Field testing of multi-variable individual pitch control on a utility-scale wind turbine

Abstract

The ongoing increase in size of modern wind turbines creates the demand of better control algorithms to counteract the increased loads acting on them. Individual blade pitch control (IPC) algorithms to alleviate blade loads are becoming a crucial part of current wind turbine control systems. The state-of-the-art individual blade pitch load reduction controller mostly relies on multiple single-input single-output (SISO) designs. This approach, however, constrains the achievable bandwidth of the controller as relevant couplings in the dynamics are ignored. These couplings become more pronounced for bigger turbines. Model-based multiple-input multiple output (MIMO) control designs can be used to account for these couplings and hence reduce loads for future, larger turbines. In this article we present the results of an intensive field test campaign of an H∞-design based MIMO IPC. This controller is designed for and tested on the utility-scale 2.5 MW Clipper Liberty research turbine operated by the University of Minnesota. The article guides the reader through the turbine’s open loop dynamics, the IPC design and the relevant implementation steps on the turbine. Finally, the developed H∞-based IPC is compared using experimental field data against no IPC (collective blade pitch control) and a classical IPC based on decoupled SISO loops.