Abstract
Wind turbines driven by permanent magnet synchronous generators (PMSG) are widely used in the present utility industry because of their better efficiency in terms of energy generation and design simplicity. However, the performance of such a system is greatly reduced when subjected to the intermittency of a wind speed or sudden wind gust. An efficient approach was proposed by integrating the neural network (NN)-based wind turbine model with a PMSG system to accurately and efficiently estimate the maximum power and optimal turbine speed to drive the PMSG rotor. This strategy employs the speed control method for the PMSG system, where the proportional-integral controller is feedback with the error signal of the PMSG output rotor speed and the reference speed generated by the NN-based wind turbine model. The performance of the wind turbine NN-based PMSG model was investigated using both MATLAB and SIMULINK tools. The robustness of the control strategy is verified by subjecting the model to the wind speed ranging from 4 to 13 m/s to obtain the maximum power with an error as low as 0.0025%. Further, a case study was performed by using the real-time wind speed data for Hafar Al-Batin, KSA and the potential for wind power generation is explored. The maximum power and optimum reference speed to drive the PMSG rotor was estimated using the proposed model for this location and compared with the theoretically calculated maximum power and optimum reference speed with an error of less than 0.27%.
| Original language | English |
|---|---|
| Pages (from-to) | 14969-14981 |
| Number of pages | 13 |
| Journal | Arabian Journal for Science and Engineering |
| Volume | 47 |
| Issue number | 11 |
| DOIs | |
| State | Published - Nov 2022 |
Bibliographical note
Funding Information:The authors would like to thank the Interdisciplinary Research Center for Renewable Energy and Power Systems (IRC-REPS), King Fahd University of Petroleum and Minerals (KFUPM), KSA. The author, MKH acknowledges King Abdullah City for Atomic and Renewable Energy (K· A· CARE) for funding support through project K· A· CARE 182-RFP-07. K· A· CARE is also acknowledged for providing an actual onsite dataset of KSA.
Funding Information:
The authors would like to thank the Interdisciplinary Research Center for Renewable Energy and Power Systems (IRC-REPS), King Fahd University of Petroleum and Minerals (KFUPM), KSA. The author, MKH acknowledges King Abdullah City for Atomic and Renewable Energy (K · A · CARE) for funding support through project K · A · CARE 182-RFP-07. K · A · CARE is also acknowledged for providing an actual onsite dataset of KSA.
Publisher Copyright:
© 2022, King Fahd University of Petroleum & Minerals.
Keywords
- Artificial neural network (ANN)
- Maximum power point tracking (MPPT)
- Optimum turbine speed
- Wind energy PMSG system
- Wind energy conversion system (WECS)
ASJC Scopus subject areas
- General
