Article
Thermodynamics
Guo Nai-Zhi, Zhang Ming-Ming, Li Bo
Summary: A data-driven analytical wake model is proposed in this paper, which extracts local inflow information and wake expansion features from measured data of wind farms, trains a machine learning model to establish the relationship between the two, and improves the wake prediction performance by over 20% compared to traditional analytical models.
ENERGY CONVERSION AND MANAGEMENT
(2022)
Article
Computer Science, Artificial Intelligence
S. Ashwin Renganathan, Romit Maulik, Stefano Letizia, Giacomo Valerio Iungo
Summary: This paper introduces a machine learning-based approach to construct predictive models of wind turbine wake flows using real-world measurement data. The mapping between the parameter space and wake flow fields is learned using deep autoencoders and deep neural networks. Probability machine learning technique and variational Gaussian process models are employed to address data uncertainty and large datasets. Active learning is also introduced to improve the predictive capability of the model.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Celal Cakiroglu, Sercan Demir, Mehmet Hakan Ozdemir, Batin Latif Aylak, Gencay Sariisik, Laith Abualigah
Summary: This study estimates the power produced in a wind turbine using six different regression algorithms based on machine learning. The XGBoost algorithm performs the best according to the R2 performance metric, while the LightGBM model is the most efficient in terms of computational speed. Wind speed is shown to have the most significant impact on the model predictions according to the SHAP algorithm.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Green & Sustainable Science & Technology
Ravi Pandit, David Infield, Matilde Santos
Summary: Continuous assessment of wind turbine performance is crucial for maximizing power generation at a low cost. This study aims to quantify and analyze the impact of wind shear and turbulence intensity on wind turbine power curves. The results show that taking these factors into consideration can improve the accuracy and reduce the uncertainty of power curve models.
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
(2023)
Article
Energy & Fuels
Ravi Pandit, Davide Astolfi, Anh Minh Tang, David Infield
Summary: Offshore wind turbines have gained popularity worldwide in recent years due to their ability to generate a large amount of electrical power. However, they are vulnerable to damage from high-speed winds, making accurate weather forecasting crucial for optimizing their operation and maintenance. This paper proposes three sequential data-driven techniques for long-term weather forecasting and compares their strengths and weaknesses. The study results suggest that the proposed technique can generate realistic and reliable weather forecasts and respond well to seasonality.
Article
Computer Science, Information Systems
Wisdom Udo, Yar Muhammad
Summary: The analysis predicts a significant increase in offshore wind capacity by 2026, emphasizing the need for complex maintenance of wind turbines. Offshore wind operation and maintenance costs make up a substantial portion of the total electricity cost, highlighting the importance of exploring predictive maintenance methods and optimizing production output. The proposed approach of monitoring critical components using historical SCADA data and advanced models shows promise in improving operational reliability and reducing maintenance costs of wind turbines.
Article
Energy & Fuels
Martin Geibel, Galih Bangga
Summary: Data driven approaches are used for optimal sensor placement and velocity prediction in wind turbine wakes. Various methods are investigated for clustering analysis and predicting the flow field's time history. The studies show that a combination of classification-based machine learning algorithm and Bi-LSTM can predict periodic signals accurately, while a more advanced technique is needed for complex turbine near wake data.
Article
Thermodynamics
Yun Wang, Xiaocong Duan, Runmin Zou, Fan Zhang, Yifen Li, Qinghua Hu
Summary: Existing wind turbine power curve (WTPC) models have limited performance in capturing the complex relationship between wind speed and wind power due to their inadequate nonlinear fitting abilities. Deep learning (DL) excels at describing complex relationships. This study proposes a novel data-driven DL approach mELM-CA-CNN to establish WTPCs based on multiple extreme learning machines (ELMs), channel attention (CA), convolutional neural network (CNN), and Huber loss (HL). Comparisons with some popular WTPC models demonstrate that mELM-CA-CNN obtains the most accurate WTPCs on four wind datasets, showing the superiority of the proposed DL approach. Moreover, the roles of the different modules of mELM-CA-CNN in improving model performance are verified.
Article
Green & Sustainable Science & Technology
Yanting Li, Wenbo Jiang, Guangyao Zhang, Lianjie Shu
Summary: This paper proposes a fault diagnosis method for wind turbines based on parameter-based transfer learning and convolutional autoencoder, suitable for small-scale data. The method can transfer knowledge from similar wind turbines and shows advantages in fault diagnosis.
Article
Automation & Control Systems
Hongyang Dong, Xiaowei Zhao
Summary: This article proposes a novel data-driven control scheme to maximize the total power output of wind farms subject to strong aerodynamic interactions among wind turbines. The proposed method is model-free and has strong robustness, adaptability, and applicability. The effectiveness, robustness, and scalability of the proposed method are tested by prototypical case studies with a dynamic wind farm simulator.
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY
(2023)
Article
Computer Science, Interdisciplinary Applications
Sven F. Falkenberg, Stefan Spinler
Summary: A predictive maintenance model for material handling equipment is built using novel data sources, which can forecast breakdowns. The study explores statistical learning methods for failure detection and shows that the standard sensors in the equipment provide sufficient data for predicting the majority of breakdowns. The research also evaluates the cost-effectiveness of different statistical learning methods and finds that K-Nearest-Neighbors and Random Forest Classifier are the most optimal choices. Additionally, the study emphasizes the importance of considering both time and condition in maintenance and presents a prediction model that incorporates both variable types. Recommendations on data collection and understanding the cost ratio between breakdowns and preventive maintenance services are provided from a managerial perspective.
COMPUTERS & INDUSTRIAL ENGINEERING
(2022)
Article
Engineering, Electrical & Electronic
Davide Astolfi, Francesco Castellani, Andrea Lombardi, Ludovico Terzi
Summary: This study formulates a method for estimating wind turbine performance decline with age based on long term SCADA data analysis, finding that in the considered test cases, the average rate of performance decline with age is approximately -0.2% per year, compatible with recent analyses based on cumulative data. It is also concluded that gearbox aging does not contribute to the performance decline, while generator aging does.
ELECTRIC POWER SYSTEMS RESEARCH
(2021)
Article
Engineering, Chemical
Christopher A. K. Gordon, Efstratios N. Pistikopoulos
Summary: Prescriptive maintenance improves system effectiveness and safety, but system disruptions may lead to abnormal operations and safety incidents. This research proposes a multiparametric framework for safety-aware process control, utilizing ensemble classifiers for fault detection, mixed-integer nonlinear programming for scheduling, and multiparametric model predictive control for fault-tolerant tracking. Results show superior performance in fault detection and ability to reconfigure control actions based on disruptions. The approach has been demonstrated on chemical and cooling water systems to enhance industrial process safety and productivity.
Article
Green & Sustainable Science & Technology
Jie Yan, Akejiang Nuertayi, Yamin Yan, Shan Liu, Yongqian Liu
Summary: This paper proposes a hybrid physical and data-driven model for simulating the dynamic operating characteristics of wind turbines. By building a dynamic characteristics database and using sequence learning and LSTM network, the time sequence dynamic characteristics of wind turbine operating conditions and parameters under various wind conditions are accurately excavated and learned. Through numerical example analysis, the method improves the simulation accuracy and calculation efficiency of wind turbine output power and load dynamic characteristics.
Article
Engineering, Multidisciplinary
M. El-Naggar, A. Sayed, M. Elshahed, M. EL-Shimy
Summary: This paper proposes a detailed approach for maintenance decisions based on failure analysis to select the optimum maintenance strategy for different subassemblies of wind turbines, aiming to improve their performance and efficiency.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Energy & Fuels
Samet Ozturk, Vasilis Fthenakis, Stefan Faulstich
Review
Physics, Applied
K. Rohrig, V. Berkhout, D. Callies, M. Durstewitz, S. Faulstich, B. Hahn, M. Jung, L. Pauscher, A. Seibel, M. Shan, M. Siefert, J. Steffen, M. Collmann, S. Czichon, M. Doerenkaemper, J. Gottschall, B. Lange, A. Ruhle, F. Sayer, B. Stoevesandt, J. Wenske
APPLIED PHYSICS REVIEWS
(2019)
Article
Energy & Fuels
Marc-Alexander Lutz, Philip Goerg, Stefan Faulstich, Robert Cernusko, Sebastian Pfaffel
Article
Chemistry, Multidisciplinary
Sebastian Pfaffel, Stefan Faulstich, Kurt Rohrig
APPLIED SCIENCES-BASEL
(2020)
Article
Economics
Katherina Grashof, Volker Berkhout, Robert Cernusko, Maximilian Pfennig
Article
Energy & Fuels
Philipp Beiter, Aubryn Cooperman, Eric Lantz, Tyler Stehly, Matt Shields, Ryan Wiser, Thomas Telsnig, Lena Kitzing, Volker Berkhout, Yuka Kikuchi
Summary: The cost of wind power has decreased significantly over the past 20 years, making it more competitive with conventional sources. Forecasts suggest further cost reductions for onshore and offshore wind, with midrange estimates predicting levelized cost of energy to be between $20-30/MWh for onshore and $40-60/MWh for offshore by 2050, approximately half of today's levels. Optimistic forecasts even anticipate reaching these levels as early as 2030.
WILEY INTERDISCIPLINARY REVIEWS-ENERGY AND ENVIRONMENT
(2021)
Article
Chemistry, Physical
Philipp Beiter, Lena Kitzing, Paul Spitsen, Miriam Noonan, Volker Berkhout, Yuka Kikuchi
Summary: Comparing prices from competitive procurement of renewable energy projects may lead to misleading conclusions, as differences in project revenue and value must be considered. By using a cash flow model to evaluate factors such as support regimes, market sales, and tax incentives, decision makers and researchers can make more comprehensive comparisons of procurement costs.
Proceedings Paper
Energy & Fuels
S. Pfaffel, S. Faulstich, S. Sheng
16TH DEEP SEA OFFSHORE WIND R&D CONFERENCE
(2019)
Article
Energy & Fuels
Samet Ozturk, Vasilis Fthenakis, Stefan Faulstich
Proceedings Paper
Energy & Fuels
Berthold Hahn, Thomas Welte, Stefan Faulstich, Pramod Bangalore, Cyril Boussion, Keith Harrison, Emilio Miguelanez-Martin, Frank O'Connor, Lasse Pettersson, Conaill Soraghan, Clym Stock-Williams, John Dalsgaard Sorensen, Gerard van Bussel, Jorn Vatn
14TH DEEP SEA OFFSHORE WIND R&D CONFERENCE, EERA DEEPWIND'2017
(2017)