Abstract

Gain-scheduling is a conventional approach for adjusting the wind turbine controller performance against operating point variations in industrial applications. Although the scheduling idea seems uncomplicated, several configurations available in research papers and technical reports are subject to inaccuracy. This paper discusses the effect of inaccurate controller realization by choosing NREL 5-MW, the ubiquitous wind turbine in the research community, and citing various controller configurations available in the literature. By performing a nonlinear analysis, it is shown that hidden terms emerge from inaccurate realizations, which could deteriorate the controller performance or lead to controller structures with different performance indexes from the original design. To remedy this issue, four implementation modifications are introduced to maintain the intended performance of the gain-scheduled controller. Although the paper is focused on the baseline gain-scheduled PI controller, modifications are generalized to arbitrary pitch control systems. Simulations with the full-order dynamics of the FAST code for the pitch angle-based and wind speed-based gain-scheduled baseline controllers and a more complex gain-scheduled fractional-order PI controller verify the presented analysis.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call