Real-Time testing of novel robust digital pitch controller for digital hydraulic pitch system in wind turbine

ABSTRACT

This research work draws insight on developing a novel digital hydraulic pitch system for a large-scale wind turbine. Digital hydraulic pitching can be appropriate for wind turbine application due to its high response, high power-to-weight ratio, and robustness. To perform digital pitch control, a machine learning-based robust digital pitch controller was designed and implemented in the developed real-time test setup. The proposed real-time test setup consist of sub-systems such as digital hydraulic system hardware, real-time interface, robust digital pitch controller, and 5 MW benchmark wind turbine model. Real-time testing was conducted on proposed controller using the developed test setup. Robustness test exhibited robust pitch control tracking when subjected to high wind gusts/fluctuations. In this test, the proposed controller reduced the pitch control error (mean absolute) by 68.10% and 51.75% when compared to feed forward and Elman controller, respectively. Subsequently, a power regulation test was performed on the proposed controller under real-time turbulence wind speed. Relative to the feed-forward and Elman controller, the proposed controller regulates power close to the rated value and improved the mean generator power by 3.39%, and 3.72%, respectively. A comparative study conducted with existing controllers (in the literature) demonstrate the superior performance of the proposed controller. This study reports that intelligent digital pitching action of proposed controller generates power close to rated power when compared to existing controllers in the literature.

KEYWORDS:

• Digital hydraulics pitch system
• machine learning
• power regulation test
• real-time robustness test
• robust digital pitch controller

Acknowledgments

This work was funded by DST-SERB [Grant No: TAR/2019/000008]. We would like to express our deepest thanks to Dr. Raghunathan Rengaswamy for his support and guidance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Science and Engineering Research Board [TAR/2019/000008].