Article
Green & Sustainable Science & Technology
Yubo Huang, Shuyue Lin, Xiaowei Zhao
Summary: This paper proposes a control system for a wind farm with a new type of wind turbines using multi-agent reinforcement learning. The multi-agent policy optimization algorithm allows the turbines to gradually improve their control policies, leading to increased power generation in the wind farm.
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
(2023)
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
Automation & Control Systems
Hongyang Dong, Xiaowei Zhao
Summary: This article presents a novel preview-based robust deep reinforcement learning method for wind-farm power tracking problem, which can handle tasks subject to uncertain environmental conditions and strong aerodynamic interactions among wind turbines. The control problem is transformed into a zero-sum game to quantify the influence of unknown wind conditions and future reference signals.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Energy & Fuels
Hongyang Dong, Jincheng Zhang, Xiaowei Zhao
Summary: This study proposes a wind farm control scheme based on deep reinforcement learning, which uses a reward regularization module and composite learning controller to optimize power generation and yaw tracking. The scheme demonstrates robustness and adaptability in handling uncertain wind conditions.
Article
Mechanics
B. F. Zhang, D. W. Fan, Y. Zhou
Summary: This work proposes the use of machine learning or artificial intelligence to control the low-drag Ahmed body and find efficient drag reduction strategies. By blowing air from microjets along the edges of the rear window and vertical base, and using pressure sensors and a controller based on the ant colony algorithm, the near-optimal control law is learned and implemented. The highest drag reduction of 18% is achieved by altering the flow structure.
JOURNAL OF FLUID MECHANICS
(2023)
Article
Mechanics
Jichao Li, Mengqi Zhang
Summary: This research utilizes Reinforcement Learning for flow control by setting up synthetic jets on both sides of a cylinder to suppress vortex shedding. Through global linear stability and sensitivity analyses, and incorporating physical results in the design of control policies, the study finds that the controlled wake converges to the unstable base flow, requiring persistent oscillating control for maintaining this state. Additionally, embedding flow stability information in the reward function can lead to a more stable controlled flow.
JOURNAL OF FLUID MECHANICS
(2021)
Article
Mechanics
Guy Y. Cornejo Maceda, Yiqing Li, Francois Lusseyran, Marek Morzynski, Bernd R. Noack
Summary: We stabilize flow past a cluster of three rotating cylinders, known as the fluidic pinball, using automated gradient-enriched machine learning algorithms to optimize control laws and achieve better performance through increasingly richer search spaces.
JOURNAL OF FLUID MECHANICS
(2021)
Article
Green & Sustainable Science & Technology
Arslan Salim Dar, Guillem Armengol Barcos, Fernando Porte-Agel
Summary: Wind-tunnel experiments were conducted to investigate the impact of minor modifications to the roof edge shape on the power performance and wake characteristics of a horizontal-axis wind turbine on a cube-shaped building. The study found that the power performance of the turbine and the wake characteristics are influenced by the roof edge shape.
Article
Green & Sustainable Science & Technology
Yubao Zhang, Xin Chen, Sumei Gong, Jiehao Chen
Summary: "This study proposes a novel communication-based multi-agent deep reinforcement learning approach for controlling power generation in large-scale wind farms. By introducing a multivariate power model to analyze the impact of wake effects, and using a hierarchical communication multi-agent proximal policy optimization algorithm to coordinate continuous controls, the proposed approach significantly increases wind farm power output. Importantly, there is no significant increase in wind turbine blade fatigue damage as the wind farm scale increases."
Article
Green & Sustainable Science & Technology
Yu Ding, Nitesh Kumar, Abhinav Prakash, Adaiyibo E. Kio, Xin Liu, Lei Liu, Qingchang Li
Summary: This study conducted a space-time performance comparison of a wind farm with 66 turbines over nearly four years, taking into account wind and environmental inputs. The analysis showed quantitative and global differences among turbines and how they changed over time.
Article
Chemistry, Multidisciplinary
Wen-Hui Lin, Ping Wang, Kuo-Ming Chao, Hsiao-Chung Lin, Zong-Yu Yang, Yu-Huang Lai
Summary: This study focuses on determining the proper hyperparameters for deep learning network models using a Q-learning scheme for wind-power forecasting. Experimental results show that the developed temporal convolution network model achieves higher accuracy in 168-hour wind power prediction and outperforms classical recurrent network models.
APPLIED SCIENCES-BASEL
(2022)
Article
Green & Sustainable Science & Technology
Jian Wei Lin, Wei Jun Zhu, Wen Zhong Shen
Summary: In this paper, two new analytical wake models are developed to predict the wind velocity distribution in the wake region of a wind turbine. These models are validated and compared with existing engineering wake models, showing lower relative errors and better performance. They are recommended for wind farm design.
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
Environmental Sciences
Jinyuan Xin, Daen Bao, Yining Ma, Yongjing Ma, Chongshui Gong, Shuai Qiao, Yunyan Jiang, Xinbing Ren, Tao Pang, Pengcheng Yan
Summary: In this study, various methodologies were adopted to explore the predictability and optimization of wind speed in a Gobi grassland wind farm. The results showed that factors such as the influence of upwind turbine wakes and the deviation between observations and simulations greatly affected the accuracy of wind speed forecasting. Error reduction was achieved through postprocessing methods and machine learning algorithms. Furthermore, the application of data assimilation, parameterization scheme optimization, and high-resolution topographic data had the potential to improve the accuracy of wind speed prediction.
Article
Thermodynamics
M. E. Nakhchi, S. Win Naung, M. Rahmati
Summary: The main objective of this study is to use the Extreme Gradient Boosting (XGBoost) machine learning algorithm to predict the power, wake, and turbulent characteristics of horizontal-axis wind farms under yaw-controlled conditions. It is observed that XGBoost is more accurate for wake prediction compared to ANN, and it requires a much shorter training time.
ENERGY CONVERSION AND MANAGEMENT
(2023)
Article
Thermodynamics
Navid Zehtabiyan-Rezaie, Majid Saffar-Avval, Mostafa Mirzaei
INTERNATIONAL JOURNAL OF THERMAL SCIENCES
(2015)
Article
Thermodynamics
Navid Zehtabiyan-Rezaie, Mostafa Mirzaei, Majid Saffar-Avval
HEAT TRANSFER ENGINEERING
(2017)
Article
Thermodynamics
Navid Zehtabiyan-Rezaie, Amir Arefian, Mohammad J. Kermani, Amir Karimi Noughabi, M. Abdollahzadeh
ENERGY CONVERSION AND MANAGEMENT
(2017)
Article
Engineering, Electrical & Electronic
Navid Zehtabiyan-Rezaie, Majid Saffar-Avval, Kazimierz Adamiak
JOURNAL OF ELECTROSTATICS
(2018)
Article
Green & Sustainable Science & Technology
N. Zehtabiyan-Rezaie, N. Alvandifar, F. Saffaraval, M. Makkiabadi, N. Rahmati, M. Saffar-Avval
SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS
(2019)
Article
Engineering, Electrical & Electronic
Navid Zehtabiyan-Rezaie, Majid Saffar-Avval, Kazimierz Adamiak
JOURNAL OF ELECTROSTATICS
(2020)
Article
Engineering, Chemical
Navid Zehtabiyan-Rezaie, Majid Saffar-Avval, Kazimierz Adamiak
Summary: This study numerically examines the performance and energy consumption of an electrohydrodynamic desalination system for nightly production of fresh water in high-humidity regions. The results show that this system has the potential to be an efficient supplement to solar stills, with high efficiency and low energy consumption.
Article
Thermodynamics
Navid Zehtabiyan-Rezaie, Majid Saffar-Avval, Kazimierz Adamiak
INTERNATIONAL COMMUNICATIONS IN HEAT AND MASS TRANSFER
(2020)
Review
Engineering, Electrical & Electronic
Navid Zehtabiyan-Rezaie, Kazimierz Adamiak, Majid Saffar-Avval
Summary: Impressive progress has been made in enhancing evaporation from liquid surfaces due to electrohydrodynamic flow, particularly through numerical simulations, over the past two decades. Sharp electrodes placed over the liquid surface have been observed to increase evaporation rate. Different water models have been proposed in numerical simulations, with conducting models being the physically correct ones.
JOURNAL OF ELECTROSTATICS
(2021)
Review
Green & Sustainable Science & Technology
Navid Zehtabiyan-Rezaie, Alexandros Iosifidis, Mahdi Abkar
Summary: With the increasing number of wind farms, research in wind-farm flow modeling is shifting towards data-driven techniques. However, the complexity of fluid flows in real wind farms poses unique challenges for data-driven modeling, requiring models to be interpretable and have some degree of generalizability.
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY
(2022)
Article
Mechanics
Ali Eidi, Navid Zehtabiyan-Rezaie, Reza Ghiassi, Xiang Yang, Mahdi Abkar
Summary: This study quantifies the model-form uncertainties in RANS simulations using a data-driven machine-learning technique. By applying a two-step feature-selection method and the extreme gradient boosting algorithm, more accurate representations of the Reynolds stress anisotropy are obtained. The proposed framework provides optimal estimation of uncertainty bounds for the RANS-predicted quantities of interest.
Article
Engineering, Civil
Navid Zehtabiyan-Rezaie, Mahdi Abkar
Summary: Analytical wake models require formulas to replicate the effect of wind turbines on turbulence levels in the wake region. One such formula proposed by A. Crespo, J. Hernandez in 1996 relates added turbulence to turbine induction factor, ambient turbulence intensity, and normalized distance from the rotor using one coefficient and three exponents. However, an incorrect exponent for ambient turbulence intensity has been mistakenly used in the literature. This study implemented both the correct and incorrect formulations of turbine-induced added turbulence in a Gaussian wake model to assess their impact on the Horns Rev 1 wind farm. Results indicate differences of 1.94% and 3.53% in predicted turbulence intensity and normalized power of waked turbines between the correct and incorrect formulations, respectively, at an ambient turbulence intensity of 7.7%. These discrepancies increase to 2.7% and 4.95% at an ambient turbulence intensity of 4%.
JOURNAL OF WIND ENGINEERING AND INDUSTRIAL AERODYNAMICS
(2023)
Article
Engineering, Multidisciplinary
N. Zehtabiyan-Rezaie, S. Rahimi Damirchi-Darasi, M. H. Fazel Zarandi, M. Saffar-Avval