Session: 09-01-10: Offshore Wind Energy - Data science and Digital twins

Paper Number: 101930

101930 - Gradient-Based Design Optimization of Fully-Flexible Floating Wind Turbines Using Modal Analysis 

To meet growing demand, a variety of different floating foundation concepts for offshore wind have been developed.  Multidisciplinary design optimization can enable efficient exploration of the broad floating wind turbine design space and guide further design work.  Previous design optimization studies have been limited by model simplifications and gradient-free optimization approaches.  Typical model simplifications have included the assumption of rigid body motions, the use of few design variables, and a focus on a single floating foundation concept.  These limitations make it difficult to consider detailed structural models and geometries, and high computational costs of gradient-free optimization make it infeasible to include fatigue-based design constraints.  In this work, the use of gradient-based optimization methods and an efficient coupled analysis model based on flexible modes makes it feasible to consider a detailed structural model with reasonable computational cost.  This work builds on previously developed gradient-based optimization models for spar and tension-leg platform wind turbines by introducing increasingly generic methods for modeling and design optimization of flexible floating wind turbines.

The design optimization model is based on a linearized aero-hydro-servo-elastic model of a generic floating wind turbine.  The current implementation of the model considers a taut-moored, tension-leg platform design, though the approach can be adapted for a wide variety of platform types.  The optimization algorithm varies platform and tower sizing parameters, and the model computes each design’s motion and stress responses to wind and wave forcing in given environmental conditions.  The model uses a state-space equation of motion to combine physical and control system inputs and outputs in the frequency-domain.  To enable consideration of a wide range of flexible designs, the model computes forcing and response in generic modes rather than conventional rigid body motions.  Implementation of the model with analytical gradients in the OpenMDAO framework allows for efficient optimization, even when considering hundreds of design variables.

The optimization seeks to minimize construction cost for the design, considering material and manufacturing costs, while satisfying constraints in both fatigue and extreme environmental conditions.  Fatigue constraints are applied to the hull and tower structure, while the extreme constraints are applied for maximum motions and mooring tensions.  Additional design constraints are applied to ensure manufacturability.  The optimization is repeated with different wind turbine rotor designs to show how the optimal substructure scales with turbine power rating.  Finally, the applications of this design optimization method to even more generic floating wind turbine design models are discussed.

Presenting Author: Peter J. Rohrer Norwegian University of Science and Technology (NTNU)

Presenting Author Biography: Peter Rohrer is a PhD candidate in the Department of Marine Technology at the Norwegian University of Science and Technology (NTNU). He earned bachelor’s and master’s degrees from the University of Michigan in Naval Architecture and Marine Engineering, with a minor in Oceanography. At NTNU he is part of the SFI BLUES center, which focuses on floating structures for the next generation of ocean industries. Specifically, his work seeks to expand the use of multidisciplinary design optimization for floating wind turbine designs. His research interests include generic modeling and analysis of floating structures, novel optimization metrics, and lifecycle assessment of floating installations.

Authors:

Peter J. Rohrer Norwegian University of Science and Technology (NTNU)
Erin E. Bachynski-Polić Norwegian University of Science and Technology (NTNU)
John Marius Hegseth Dr.Techn. Olav Olsen