Lucy Y. Pao

This article addresses the problem of wake steering control for wind farms that explicitly consider the tradeoff between farm-level power generation and yaw duty cycle under variable and uncertain wind conditions. A novel stochastic model predictive control (MPC) algorithm is presented, which utilizes a stochastic model of the freestream wind field components in a receding horizon framework to compute optimal yaw set points that maximize the expected value of the farm power while constraining the yaw actuation. Different configurations of the algorithm are evaluated using a steady-state wind farm simulator. The proposed stochastic MPC algorithm can plan control actions over a future prediction horizon based on probabilistic estimates of the incoming wind magnitude and direction.

Floating offshore wind turbines (FOWTs) unlock superior wind resources and reduce operational barriers. The dynamics of FOWT platforms present added engineering challenges and opportunities. While the motion of the floating platform due to wind and wave disturbances can worsen power quality and increase structural loading, certain movements of the floating platform can be exploited to improve power capture. Consequently, active FOWT platform control methods using conventional and innovative actuation systems are under investigation. This paper develops a novel framework to design nonlinear control laws for six degrees-of-freedom platform motion. The framework uses simplified rigid-body analytical models of the FOWT. Lyapunov’s direct method is used to develop actuator-agnostic unconstrained control laws for platform translational and rotational control. A model based on the NREL-5MW reference turbine on the OC3-Hywind spar-buoy platform is utilized to test the control framework for an ideal actuation scenario. Possible applications using traditional and novel turbine actuators and future research directions are presented.

This chapter introduces a simple payoff-based learning rule that is completely decentralized and leads to an efficient configuration of actions in any n-person finite strategic-form game with generic payoffs. Agents respond solely to changes in their own realized payoffs, which are affected by the actions of other agents in the system in ways that they do not necessarily understand. The methodcan be applied to the optimization of complex systems with many distributed components, such as the routing of information in networks and the design and control of wind farms.

The expected growth of floating offshore wind energy is remarkable, with the U.S. Department of Energy projecting exponential increases in installed capacity persisting over the next few decades [1] . Reductions in cost of energy that have attracted industry adoption have in part been brought about through control advancements, including the introduction of auxiliary feedback loops in addition to conventional single-input, single-output (SISO) controllers. However, industry projects tend to be risk-averse, favoring incremental pragmatic solutions to immediate problems [2] , so many sophisticated control methods in academia have not been applied at scale due to their complexity. There is a use-case for simple intuitive controllers that achieve improved performance without significantly increasing the complexity of the controller, such as structured multiloop control (see “Summary”).

As the global demand for renewable energy sources intensifies amid the urgent fight against climate change [1] , offshore wind energy has emerged as a promising and crucial component of the sustainable energy portfolio. Fixed-bottom offshore wind farms have already demonstrated their potential; however, they are limited to relatively shallow waters, typically no deeper than 60 m.

Climate change poses an existential threat to humanity. It is now indisputable that the primary cause of this threat is human activity resulting in high greenhouse gas (GHG) emissions, which began during the Industrial Revolution and have continued to rapidly accelerate. The first warnings of impending and irreversible climate change were sounded decades ago, when governmental and intergovernmental policy makers had sufficient time to enact the changes needed to avoid the dire situation we find ourselves in today.

To develop better wind farm controllers that can meet more complex objectives, methods of modeling the wind turbine wakes at low computational expense are needed. Gaussian process (GP) regression offers a computationally inexpensive framework for learning complex functions from noisy measurements with very few datapoints. In this work, an online learning approach is presented to learn the rotor-averaged wind velocity at downstream wind turbines with GPs, using the available datastream of wind field measurements and wind turbine control set-points. This framework can readily be integrated into model-based controls methods because the model a) is updated online at low computational expense, b) assumes a mathematically favorable Gaussian form, and c) explicitly quantifies the stochastic nature of the wake field so that the trade-off between exploration and exploitation, and the uncertainty in the prediction, can be utilized. We show that a GP-learned model can match true values with errors within 0.5% on average, with as few as 5 training data points.

In this final report, we summarize activities and results from the Ultraflexible Smart Floating Offshore Wind Turbine (USFLOWT) project. The project involved co-design of a 10-MW floating offshore wind system, especially the substructure. This substructure was based on the SpiderFLOAT, an ultra-compliant, ultralight, and modular system with actuators for smart platform control that was promising for control co-design. The project’s overall goal was to unlock the floating offshore wind market based on conditions on the North Atlantic coast of the United States. Tasks summarized in the report include Task 2.3 (including controller design for the turbine and the substructure and final design of the SpiderFLOAT), Task 2.4 (final design of the active can ballasting actuator), Task 2.5 (planning for a scale demonstrator), Task 3.1 (technology to market plan), and Task 3.3 (approval in principle). The report also summarizes project dissemination activities at a variety of venues including the lead author and/or presenter for each.