Geometrical Multidisciplinary Design Analysis and Optimisation for Airborne Wind Energy Systems

Abstract

The transition to sustainable energy sources is increasingly urgent, driving innovation in renewable energy technologies. Airborne wind energy (AWE) is a promising emerging alternative to conventional wind turbines, using tethered kites to harvest wind energy at higher altitudes where wind speeds are stronger and more consistent, while requiring significantly less material. Kitepower B.V. employs a leading-edge inflatable (LEI) kite flying in crosswind patterns to generate traction force on a tether that is reeled out from a ground-based drum connected to a generator. Energy is produced through repeated reel-out and energy-efficient reel-in cycles. As Kitepower is currently in active development, the kite design process requires many iterative design cycles and extensive testing, which are time-consuming, expensive, and heavily reliant on expert knowledge. This thesis investigates the application of multidisciplinary design analysis and optimisation (MDAO) to the geometric design of a LEI kite for ground-generation crosswind AWE systems, with specific focus on the system developed by Kitepower B.V. A computational framework is propose that integrates kite shape parametrisation, aerodynamic and structural analysis, and performance modelling within an MDAO toolchain. The kite geometry is described using a set of decoupled parameters that enable systematic reshaping while maintaining consistent reference properties. Two simulation frameworks are proposed to support this concept. The first is a conceptual high-fidelity toolchain that integrates aerodynamic, structural, and dynamic performance models, intended to guide future development of a fully coupled MDAO environment. The second is a simplified, computationally efficient toolchain that was implemented to enable quantitative optimisation under realistic time constraints. Using annual energy production as the optimisation objective, the implemented toolchain is employed to verify the optimisation methods and evaluate performance improvements relative to a reference kite design within a constrained design space. The results demonstrate that MDAO can effectively support early-stage geometric kite design for AWE applications, offering a fast, systematic alternative to traditional expert-driven design iteration.