The architecture used for supervisory and closed-loop control of wind turbines has changed relatively little over the last 20 years or so. The “eco4wind” funded project focuses on moving forward in controlling wind turbines. The aim is to improve the energy eﬃciency of wind turbines, make the supply of energy more reliable and to unlock new installation sites. For this purpose, IAV, as lead partner, is working in a partnership with Albert Ludwig University of Freiburg, Otto von Guericke University of Magdeburg, Ruhr University of Bochum and Senvion GmbH.
At the moment, the architecture for supervisory and closed-loop control of wind turbines is based on the separation between stationary supervisory control and dynamic base-layer control. At partial load control, the primary control objective centers on achieving the maximum energy yield (power control). In this case, the control system uses the power controller – with manipulatedvariable “generator torque” – to adjust generator speed, and hence rotor speed, in a way that attains the optimum tip-speed ratio.
At full load control, there is a shift in the primary control objective. Here, the focus is on stabilizing rotor speed (speed control). This is achieved by intervening to control the collective pitch angle of all rotor blades. In addition, generator torque is kept constant so that the turbine feeds in its nominal output. Over the course of time, the primary goals in both the “part-load” and “full-load” operating range have been complemented by secondary control objectives to reduce strain on various critical components. As these control strategies have evolved over the years, they involve a high degree of complexity, making it difficult to implement changes.
Potentials remain unused
This approach could prove less appropriate for modern turbines of even higher performance in future. “It hinders their potential from being exploited to the full”, explains Dr. Axel Schild, project manager at IAV for eco4wind. “For this reason, the project partners want to use the latest research results from several specialist areas of control engineering and automation technology to provide state-of-the-art technology for operating wind turbines.”
eco4wind aims at a complete revision of the present architecture for supervisory and closed-loop control of wind turbines. The focus here is on adding a new core module to existing turbine automation: an innovative, real-time supervisory controller based on nonlinear model-predictive control. This technology has already proven successful in other segments for the real-time control of large-scale plants – real-time control has been state of the art in the process industry for decades. However, given the speciﬁc requirements of wind turbines, applying existing solutions to the wind power sector is not as easy as it might seem.
Modular system permits customized adjustment
The functional configuration of the real-time supervisory controller for wind turbines is geared towards a modular system typical for the automotive industry, making it possible to customize functionalities for individual turbines and manufacturers, even at a later stage. The real-time supervisory control is to provide a completely new and intelligent way of planning and controlling plant dynamics. “This is, in fact, equivalent to a technical revolution because real-time supervisory control can replace the conventional way of controlling speed and power output as well as damping tower and drive train oscillations”, Schild says. “This will allow wind turbine operators to use their existing turbine infrastructure far more efficiently, and it also means it will be possible to save on material in rating future wind turbine systems on account of their better control.”
Levelized costs of energy to fall by two percent
The higher energy yield in conjunction with lower material costs will reduce levelized
costs of energy (LCOE) by at least two per cent. Power supply from wind turbines will also become more reliable: the demand for balancing energy will decrease and wind turbines will even be able to provide balancing energy themselves. On top of this, the results from eco4wind will facilitate new turbine designs, which will allow to access previously unprofitable sites.
The five partners are channeling their specific, complementary experience and expertise into the project. “Besides new methodological primary research and broadening existing theory, the focus is also on developing innovative control concepts and real-time-feasible algorithms”, reports Christian Schulte, head of the IAV department in charge. “The new components will then be integrated into a real-world platform and assessed in simulations and field tests.”
In the frame of the project the partners aim to achieve three central results: a fullyfledged infrastructure implemented in prototype form for controlling wind turbines (hardware and software); a fully designed engineering tool-chain implemented in prototype form that will enable energy providers to integrate the new real-time control system in their wind turbines; and test data from one of the wind turbines operating on the basis of the new methods.
The eco4wind project was launched in January 2017 and will be completed in December 2019. In total, it is being funded by the German Federal Ministry for Economic Affairs and Energy to the tune of around € 3 million.