Using neural network super‐twisting sliding mode to improve power control of a dual‐rotor wind turbine system in normal and unbalanced grid fault modes

Adil Yahdou,Abdelkadir Belhadj Djilali,Elhadj Bounadja,Habib Benbouhenni
DOI: https://doi.org/10.1002/cta.3960
IF: 2.378
2024-02-17
International Journal of Circuit Theory and Applications
Abstract:This paper presents a modified super‐twisting sliding mode control (STSMC) strategy based on neural network (NN) algorithms to regulate the stator powers of a dual‐rotor wind turbine with a doubly‐fed induction generator in both normal and unbalanced grid fault modes. The strategy involves replacing the conventional STSMC gains with NN algorithms, resulting in very positive outcomes, including improved current quality, perfect stator power tracking, reduced stator power ripples, and excellent robustness. Summary According to recent research work, increasing electric power generation is one of the significant advantages of the dual‐rotor wind turbine (DRWT) compared to the other types for the same wind speed. In this research work, a modified super‐twisting sliding mode control (STSMC) based on the neural network (NN) is suggested to regulate the stator powers of a DRWT‐based doubly‐fed induction generator (DFIG) in normal and unbalanced grid fault modes. The design of this strategy involves replacing the gains of conventional STSMC with the NN algorithm to enhance robustness, mitigate the impact of unbalanced grid voltage, and consequently improve the quality of the generated power of DRWT‐based DFIG. This forms the primary contribution of this work. The suggested strategy is compared with vector control (VC) and conventional STSMC in terms of reference tracking, power ripples, response dynamics, harmonic distortion of stator current, and the effect of an unbalanced grid fault. Finally, the utility and effectiveness of the designed controller are confirmed through computer simulations. Furthermore, when the grid is subjected to a 20% voltage drop, the results demonstrate that the suggested strategy reduced the total harmonic distortion (THD) value of the stator current by 12.92% compared to VC and by 9.29% compared to conventional STSMC.
engineering, electrical & electronic
What problem does this paper attempt to address?