Numerical optimal control methods for robustness optimization of multi-wing airborne wind energy systems

空中风能（AWE）是一种可再生能源技术，旨在收集传统风力涡轮机无法达到的高海拔风，其资源仅为后者的一小部分。它通过抛弃传统风力涡轮机的基础、塔架和内转子部件，并通过用系留的自主机翼代替转子尖端来实现快速逆风飞行。电力通常是通过机翼上的小型涡轮机产生的，或者是通过系绳的周期性放出和收回来驱动地面上的绞盘。将AWE建立为可行的可再生能源技术的主要挑战之一将是公用事业规模的成本效益，这与农场配置中AWE系统可以实现的功率密度密切相关。目前，最先进的AWE系统几乎完全基于单翼配置，其特征在于大的轨迹足迹和有限的操作高度。出于安全和效率的原因，这些系统因此需要与传统风力涡轮机类似的农场间距。相比之下，在多翼AWE系统（MW-AWES）中，多个系留翼围绕共享主系绳飞行圆形反风轨迹，从而在几乎任意的操作高度实现更紧凑的轨迹。因此，它们可以在地面上更紧密地堆积在一起，同时垂直堆叠以避免尾流相互作用。因此，MW-AWES有可能在农场的地面区域提供更高的功率密度。此外，与单翼设计相比，MW-AWES设计通常使系统效率加倍，并且它们具有有益的模块化升级特性。然而，多翼布局的复杂性大大增加，到目前为止，MW-AWES仅在计算机模拟中进行了研究，其中很大一部分由申请人的研究小组进行。这一拟议研究项目的总体目标是为MW-AWES的稳健和自主运行奠定基础。为此，我们制定了三个目标。第一个目标是开发有效的问题配方的MW-AWES的鲁棒性优化。这应该允许规划协调的多翼飞行路径，满足相互依赖的约束鲁棒。第二个目标是开发MW-AWES的状态估计和模型预测控制的数值算法。这项工作的结果应允许在线规划的干扰抑制机动的整个多翼系统在很长一段时间内。第三个目标是开发基于高保真模型的MW-AWES轨迹优化算法，特别是为了考虑风场中的感应效应。这将缩小一个重要的差距，以便能够获得准确的飞行路径和性能预测，以获得年平均功率输出或可达到的功率密度。

Airborne Wind Energy (AWE) is a renewable energy technology that aims at harvesting high-altitude winds that cannot be reached by conventional wind turbines, at a fraction of the resources of the latter. It does so by discarding the foundation, tower and inner rotor parts of a conventional wind turbine, and by replacing the rotor tips by tethered autonomous wings that fly fast crosswind maneuvers. Electricity is most commonly generated via small turbines on board of the wing, or by the periodic reeling-out and -in of the tether to drive a winch on the ground. One of the main challenges to establish AWE as a viable renewable energy technology will be the cost-effectiveness at utility-scale, which is closely related to the power density that can be achieved by AWE systems in farm configurations. Currently, state-of-the-art AWE systems are almost exclusively based on single-wing configurations, characterized by a large trajectory footprint and a limited operating height. For reasons of safety and efficiency, these systems hence require a similar farm spacing as conventional wind turbines. By contrast, in multi-wing AWE systems (MW-AWES), multiple tethered wings fly circular crosswind trajectories around a shared main tether enabling more compact trajectories at almost arbitrary operating heights. Hence, they can be packed more closely together on the ground, while being stacked vertically to avoid wake interaction. Therefore, MW-AWES potentially offer higher power densities with respect to the farm's ground area. Moreover, MW-AWES designs typically double the system efficiency compared to single-wing designs, and they have beneficial, modular, up-scaling properties. However, multi-wing configurations come with a considerably increased level of complexity, and so far, MW-AWES were only investigated in computer simulations, to a significant part by the applicant’s research group. The overall aim of this proposed research project is to lay the groundwork for the robust and autonomous operation of MW-AWES. To achieve this, we formulate three objectives. The first objective is to develop efficient problem formulations for robustness optimization of MW-AWES. This should allow to plan coordinated multi-wing flight paths that satisfy interdependent constraints robustly. The second objective is to develop numerical algorithms for state estimation and model predictive control of MW-AWES. The outcome of this work should allow for online planning of disturbance rejection maneuvers for the overall multi-wing system over a long time horizon. The third objective is to develop algorithms for the optimization of MW-AWES trajectories based on high-fidelity models, in particular in order to account for induction effects in the wind field. This will close an important gap to be able to obtain accurate flight paths and performance predictions for yearly average power output or attainable power densities.