New Methods for Digital Twin Modelling of Wave and Wind Energy Systems
Renewable energy is a rapidly growing sector within the realm of electricity generation. While the field itself is incredibly broad, wind and wave energy are the primary focus of this thesis. Wind turbines have existed for hundreds of years, with their first application to electricity production occurring over 100 years ago. Wave energy, on the other hand, is a comparatively new field, having been seriously explored within only the past 20 years. However, both sectors are intimately related to the field of structural dynamics, and the goal of producing electricity at a rate competitive with traditional sources of energy plays a large role in the development of new technologies in these fields. In this respect, validated design tools are one of the key needs for achieving (in the case of wave energy) and maintaining (in the case of wind energy) a market-competitive levelized cost of energy (LCOE). In this thesis, new methods for developing digital twin models based on dynamic modelling and multiple uses of the resulting models are derived, detailed, and discussed for both wave and wind energy systems. Wave energy is the focus of the first half, in which a coupled dynamics model of a point absorber-style wave energy converter (WEC) and the bridge to which it is mounted is developed and validated against experimental data. The primary benefits of the resulting model are its simplicity and decreased simulation time relative to other available WEC models while still providing an acceptable degree of accuracy. Such a model would be particularly useful in the realm of controls system development, which will become increasingly important in the field for larger-scale devices. The model is then utilized to explore the relationship between power production and fatigue damage in the realm of wave energy. This issue is of great importance in the young field of wave energy as the feasibility of WECs lies in their ability to produce power at a sufficiently low LCOE. The second half of this thesis focuses on wind energy and a novel technique for developing an aero-structural digital twin of an existing utility-scale 1.5MW wind turbine. In this technique, experimental data from an operating wind turbine is used to calibrate the properties of a baseline turbine model to represent the dynamic behavior of a target wind turbine. The blade aerodynamics and structural dynamics are the primary focus of this technique, in which the power curve and thrust coefficient data of an experimental turbine is used to calibrate the blade aerodynamic properties of the model, and experimental blade total mass, center of gravity, and natural frequencies are used to calibrate the model blade mass and stiffness property distributions, respectively. A primary benefit of this methodology is its relative ease of implementation in the creation of models of multiple similar turbines (such as those in a fleet) once a baseline model is in place. In short, the overarching goal of this thesis is to provide and explore two unique dynamic model development methodologies for existing renewable energy systems (i.e., digital twins) and explore their potential benefits in the realm of LCOE reduction.