coastal regions. By providing a robust framework for energy planning, it offers a blueprint for sustainable urban development with significant implications for reducing carbon emissions and improving resource efficiency. • Balancing technical & environmental aspects of energy hub design in an integrated way. • Boosting efficiency with microturbines, solar panels via optimized renewable energy models. • Addressing sustainability challenges through environmental design & systematic analysis. • Driving innovation & efficiency with water cycle algorithm for energy optimization. • Validating framework through modeling ensures robustness, reliability & applicability. [ABSTRACT FROM AUTHOR]

Cost-minimizing electricity system models are important tools for understanding conditions for the development of the electricity system. Since the variability of wind and solar power outputs differs between years, a satisfactory representation of variability requires a high time resolution, as well as data that cover multiple decades. This work proposes a weather-year selection method that represents power generation variability by selecting a set of weather years to represent the net-load variability of a broader span of historical weather years (in this work, 39 years). The representativeness is captured in terms of net-load amplitude and duration, such that the electricity demand, as well as the wind- and solar-generation profiles, are considered in their chronologic order, rather than simply as discrete data-points. The weather-year selection method is applied to modeling the North European electricity system with the aims of evaluating the method and investigating the impacts of extreme net-load events on the electricity system composition. The results show that the proposed method can represent the net-load variability of multiple decades using a few selected weather-years. In addition, when the probability of extreme net-load events is accounted for, these extreme events mainly increase the peak thermal capacity and long-term biogas fuel storage capacity. • Proposes and evaluates novel methodology for selecting representative weather-years. • Benchmarks model results using 39 simultaneously modeled weather-years. • As Few as 3 weather-years may suffice to represent 39 years of variability. • Investigates the impact of extreme net load events on optimal system composition. • Thermal peak capacity with fuel storage manages extreme net load events at low cost. [ABSTRACT FROM AUTHOR]

Utilizing hydropower to mitigate the variability of wind power and photovoltaic has been proven to be an effective strategy for enhancing their utilization. However, the integration scale depends largely on hydropower regulation capacity. This paper compares the technical and economic differences between pumped storage and electrochemical energy storage enhancement modes for hydro-wind-photovoltaic systems. Pumped storage retrofits involve adding pumping stations between adjacent reservoirs. Two detailed coupling models are developed, and a fine-grained simulation optimization approach is used to capture operational details. Moreover, economic indicators are established from an engineering project perspective to evaluate their profitability. Taking a cascaded hydropower in China as a case study. The results show that: (1) Pumping station mode has 2.58 times more annual incremental revenue than battery storage mode. The differences can be attributed to energy storage and transmission capacity occupations variances. (2) Considering the high replacement cost of batteries, the net revenue of the battery storage mode over the project's life is negative and economically infeasible. In contrast, the net revenue from the pumping station mode amounts to 251 million CNY. (3) Initial battery cost must be reduced to at least 75 %, and transmission capacity needs to be increased to realize the economics. • Two hydropower storage retrofit modes are assessed technically and economically. • The optimal energy storage enhancement in Chinese hydropower is identified. • Pumping station retrofit is superior in storage duration and power absorption. • Initial cost and channel capacity are critical for battery retrofit. [ABSTRACT FROM AUTHOR]

In the context of large-scale wind power grid integration, accurate wind power forecasting is crucial for optimizing grid scheduling and ensuring safe grid connection. This study proposes a recurrent attention encoder–decoder network for multi-step interval wind power forecasting, combining Numerical Weather Prediction (NWP) inputs with deep learning techniques. The approach leverages a sequence-to-sequence neural network and temporal attention mechanism, enabling better capture of latent patterns in historical data that are useful for future predictions, directly generating multi-step time series and final prediction intervals. Additionally, a moving window training scheme, integrating bifurcated sequences and hidden layers, is employed to organize historical data and improve the stability and performance of the sequence. Using offshore wind farm data, the wind speed and direction components (U, V) are decomposed, and experiments show that the proposed method outperforms existing methods in metrics such as a minimum PINAW of 0.119 and an average reduction of 19.37% in CWC. These results demonstrate high accuracy and reliability in interval forecasting, providing strong support for wind farm scheduling and grid optimization. [Display omitted] • Propose AMQRNN: Integrate NWP data to enhance wind power forecast accuracy. • Use Seq2Seq and Attention for direct multi-step time series prediction. • Innovative training: Forked sequences and moving windows improve model performance. • Model understanding: Context fusion boosts accuracy of historical data predictions. [ABSTRACT FROM AUTHOR]

Integrating renewable energy in power systems can significantly reduce emissions in the energy sector, resulting in remarkable environmental and human health benefits. One of the major barriers for renewable energy integration is transmission congestion, which can be effectively mitigated through optimal utilization of transmission flexibilities. Distributed flexible AC transmission systems (D-FACTS) are cutting-edge devices that can provide premium flexibility to electric power transmission systems when optimally allocated and configured. However, the optimal D-FACTS allocation and configuration problem is extremely computationally challenging. This study aims to present a computationally efficient algorithm that can optimally allocate and configure variable-impedance D-FACTS to minimize (1) power system operating costs, (2) global warming potential (GWP), and (3) human toxicity potential (HTP), considering uncertainties in load and renewable energy generation. The model was implemented on a modified RTS-96 test system with a high penetration of wind energy, and results show that optimally allocating and configuring D-FACTS can reduce power system operating costs, GWP, HTP, and renewable energy curtailment. The results also indicate an inverse relationship between the first objective and the other two, showing the necessity to choose a proper trade-off between cost savings, environmental and human health impacts. • A multi-objective optimization model is proposed to allocate and configuredistributed flexible AC transmission systems. • The optimization model minimizes not only power system operating costs but also environmental and human health impacts. • A custom-made evolutionary algorithm is proposed to solve the optimization problem with high computational efficiency. • The model can optimally decide the locations as well as the quantity and set points of D-FACTS modules at each location. • The inverse relationship between the objectives, namely, the power system operating cost and the GWP/HTP, is analyzed. [ABSTRACT FROM AUTHOR]

Thermal power has been encouraged to collaborate with large-scale variable renewable energy (VRE) in hybrid power systems (HPSs). To optimize the VRE capacity and explore the deep peak-shaving performance of ultra supercritical units, we propose a comprehensive evaluation framework that optimizes the VRE size integrated with thermal power. First, we proposed a scenario-based uncertainty modeling method considering wind and PV power correlations. Second, a short-term daily optimization model was designed, along with a two-stage adaptive complementary strategy. Finally, a lifecycle techno-economic evaluation framework was established to determine the cost-effective VRE size. The framework was applied to a case study in Qinghai Province, China. The results show that (1) The optimal VRE capacity configuration is 2.5 times that of thermal power, including 3333 MW of wind power and 1667 MW of PV power, yielding a net present value of 4.43 billion CNY. (2) Converting traditional units to ultra-supercritical units can effectively improve the efficiency of coordinated operation and lead to more than 15% decrease in unit generation costs. (3) Relaxing power transmission constraints can effectively improve the operating performance of thermal units, and the unit generation cost accordingly decreases by 22.5%. • A novel uncertainty modeling method was developed considering VRE power correlations. • A two-stage adaptive complementary strategy is used to enhance the complementarity. • A short-term operation optimization model of the wind-PV-thermal HPS was designed. • A techno-economic evaluation model was used to determine the optimal VRE capacity. • The impacts of different system components on the sizing scheme are quantified. [ABSTRACT FROM AUTHOR]

To alleviate the intermittency and fluctuation of power output caused by renewable energy, a spectral splitting solar and wind complementary power generation system with hydrogen storage technology is proposed in this study. Through spectral splitting, the full spectrum utilization of solar energy is achieved, which enhancing the hydrogen production from methanol decomposition significantly. Then, the stored hydrogen doped with methane is converted into electricity by a natural gas combined cycle to guarantee the electric demand without any deficiencies. This study focuses on the scale configuration optimization based on the complementarity evaluation of natural resources. To gain comprehensive insights into the proposed system, energy, exergy and economic analyses are performed under the optimal scale configuration. Results show that the fluctuation of solar and wind power output for three cities with different meterological conditions are mitigated, while the total energy utilization efficiency ranges from 34.64 % to 38.70 %. Highest exergy loss is observed in the solar and wind power generation processes, accounting for around 38.85 %–49.60 % of the total exergy input for three cities. Overall, the feasibility of solar and wind hybrid power generation is demonstrated in this study, which providing a promising approach for the sustainable energy production and utilization. • A solar and wind hybrid power system integrated with hydrogen storage is proposed. • Full spectrum solar utilization is achieved by a spectral beam splitter. • Scale configuration based on complementarity of solar and wind energy is carried out. • Assessing the impact of meteorological conditions and scale configurations. [ABSTRACT FROM AUTHOR]

As electrochemical energy storage (EES) becomes increasingly prevalent in electricity markets, accurately assessing their techno-economic performance is crucial. This paper introduces the novel concept of the techno-economic feasible region (TEFR) for EES participation in electricity markets, providing a new analytical framework for optimizing EES market strategies. The main contributions are: 1) A bilevel game-theoretic model is developed for both independent energy storage (IES) and wind-storage system (WSS) to capture the complex interactions between EES and the independent system operator (ISO); 2) A systematic method is proposed for determining TEFR, and geometric metrics including surface area, normalized volume-to-surface area ratio, and roundness are introduced to quantify TEFR characteristics. Based on the IEEE 24-bus system analysis, it is found that WSS exhibits 10–15 % larger operational boundaries than IES, but with similar utilization efficiency; 3) Sensitivity analyses reveal that charge/discharge power has the most significant impact on TEFR characteristics, with the normalized volume-to-surface area ratio increasing by 10.8 % and roundness rising by 124 % when power increases from 28 MW to 30 MW. This study provides an important theoretical foundation for optimizing EES design and market participation strategies. • A novel concept of the TEFR for EES participation in electricity markets is proposed. • The TEFR of wind-storage system is 10–15 % larger than independent energy storage. • Impact of storage capacity, charge/discharge power, and wind power forecast error on TEFR is investigated. [ABSTRACT FROM AUTHOR]

The deployment of renewable energy technologies, though necessary to decarbonise our society, poses a risk stemming from the massive increase in the use of critical raw materials. This work presents a prospective evaluation of a national electricity generation mix up to 2050 and discusses the increase in several critical and strategic materials used in this transition. Results indicate that the deployment of solar photovoltaics and wind energy will raise material criticality concerns in the coming decades. When comparing a decarbonisation scenario aligned with the 2030 Spanish policy with a business-as-usual scenario, results show that a higher penetration of renewables would involve increases of up to 53 % in silicon, 27 % in aluminium, 11 % in copper, and less than 1 % in other materials by 2050. Overall, the decarbonisation scenario would involve up to 12 % more materials. Furthermore, criticality indicators show increases of 0.06 % and 5 % by 2050 depending on the selected indicator. Differences in figures highlight discrepancies in the way criticality is evaluated, suggesting that further research is needed. Nevertheless, national long-term energy policies such as the Spanish one are urged to implement criticality issues in their formulation. Consequently, the authors recommend including critical material usage within energy and climate planning models. • A national power system optimisation model is created to evaluate decarbonisation scenarios. • Quantitative results confirm the rising demand for critical and strategic raw materials, and increasing criticality scores. • By 2050, the demand for critical and strategic raw materials for power generation in Spain could be 12 % higher. • Criticality measures of decarbonisation scenarios show slight relative increases. • Authors recommend including critical raw materials in energy planning models. [ABSTRACT FROM AUTHOR]

To address the challenges associated with the variability of wind power and to optimize the use of renewable energy, this study investigates the integration of large wind farms with a power-to-gas system. This integration is designed to convert surplus wind energy into hydrogen, thereby enhancing grid stability and reducing energy curtailment. In this paper, an optimal planning and operation strategy is proposed for a large wind farm coupled with a power-to-gas system. The key contributions include: (1) a mixed integer linear programming-based switch control model that accurately captures the realistic operation of the electrolyzer within the power-to-gas system, and (2) a novel distributionally robust optimization method that considers variability in wind speed distributions. The electrolyzer requires a continuous and stable output to preserve its lifespan, and the switch control model ensures realistic operational conditions. The proposed distributionally robust optimization addresses regional variations in wind speeds, balancing robustness with flexibility. A case study demonstrates that the proposed approach outperforms conventional methods. Furthermore, incorporating mixed integer linear programming-based constraints for the electrolyzer switch control led to a 20 % improvement in economic performance over systems without power-to-gas integration and reduced curtailed energy from the wind farm by up to 94.5 %. • A novel ambiguity set based distributionally robust optimization. • Mixed linear integer programming-based switch control model for hydrogen production. • Optimal planning and operation of large wind farm system with power to gas. • Improvement of potential renewable energy sources penetration rate. [ABSTRACT FROM AUTHOR]

In response to the global climate crisis, South Korea has committed to achieving net-zero emissions by 2050, requiring a transformation of its energy system. This study explores the potential sector coupling between the power and heating sectors, referred to as power-to-heat (P2H) in South Korea's 2050 net-zero energy system. Using the open-source EnergyPLAN model, we simulated the future energy scenario with government-projected data and assumptions about energy infrastructure. EnergyPLAN effectively models interactions within systems where distribution is critical, such as electricity, heat, and gas. South Korea's net-zero scenario served as the baseline input, allowing us to assess feasibility and quantify P2H's role in supporting net-zero goals. Our findings suggest that by 2050, South Korea's projected infrastructure could lead to an overbuilt system, with electricity and heating capacities exceeding demand. Variable renewable energy (VRE) capacity is expected to surpass hourly needs, even with storage and sector coupling. Annually, electricity supply may exceed demand by about 89 TWh, with a target demand of 1257 TWh. In district heating, approximately 4.7 TWh of surplus VRE could be used by P2H, meeting only 14.5 % of heating demand, indicating limited absorption of the surplus. Sensitivity analyses on flexible resources, such as electricity and thermal storage, showed limited cost-effectiveness. Increasing wind power's share rather than solar PV is recommended to enhance net-zero feasibility, given South Korea's capacity factors. • The net-zero energy system in 2050 of South Korea was evaluated. • The variable renewable energy (VRE) can be managed by sector coupling. • An increase in the share of district heating increase Power-to-Heat (P2H) potential. • The net-zero scenario of Korea has excessive solar PV capacities. • It is required to adjust the electricity mix to reinforce the future energy system. [ABSTRACT FROM AUTHOR]

Most of the current design methodologies for ducted wind-turbines rely on an uncoupled design-procedure of the duct and the rotor. The aim of this work is to assess, for the first time, the reliability of this strategy by quantifying the differences with an advanced fully-coupled approach. Therefore, two different aero-designs are performed using these strategies. In the uncoupled approach the duct geometry is optimized to maximize the ingested mass flow rate disregarding the rotor presence. Then, a free-vortex disk is introduced, and its load is varied to maximize the power coefficient based on the device frontal-area (C P , e x). In the coupled design, the duct geometry and the rotor load are simultaneously optimized. In both cases, the CFD-actuator-disk is used as analysis method along with a gradient-based optimizer. The coupled strategy yields a higher C P , e x (0.68) adopting a compact diffuser and high stagger-angle. Contrarily, the uncoupled procedure leads to a lower C P , e x (0.59) using a large chord and a moderate stagger. A physical explanation of these differences is offered. Finally, while the presented cases refer to a given shape of the duct, the conceptual generality of the study in terms of the effective reliability of the uncoupled design remains valid in general. • Comparison of uncoupled and coupled design strategies to maximize the C P , ex of DWTs. • Coupled and uncoupled designs are carried out via optimization using the CFD-AD. • The uncoupled design leads to a high-chord low-stagger duct with a reduced C P , ex • The coupled design yields a compact high-performance device with C P , ex = 0. 68 > 16 / 27 • A physical interpretation of the results proposed by the design methods is offered. [ABSTRACT FROM AUTHOR]

This study aims to employ Genetic Algorithms (GA) to optimize the Archimedes Wind Turbine's (AWT) structure. The distinct design of the AWT necessitates a different methodology than standard lift-type wind turbines. Computational Fluid Dynamics (CFD) was used to evaluate the performance of the design based on the SST k-ω model. In order to validate the simulation results, the acquired data was compared with those from earlier design studies. Simulation results showed that the differences between the two studies' Mean Absolute Error (MAE) were only 5.9 %. GA was selected in an iterative link with ANSYS software to find the optimal turbine's structure. This study investigated several scenarios in which the design variables including opening angle, pitches, and rotational speed were changed, either separately or in combination. In the most comprehensive scenario, an optimized AWT with opening angle of 63.49°, Tip Speed Ratio (TSR) of 1.12, pitch1 value of 115.03 mm, and pitch2 value of 389.54 mm was obtained, where the search space includes all parameters. This design resulted in a power coefficient equal to 0.2644, which shows a 27.72 % increase in efficiency and 7.94 % reduce in thrust force. • A Genetic–CFD algorithm has been developed to achieve the optimal design of an Archimedes Wind Turbine (AWT). • CFD and SST k-ω model used for performance evaluation. • Comparing simulation results to previous studies, Mean Absolute Error (MAE) was 5.9 %. • Optimal AWT with 27.72 % efficiency increase and 7.94 % thrust force reduction achieved. [ABSTRACT FROM AUTHOR]

Wind power generation is considered an important green renewable energy source due to its ability to meet the world's power requirements over time. However, optimizing the layout of a wind farm to alleviate the wake effect and maximize power generation is a challenging optimization problem due to being non-convex and NP-hard. Several optimization approaches have been recently proposed for tackling this problem; however, they still suffer from low quality of final results and slow convergence speed. Therefore, this study proposes a new, effective approach, namely GBOT, based on the gradient-based optimizer and the recently proposed encoding mechanism. Despite that, GBOT still suffers from a slow convergence rate as the number of wind turbines increases. Therefore, it is improved by replacing the updating scheme of GBO with a novel one to aid in improving the exploration and exploitation performance along the optimization process. This improved variant is known as IGBOT. Both GBOT and IGBOT are compared with several state-of-the-art methods based on two wind scenarios. This comparison is conducted in terms of several performance metrics, including best power output, average power output, worst power output, standard deviation, Wilcoxon rank sum test, Friedman mean rank, multiple comparison test, and convergence curve. • Novel optimization methods for solving wind farm layout problem. • Rapidly and accurately detect the near-optimal location for each turbine. • Validating the performance of proposed methods using two different scenarios. • Comparing effectiveness of proposed methods with several state-of-the-art methods. • The results obtained are better than those obtained by other methods. [ABSTRACT FROM AUTHOR]

Optimizing wind farm layouts is critical to maximizing wind power generation. The wake effect significantly impacts turbines located downwind, making farm layout a key determinant of power generation efficiency. Traditional algorithms often overlook the value of leveraging historical information, which can lead to entrapment in local optima. Our survey reveals that previous studies on wind farm layout optimization (WFLO) have not adequately integrated the historical data of particle swarm optimization (PSO) with reinforcement learning's empirical pool, resulting in the loss of valuable information. Here, we present a novel approach that enhances algorithm development and exploration by utilizing historical data and integrating proximal policy optimization from reinforcement learning with an experience pool. This method markedly outperforms the conventional genetic PSO in terms of performance. Extensive numerical experiments across wind farms of various sizes and four distinct wind scenarios demonstrate the superior efficacy of our reinforcement learning-based particle swarm optimization (RPSO) algorithm compared to 12 state-of-the-art methods. Under four wind scenarios, the average power conversion efficiencies of RPSO for the three turbine scales reach 98.68%, 98.14%, and 97.33%, respectively, underscoring the high competitiveness of the proposed RPSO for WFLO in diverse wind conditions. • A novel combination of PPO combined with PSO for WFLO is presented. • Use RL's experience playback to dynamically adjust PSO parameters. • A reasonable blend of RL's experience pool and PSO's historical information. • PPO significantly improves the convergence and performance of the PSO. • RPSO achieves state-of-the-art in WFLO under different scenarios. [ABSTRACT FROM AUTHOR]

As wind power continues to develop, advancing wind power prediction becomes more and more crucial. This study focuses on advancing wind power prediction by addressing data privacy and enhancing model applicability at multi-spatial scales, from individual turbines to entire farms. Traditional methods are typically confined to a single scale, lacking flexibility in application, requiring extensive data from farms, which potentially compromises energy data privacy. To tackle these challenges, we introduce an innovative Divide-Merge Federated Learning with Active Private Clustering (D-M APCFed) approach. This approach strategically employs federated learning to train models within privacy-preserving boundary, overcoming the adverse effects of wind power data heterogeneity through a novel APC method and knowledge merge technique. The primary innovation of this study is a scalable and accurate wind power prediction model that operates effectively at multi-spatial scales while safeguarding energy data privacy. In case study of two spatial scales, the D-M APCFed approach achieves an average prediction accuracy of 87.11 % in the twelve federated farms and 81.69 % in the twenty federated turbines. This approach enables a more generalized model through the secure use of data from diverse sources at multi-spatial scales, enhancing prediction accuracy and ensuring the confidentiality of sensitive information. [Display omitted] • Propose a novel federated learning approach scalable across multi-spatial scales. • This approach could balance energy data privacy and prediction accuracy. • Active Private Clustering method can offset negative impact of data heterogeneity. • Transferability and scalability of the approach are validated in different scenarios. [ABSTRACT FROM AUTHOR]

Predicting wind power is crucial for wind farm operations and power system stability. Most existing prediction methods use cabin wind speed as the input variable, but but few of them correct the wind speed data and consider the correlation between input data. This paper proposes a hybrid probabilistic neural network model for short-term wind power probabilistic prediction, which primarily consists of two deep neural network models connected in series. The first model corrects SCADA wind speed using an LSTM neural network with mechanism information. The second model uses a self-attention mechanism to strengthen the correlation among input time series and constructed a probabilistic prediction model named SA-DeepAR. Using real wind farm data to verify results shows the corrected wind speed improves power prediction accuracy by 44 %, and the prediction accuracy of the SA-DeepAR model improved by about 15 % in RMSE and MAE compared to the DeepAR model, and by about 6 % in R2. In terms of probability prediction, the SA-DeepAR model can still maintain an average prediction interval coverage probability of 95 % at a 40 % confidence level. The proposed model can predict short-term wind power generation effectively and offer reliable data for decision-making. • A hybrid neural network model is proposed for short-term wind power probability prediction. • Correcting cabin wind speed using mechanism information and an LSTM neural network. • Applying the self-attention mechanism to model the interrelation among variables. • The power probability prediction results show a narrow confidence interval with high coverage probability. [ABSTRACT FROM AUTHOR]

Harnessing wind power for hydrogen production is a promising solution to tackle the challenges of climate change. However, the high cost associated with wind-powered hydrogen production systems emerges as a significant barrier. Hence, this study aims to present an economic analysis of wind-powered hydrogen production systems, taking into account the heterogeneity of landforms. The analysis utilizes hourly wind speed data from 290 anemometer towers throughout China, providing a comprehensive and accurate assessment of wind power generation. The findings reveal that landform heterogeneity has a significant impact on the levelized cost of hydrogen (LCOH), but such impact is expected to diminish over time. Moreover, the LCOE in various provinces ranges from ¥0.31–0.52/kWh, and the LCOH ranges from ¥25.75–36.31/kg, with an average of ¥30/kg. Electricity costs and CAPEX account for over 75 % of the total hydrogen production costs. Inner Mongolia, Xinjiang, and Jiangsu exhibit the lowest LCOH due to their lower wind power costs. Furthermore, the predicted results indicate a swift descent in the LCOH from 2020 to 2030. By the years 2030 and 2060, the nationwide average LCOH is projected to reach ¥25/kg and ¥22/kg, respectively, signifying an 18 % and 28 % reduction compared to the year 2020. [Display omitted] • Landform heterogenous is considered to analyze the cost of wind-to-hydrogen system. • Hourly wind speed data from 290 anemometer towers throughout China is utilized. • Levelized cost of hydrogen across China ranges from ¥25.75–36.31/kg. • Inner Mongolia, Xinjiang, and Jiangsu own the lowest LCOH. • Average LCOH is projected to reach ¥25/kg and ¥22/kg, respectively, by 2030 and 2060. [ABSTRACT FROM AUTHOR]

Accurate wind speed forecasting is essential for optimizing energy dispatch and enhancing grid stability. This study presents a novel hybrid wind speed forecasting model (WSDTW-CLA), emphasizing the spatial transfer characteristics of wind speed while mitigating the inherent errors in existing models. The proposed method employs the Wind Speed Dynamic Time Warping (WSDTW) algorithm to align wind speed data from neighboring stations, effectively facilitating the capture of spatial transfer patterns during the preprocessing phase. This alignment generates a wind speed spatial matrix that incorporates future-relevant information, providing precise input for forecasting module. The model employs a hybrid neural network combining a convolutional neural network (CNN), a long short-term memory (LSTM) network, and an autoencoder (AE) to predict wind speeds by establishing feature connections from the preprocessed data. The performance of the WSDTW-CLA model is evaluated using seasonal datasets from March, June, September, and December in Yunnan Province, China. A multi-step comparative analysis involving seven established models and seven sub-models within the proposed framework demonstrates that the WSDTW-CLA model significantly outperforms other similar models, with all evaluation metrics showing improvements of over 30 %. This proposed method enhances the utilization of wind energy resources, thereby promoting the advancement of the wind power industry. • Prediction of wind speed with spatial transmissibility. • Wind speed dynamic time warping (WSDTW) strategy is introduced to match the future wind speed sequence of the target site. • A hybrid wind speed forecasting model, WSDTW-CLA, is introduced utilizing WSDTW, convolutional neural network (CNN), long short-term memory neural network (LSTM), and autoencoder (AE). • Explains the utilization and transmission of spatial wind speed information within the prediction process. [ABSTRACT FROM AUTHOR]

Wind farms are becoming larger, presenting opportunities to improve their efficiency and effectiveness. Although wind farm control can be used to adjust a wind farm's power output to meet grid requirements, its development is restricted by the limited flexibility of wind turbine operation. Wind turbines are typically designed to operate at the power output determined by the wind speed, making it difficult to adjust their power output quickly and easily in response to changes in grid demand. A new approach to wind farm control that provides full flexibility for both wind farms and turbines is proposed. This method adjusts the wind farm's power production using predictive control of each turbine to meet the grid demands. The power generated by each turbine is adjusted to achieve this, keeping fluctuation in the wind farm power output low. A wind farm simulation is conducted under varying wind speeds using the discretized C MEX Matlab/SIMULINK R ○ model of the 5 MW Supergen turbine and the wind turbine model from NREL. The results demonstrate that the wind farm power output tracks a reference value that can be set by the grid. The results are also analyzed on the torque/speed plane, and they indicate that the turbines operate within their specified safe operating zones under both normal and gusty wind operating conditions at the same time. • A wind farm control method to provide full flexibility for wind farms and turbines is discussed. • It adjusts the wind farm's power output using control of each turbine to meet the grid demands. • The fluctuation in the wind farm power is kept low by the proposed at the same time. • The proposed wind farm control method is scalable, hierarchical, and decentralized. • Each turbine has a feedforward MPC for improved performance and to cope with gusts better. [ABSTRACT FROM AUTHOR]

Ultra-short-term wind power forecasting (UWPF) is crucial for the large-scale grid connection of wind energy. The state grid in China has strict multi-step forecasting requirements, which pose challenges to on-site efficiency and accuracy. This paper proposes a comprehensive framework of the hybrid method for UWPF with on-site application, consisting of data decomposition, model prediction, and post-processing. In the first stage, rolling decomposition and feature reconstruction are employed to decompose the wind power data into sub-components without future information leakage. The feature extraction and model matching processes are then performed to make full use of different machine learning prediction models and distinct characteristics of wind power components. Finally, a novel error tracking strategy is proposed to enable real-time error correction for multi-step forecasting by capturing the fluctuation characteristics of wind power data. The proposed framework is evaluated on two field wind power datasets through comparative experiments with five benchmark methods and nine reference methods. The experimental results demonstrate that (a) Each module of the proposed method effectively contributes to improving forecasting accuracy. (b) The proposed method significantly outperforms traditional machine learning methods in both single-step and multi-step forecasts, indicating its capability to handle practical UWPF tasks effectively. • Comprehensive decomposition-based framework for ultra-short-term wind power forecast. • No future information leakage during the data decomposition stage. • Forecasting with extracted component feature and superior model matching strategy. • Real-time error tracking and correction for multi-step wind power forecasting. • On site application at an offshore and an onshore wind farm in China. [ABSTRACT FROM AUTHOR]

Wake effects within offshore wind farms not only impact the overall energy output but also increase the structural fatigue loads of wind turbines. In this work, we propose a wake redirection control (WRC) strategy for power and fatigue multi-objective optimisation. In particular, the steady-state engineering wake model is further augmented by incorporating a load assessment feature, so that evaluating both power and fatigue behaviours in WRC design becomes possible. More specifically, a steady-state aerodynamic wake model is used to evaluate the power output, while the wind turbine fatigue behaviour is predicted by a load indicator derived from aero-elastic simulations covering a wide range of waked inflow and yaw-offset conditions. Based on the wake model and the load indicator, multi-objective particle swarm optimisation is then used to locate the optimal wind farm yaw settings for both power optimisation and load mitigation. In order to demonstrate the effectiveness of the proposed strategy, WRC design is performed for a 3 × 3 offshore wind farm, and the results have been verified against the state-of-the-art multi-physics engineering tool FAST.Farm. It is shown that the proposed multi-objective WRC strategy could achieve a 7.49% overall power increase and a 2.15% tower-bottom fatigue load reduction, while suppressing the growth of blade-root fatigue load at the same time, over-performing the other WRC designs with different control objectives. This study provides an efficient way of structural fatigue evaluation under combined wake interactions and yaw-misalignment, enabling power and fatigue multi-objective optimisation for offshore wind farm WRC design. • A wake redirection control (WRC) framework is proposed for offshore wind farm power enhancement and load mitigation multi-objective optimisation. • The proposed WRC framework is based on a look-up table, constructed by a wake surrogate model, a fatigue load indicator, and a swarm-based optimiser. • A fatigue load indicator has been developed to facilitate the rapid assessment of wind turbine structural loads. • WRC design objectives with various combinations of overall power production, tower and blade fatigue loads, have been comparatively studied. [ABSTRACT FROM AUTHOR]

The combined pumped storage and wind power system (CPSWPS) generates the problem of low-frequency oscillations, which leads to complex dynamic characteristics and regulation control. Thus, this paper presents an optimization of the shafting and excitation dual damping controller (SEDDC) for CPSWPS. Firstly, detailed equations of the CPSWPS are constructed. Then, a novel shafting damping controller (SDC) and SEDDC are designed and applied in CPSWPS. Furthermore, an improved sparrow search algorithm that combines random walk and golden sine strategy (RGSSA) is proposed. Finally, an optimization method of SEDDC is presented, and its robustness and universality are verified. The results indicate that the designed SDC has more regulation advantages than traditional rotor-side damping controller, and SEDDC can effectively enhance the damping characteristics of CPSWPS subsystems. Compared to traditional optimization algorithms and sparrow search algorithm variants, the proposed RGSSA has a faster convergence speed and better search results. Optimization of SEDDC significantly improves the damping performance of the CPSWPS and enhances the regulation quality of the unit speed in pumped storage power stations and wind power stations. The SEDDC optimized by RGSSA is more effective in low-frequency oscillation suppression than conventional optimization algorithm, which has high robustness and universality under multi-condition disturbance. • Detailed mathematical equations of CPSWPS are constructed. • Novel shafting damping controller and SEDDC are designed and applied. • Improved sparrow search algorithm and optimization method of SEDDC are proposed. • Parameter sensitivity of SEDDC is analyzed. • Robustness and universality of the optimal SEDDC are verified. [ABSTRACT FROM AUTHOR]

For clean energy systems such as wind, solar and nuclear plants, the output power is usually regulated by controlling the motion rate of actuators, e.g. the stepping motors utilized for sun tracking of solar photovoltaic panels, yaw and pitch angle positioning of wind turbines and control rod driving of nuclear reactors. By constraining the actuators' motion rates to a finite set of values, the control system of a clean energy plant can be much simplified with obvious enhancement in operation reliability but requires developing finite-set control methods correspondingly. Motivated by the benefit of adopting finite motion rates, a finite-set control method is newly proposed for the nonlinear systems describing the dynamics of clean energy plants, compensating for the quantization and saturation effects induced by adopting a finite set of motion rates while ensuring globally bounded closed-loop stability strictly under a sufficient condition. The method is applied to design a finite-set power-level control of modular high temperature reactors, demonstrating stable power-level control during a reactor ramping-down from 100 % to 50 % reactor full power (RFP) with a constant rate of 5 % RFP/min. The simulation results also indicate that under the regulation of the finite-set control law, the steady error of hot helium temperature can eliminated, and the overshoot of neutron flux and that of hot helium temperature can be reduced by approximately 66 % and 75 % through properly adjusting control parameters, providing practical insights for engineering applications. • A novel finite-set control method is proposed for clean energy systems. • The method ensures globally bounded closed-loop stability. • Applied to modular high temperature reactors for stable power-level control. • Simulation shows reduced overshoot and faster response times for practical use. [ABSTRACT FROM AUTHOR]

This study proposes an integrated energy system using hydrogen storage to realize the efficient utilization of renewable energy sources and reduce the fluctuation when renewable energy is connected to grid. The system utilizes solar and wind energy to realize hydrogen production, desalination, and CCHP. First, the energy, exergy, and economic evaluations for the proposed system are carried out, and then an in-depth analysis of the key operating parameters is performed. The system achieves energy efficiency, exergy efficiency, and cost rate of 48.49 %, 19.98 %, and 7.969 $/h, respectively. And the exergy analysis shows that the main exergy destructions are caused by the parabolic trough solar collector and the transcritical CO 2 power cycle. The parametric analysis demonstrates when solar radiation flux and wind speed increase, the exergy efficiency and hydrogen production are increased, but the cost rate is increased accordingly. Finally, two sets of multi-objective optimization schemes are performed combining the artificial neural network with the Non-dominant genetic algorithm-II. For the optimized fresh water output, cost rate, and exergy efficiency, it is achieving improvements of 51.73 %, 8.4 %, and 3.6 %, and for the optimized hydrogen production, cost rate, and exergy efficiency, it is increased by 12.53 %, 0.564 %, and 0.75 %, respectively. [Display omitted] • Proposing an integrated energy system for hydrogen production, desalination, and combined cooling, heating, and power. • Synergizing solar and wind energy to produce power, heating, and fresh water. • Performing the multi-objective optimization scenarios using artificial neural network and Non-dominant genetic algorithm-II. • It shows the improvements of 12.53 % and 51.73 % for optimized hydrogen and freshwater production. [ABSTRACT FROM AUTHOR]

To tackle the issues of cooperative energy and reserve trading as well as fair cooperative benefit allocation among multiple integrated energy systems (IESs), this paper proposes a two-stage cooperative energy and reserve trading model for multiple integrated energy systems (MIESs). Specifically, at day-ahead stage, MIESs aim to maximize their overall profit through cooperative electricity and heat trading. At real-time stage, MIESs trade demand response (DR) reserve to minimize the overall wind power deviation compensation cost. To reduce the complexity in model solution, we transform the model into two sub-problems. In sub-problem 1, we determine the energy and DR reserve trading volumes. Here, distributionally robust optimization (DRO) is utilized to manage the severe uncertainties in wind power distribution. In sub-problem 2, based on the outcomes from sub-problem 1, we settle the energy and DR reserve trading prices. To ensure the fairness of benefit allocation, asymmetric Nash bargaining theory is applied to assess each IES's contributions in trading volumes and profit growth. Interval adaptive alternating direction method of multipliers (IA-ADMM) is used to preserve each IES's privacy. Finally, simulation results demonstrate that, compared with independent operation, cooperative trading among MIESs increases profits for all IESs, thereby motivating their participation in cooperative trading. • A two-stage cooperative energy and reserve trading model for MIESs is presented. • Asymmetric Nash bargaining ensures a fair allocation of cooperative benefit. • DRO is utilized to manage the severe uncertainties in wind power distribution. • IA-ADMM is employed to achieve the decentralized optimization of MIESs. [ABSTRACT FROM AUTHOR]

Given the variety of potential application scenarios, it is crucial to develop wind power curves that are more accurate, smoother, efficient, and applicable across a wider range of contexts. To this end, a wind power curve model combining smoothed spline with first-order moments (FOM) and density-adjusted wind speed (DAWS) strategy is proposed in this paper. First, the DAWS strategy is employed to consider the influence of meteorological factors on the wind power curve by adjusting wind speed. This strategy reduces the root mean square error (RMSE) by 5.81%–6.17 % and the mean absolute error (MAE) by 5.84%–6.44 % without adding complexity to the model. Secondly, FOM is proposed as a substitute for the original data during the modeling process. This approach reduces the number of operations on the similar data, resulting in a reduction of modeling time by 48.11%–99.89 %. Furthermore, the impact on model accuracy is minimal. Finally, a the wind power curve model based on smooth spline is constructed, which exhibits superior smoothness, a broader range of generalizability, enhanced model accuracy, and reductions in RMSE by 5.67%–23.87 % and MAE by 4.79%–22.35 % in comparison to the control method. • A Wind power curves model that combine smoothness, efficiency, and high accuracy is developed. • An air density-adjusted wind speed strategy is proposed. • An alternative to full dataset modeling with first-order moments is presented. • Model validated with data from multiple wind farms, demonstrating high reliability and practicality. [ABSTRACT FROM AUTHOR]

Wind energy (WE) is a volatile source of electricity production. In France and worldwide, the development of renewable energy sources (RES) is increasing the cost of the energy system; these will increase further to reach the WE target of 33.2 GW of installed capacity by 2028. Moreover, intermittency decreases as the installed capacity is geographically dispersed, hence the importance of investigating distributed wind deployment strategies in complementary locations. However, identifying potential wind energy locations that provide complementarity is challenging, especially given the inherent chaotic nature of wind during time. The objective of this research is to propose an adequate methodology to cluster wind time series (TS) to provide insights on smart planning considering distributed wind energy (WE) production. Results reveal that Shape Based Distance (SBD) classifiers perform best in clustering TS and appear relevant in identifying potential wind complementary locations in France. Ultimately, intermittency measures show that, in the case of complementary wind locations, availability (AVA) can increase by about 31% and the variability can decrease by about 30%. • Clustering wind time series, using a reproducible method, led to the identification of wind complementary locations. • Wind speeds are recorded by weather stations across northwest and southeast of France. • Benchmark of three instance-based clustering techniques using cluster validity index. • Clusters with negative and near-zero correlation exhibit promising potential for complementarity. • Complementary wind parks identification serves as an important step for planning future distributed wind power generation. [ABSTRACT FROM AUTHOR]

Offshore floating wind turbines will play a pivotal role in achieving net-zero emissions. This study explores a hybrid platform combining an offshore floating wind turbine with wave energy converters (WECs) from an active control perspective to mitigate turbine damage risk while maximizing wave energy conversion. Initially, the control-oriented state-space modelling of the hybrid platform and the validation of the time-domain Cummins-type models of different orders are performed by the Eulerian-Lagrangian method with model linearization and downscaling. Then, a multi-objective non-causal optimal controller is introduced, balancing wave energy capture and penalizing the nacelle acceleration. When wave energy capture maximization is set as the control target, the proposed optimal controller can improve energy output by 184% as compared against a well-tuned passive damper across all peak periods tested. However this causes peak hub acceleration to increase marginally beyond the desirable limits of 3–4 m/s 2 for wind turbine operation. When hub acceleration is penalized, the multi-objective optimal controller can reduce peak values below limits by between 40% and 61% with trivial loss of wave energy capture. • Validated a state-space model for a hybrid wind-wave energy platform. • Reduced model dimensionality for improved control efficiency and validity. • A multi-objective control balancing wave energy and wind platform stability. • Improved insights on active control for safe, efficient hybrid platform operation. [ABSTRACT FROM AUTHOR]

This paper proposes constructing a multi-energy complementary power generation system integrating hydropower, wind, and solar energy. Considering capacity configuration and optimization of the complementary power generation system, a dual-layer planning model is constructed. The outer layer aims to maximize the accessible scale of wind and solar energy, while the inner layer considers the matching degree between power output and grid load. The optimization uses a particle swarm algorithm to obtain wind and solar energy integration's optimal ratio and capacity configuration. The results indicate that a wind-solar ratio of around 1.25:1, with wind power installed capacity of 2350 MW and photovoltaic installed capacity of 1898 MW, results in maximum wind and solar installed capacity. Furthermore, installed capacity increases with increasing wind and solar curtailment rates and loss-of-load probabilities. When the loss-of-load probability is set to 3 %, the impact of wind and solar curtailment rates on the installed capacity ratio is relatively small. The complementary characteristics of wind and solar energy can be fully utilized, which better aligns with fluctuations in user loads, promoting the integration of wind and solar resources and ensuring the safe and stable operation of the system. • Proposed model optimizes wind-solar-hydropower capacity configuration for stability. • Wind-solar ratio of 1.25:1 minimizes energy curtailment and maximizes grid integration. • The model enhances system reliability by utilizing hydropower's peak-shaving capacity. • Simulation results validated using real-world data from the southwest region of China. • Future research will focus on stochastic modeling and incorporating energy storage systems. [ABSTRACT FROM AUTHOR]

The power prediction accuracy of wind farm cluster (WFC) seriously affects its consumption and the safe and stable operation of power system. The fluctuation of power between wind farms (WFs) significantly affects the wind power ultra-short-term prediction (WPUP) accuracy of WFC. In this regard, this paper proposes a graph deviation attention network (GDAN) considering improved clustering distance and learnable graph structure (LGS) for predicting and correcting the wind power of WFC. And used a weighted distance function combining sequence convergence smoothing effect and correlation to dynamically divide the WFC, and to learn and construct the graph structure. Proposed the GDAN with LGS to mine the convergence correlation of WF sub-clusters and establish power prediction model. Considering the characteristics of load peak and valley periods (LPVP), introduced a power correction coefficient to reduce the error, and used the successive variational mode decomposition (SVMD) to extract its key components to achieve power prediction and correction. The proposed method is applied to the WFC in Western Inner Mongolia, China. Compared with the comparison model before correction, the RMSE, MAE and MAPE are reduced by 4.27 %, 3.55 % and 17.92 % respectively, and the R2 and Pr are increased by 11.87 % and 9.88 % respectively. [Display omitted] • Proposed a WFC power prediction strategy considering LPVP characteristics. • Used a clustering distance with correlation and convergence effect for WFC division. • Proposed a GDAN with graph structure self-learning mechanism for power prediction. • Proposed the correction coefficient to describe the dynamic evolution of error. • Used SBOA-SVMD model to increase the predictability of correction coefficient. [ABSTRACT FROM AUTHOR]

Energy transition on small islands is limited by the scarce availability of land, restricting large-scale implementation of onshore renewable energy technologies such as solar photovoltaics and wind power. Ocean energy technologies provide novel opportunities for land-constrained islands to achieve 100% renewable energy systems. While wave power is increasingly implemented in energy system modelling research, ocean thermal energy converters are not yet a standard technology in renewable energy technology portfolios. This research aims to study the impacts of ocean thermal energy converters on the energy system of the Maldives through a structured sensitivity analysis for the two scenario clusters covering e-fuel import and domestic production. The ocean thermal energy conversion plants are modelled using spatially and temporally resolved resource data and cost assumptions from a global upscaling scenario, considering the technology's current development stage. Results show that ocean thermal energy converters play a limited role in 'purely' cost-optimised sub-scenarios due to the availability of very low-cost offshore floating photovoltaics, making it difficult for them to compete. Nevertheless, reduced requirement of energy storage technologies due to the stable electricity production of ocean thermal energy converters offers an option to diversify the renewable energy technology portfolio with only a minor increase in cost. • Role of ocean thermal energy converter on the case of Maldives studied. • Structured sensitivity analysis applied to assess the impact on the energy system. • Import and domestic production of e-fuels in 2030 and 2050 modelled. • Ocean thermal energy converter are hardly cost-competitive with solar photovoltaics. • Technology portfolio diversification options are discussed based on results. [ABSTRACT FROM AUTHOR]

To unlock the full potential of offshore wind requires investment in a more integrated offshore grid infrastructure combining generation and interconnection between countries in so-called hybrid projects. Taking the perspective of a social planner, these projects may result in more efficient allocation of scarce resources. Alas, the distribution of this welfare gain may result in disagreement on what grid design is the most attractive, what markets should be connected and whether and when to invest. To examine these dilemmas, we apply real options theory to investments projects with different grid design and market characteristics. We compare the incentives to invest in a) only the wind farm and b) both the wind farm and the grid infrastructure, assuming the project value of the last alternative can serve as a proxy for social welfare. Solutions are derived using a simulation approach called least squares Monte Carlo method. We find: 1) Non-stationary stochastic prices and/or decreasing costs may make it socially optimal as well as commercially beneficial to postpone even profitable investments. 2) Connecting markets is socially desirable but may reduce the offshore wind investor's project value. 3) Connecting markets with different characteristics is socially desirable but reduce the offshore wind investors' project value. • Considering uncertainty, profitable offshore wind investments might be postponed. • Differing private and public incentives may cause suboptimal investment. • Connecting markets is socially desirable but reduces value for a private investor. • Redistributing gains from offshore wind projects can align stakeholders' incentives. [ABSTRACT FROM AUTHOR]

With the large-scale development of wind power, high penetration wind power grid connection poses serious challenges to the safe and stable operation of the power system. However, the current accuracy of wind power forecasting is facing bottlenecks due to the limitations of Numerical Weather Prediction (NWP) data. Therefore, this article proposes a two-stage day-ahead multi-step wind power prediction (WPP) scheme that considers temporal information interaction. In the first stage, the next day prediction of wind power is based on historical power and 0–24 h NWP data. Then, an embedded deep decomposition module is used to extract predictable components and multi-scale information fusion is performed. In the second stage, the result of day-ahead WPP is obtained based on the extracted predictable components and combined with 24∼48 h of NWP data. The wind farms in Jilin and Inner Mongolia of China are used to experimental analysis. The results show that the scheme proposed in the article has a better prediction effect compared with other schemes in the paper, which can effectively improve the multi-step prediction accuracy of day-ahead wind power. • Propose a day-ahead power prediction that considering time information interaction. • Propose a multi-scale fusion TRFE feature enhancement module. • It provides new ideas for day-ahead wind power prediction for wind farm. [ABSTRACT FROM AUTHOR]

In this paper, considering the complex relationship between wind speed variation characteristics and data features, a fully connected multi-reservoir echo state network (FCMR-ESN) is proposed for wind power generation prediction, which can handle some issues such as insufficient extraction of data features and a gradual decline in memory capacity. Firstly, the Spearman correlation coefficient is used to calculate the correlation between the characteristic data. Secondly, the fully connected neurons are applied to the connections between the reservoirs. The fully connected layer can capture the complex nonlinear relationships in the wind power data and effectively map them to the prediction results. Thirdly, the reservoir parameters and the connection coefficients between the reservoirs are optimized using the improved weighted mean of vectors (INFO). Finally, The effectiveness of FCMR-ESN prediction is demonstrated through prediction experiments conducted on two sets of datasets with different lengths from different regions. In the one-day time prediction for the first set of data, the MAPE decreases to 2.90%. The MAPE for the three-day prediction in the second data set is reduced to 3.53%. • Considering the complex of wind data, a prediction method based FCMR-ESN is proposed. • Spearman correlation coefficient is used to calculate the correlation of sample data. • Fully connected layer can capture the complex nonlinear relationships of wind data. • Reservoir parameters and connection coefficients are optimized using the INFO. [ABSTRACT FROM AUTHOR]

The randomness and fluctuations in wind power generation present significant challenges for grid and wind farm dispatching. Accurate very short-term wind power forecasting (WPF) is therefore essential for the efficient operation of modern power systems. Data-driven models, such as Transformers, have demonstrated their effectiveness in WPF due to their ability to efficiently capture global features in long sequences. However, limited research has examined the impact of incorporating static data into WPF, which may limit forecasting accuracy. This paper proposes a Temporal Fusion Transformer forecasting model to address this challenge. This approach employs static data as the input features for the model. The model includes feature selection through a variable selection network and employs a specialized temporal fusion decoder to learn effectively from these static features. The case results show that the results of the proposed model are more accurate than the state-of-the-art methods, reducing MAPE by at least 1.32%, RMSE by 0.0091, and improving R 2 by 0.035 in case studies. Additionally, the model maintains a manageable computational burden, underscoring its practical applicability. • End-to-end framework with static features improves very short-term wind power forecasting accuracy. • Variable selection network mitigates high-dimensional features, enhancing relevant inputs. • Multi-head self-attention captures diverse temporal dependencies for forecast accuracy. • Case studies show the model's impact on forecasting accuracy across various input lengths. [ABSTRACT FROM AUTHOR]

Accurate wind power prediction is highly significant to the safety, stability, and economic operation of power grids. Currently, typical ensemble methods for wind power forecasting are widely designed based on the mean square error (MSE) loss, which are very suitable for the assumption that the error distribution obeys the Gaussian distribution. However, the complex nonlinear conversion of wind energy into wind power may change the statistical characteristics of errors, thus the prediction model based on the traditional MSE loss may lead to the performance degradation of the forecasting model. To address these problems, a probability density function based adaptive ensemble learning with global convergence is proposed for wind power prediction, which comprises three modules: a data preprocessing module, a prediction module, and a combination module. Firstly, an effective feature generation mechanism is employed to extract the multi-mode characteristics of wind data. Then, an auxiliary error based adaptive global convergence model is developed as benchmark predictor in the prediction module, where an adaptive updating algorithm is derived based on the Lyapunov approach to ensure the global convergence of the model weights. Moreover, considering the asymmetric characteristic of modeling error, a probability density function (PDF) based ensemble learning is created to integrate the results of benchmark predictors. Specifically, the ensemble model parameter updating is transformed into the shape control for the modeling error PDF, which can break through the limitation of MSE loss capturing only the second moment information, and emphasize the spatial distribution of errors to make an unbiased estimate of wind power. Experimental results show that the proposed ensemble model has significant advantages over other models involved in this study. • A diversity feature generation mechanism is introduced for extraction of multi-modes. • An auxiliary error based adaptive model is developed as benchmark predictor. • A probability density function based adaptive ensemble learning is created. • A rigorous convergence analysis of the learning algorithm is implemented. [ABSTRACT FROM AUTHOR]

The potential of improvement of both overall energy efficiency and penetration of renewable energy for the combined heat and power (CHP) unit was investigated by integrating the source-side and grid-side thermal energy storage (TES) systems simultaneously. The mathematical model of the proposed thermal system was established, with which the flexibility-enhancing features across diverse operating conditions were analyzed. The flexibility improvement rate, heat consumption rate, TES cycle efficiency and energy efficiency were revealed. Moreover, the wind power consumption, coal-savings and net annual revenue of CHP unit integrated with different TES were presented. The results indicated that the flexibility improvement rate of source-side TES, grid-side TES and dual TES is 2.4 %, 21.2 % and 26.2 %, respectively. The heat consumption rate of a CHP unit integrated with source-side TES system is lower compared to that of a traditional CHP unit when power load ratio is below 50 %. The CHP unit integrated with a dual TES system exhibited a maximum increase in wind power accommodation rate of 37.7 % and a maximum reduction in standard coal consumption of 7.7 %. The proposed systems offer a promising approach for enhancing the flexibility of CHP units to accommodate more renewable energy. • Renovation with dual TES systems proposals aimed at enhancing flexibility of CHP unit. • Cost-benefit assessment of the novel systems introduced. • Maximum increment of 26.2 % for feasible operation region acquired. • Energy-saving potential revealed maximum 7.7 % drop for standard coal consumption. [ABSTRACT FROM AUTHOR]

Reliable multistep ahead wind speed forecasting (MAWSF) is critical for the energy management of wind farms and the long-term maintenance of wind power systems. However, relying on inherent meteorological features such as temperature and atmospheric pressure often fails to meet the deep learning model's feature requirements for accurate wind speed forecasting (WSF). This paper introduces a hybrid multistep forecasting model that constructs a univariate wind speed feature enhancement framework, combining random forest (RF) and Transformer models for WSF. Initially, the hybrid enhancement framework decomposes the univariate wind speed data and extracts time-series features, effectively mining the latent feature information. Subsequently, the RF feature selector filters out significant features contributing to WSF and eliminates redundant features to provide stable features. Finally, the Transformer model is utilized for both short-term and long-term MAWSF. This study conducted MAWSF on data with sampling intervals of 20 min, 30 min and 1 h. The results indicate that, compared to existing state-of-the-art models, the hybrid model in MAWSF tasks reduces the dependency of models on inherent meteorological features, achieving more accurate forecasting and faster computation speeds. Ultimately, the proposed model can provide reliable technical support for energy management and maintenance guidance in real-world wind farms. • A new hybrid model is proposed for multi-step ahead wind speed forecasting. • A new feature enhancement method for univariate wind speed data is proposed and validated. • The proposed hybrid model shows superior performance compared to other state-of-the-art models. • The proposed hybrid model facilitates multi-step ahead forecasting of wind speed in wind farms. [ABSTRACT FROM AUTHOR]

The inherent volatility and intermittency of wind power present significant forecasting challenges, undermining the efficient integration of wind energy into the power grid. Existing methodologies, notably long short-term memory (LSTM) networks, encounter significant limitations due to their inefficiencies in processing long sequences, difficulties in capturing multi-scale temporal dynamics, and heightened sensitivity to noisy data, which can severely hamper model performance. To address these challenges, This paper proposes the frequency filter enhanced dual LSTM network (FDNet), a novel approach that directly addresses the constraints of the LSTM and improves the accuracy and stability of wind power forecasting. Specifically, FDNet employs the patching operation to divide the original time series into several sub-sequences, potentially boosting the computational efficiency. Furthermore, a specific frequency filter is designed and incorporated into FDNet, effectively reducing the influence of noise. Finally, a dual LSTM structure is employed, which enables FDNet to adeptly discover both short-term local temporal patterns and long-term global temporal patterns inherent in wind power data. Extensive experiments across three datasets demonstrate that FDNet significantly outperforms existing methods, achieving up to 11.0% reduction in mean absolute error and 8.1% in root mean squared error on the HL dataset, underscoring its effectiveness in wind power forecasting. • A frequency filter enhanced dual LSTM network is proposed for wind power forecasting. • Patching operation is introduced to improve the efficiency of the model. • The proposed model addresses the shortcomings of the traditional LSTM. • The proposed model achieves the best performance on three wind power datasets. [ABSTRACT FROM AUTHOR]

The depletion of fossil fuels and environmental pollution have increasingly led to the recognition of wind power as a significant sustainable energy source. However, the intermittent and unstable nature of wind energy underscores the critical importance of accurate wind power forecasting for maintaining the stability of power systems. This paper aims to achieve precise forecasting of ultra-short-term wind power generation by proposing an innovative and practical method utilizing a novel self-sustaining cascading rolling mechanism. Initially, employing a rigorous data partitioning approach to ensure the independence of the training and testing datasets, and determining a rolling decomposition window of 192 time steps through preliminary experiments. Subsequently, the decomposition window was gradually shifted backward along the temporal axis, applying the ICEEMDAN algorithm independently within each window to eliminate any possibility of information leakage. Finally, a TCN-BiLSTM-Attention forecasting model was constructed, which accepts the multiple components obtained from the rolling decomposition as input, allowing for accurate predictions of wind power fluctuations over various forecasting horizons ranging from 15 min to 1 h. The effectiveness of the hybrid algorithm was validated through comprehensive experiments. Thanks to the resolution of the information leakage issue, this hybrid method can be implemented in a simulated online context. • Pre-experimented strategy ensures 192-step rolling windows for data in- tegrity. • A self-sustaining cascading forward decomposition prevents information leakage. • The Multi-input, single-output model combines single and multi-step fore- casting. • TCN-BiLSTM-Attention forecasts wind power shifts 15 min to 1 h ahead. [ABSTRACT FROM AUTHOR]