Data-Driven Planning for Grid-Connected Advanced Air Mobility Networks

To curb road traffic congestion and reduce carbon emissions, electric vertical take-off and landing (eVTOL) aircraft are attracting attention from both academic and industrial communities. As the optimal planning of vertiport infrastructure is crucial for enabling widespread eVTOL adoption and fostering the low-altitude economy (LAE), this paper addresses the integrated planning of advanced air mobility (AAM) networks integrated with power grid systems under energy consumption uncertainty. We propose a two-stage data-driven distributionally robust optimization (DRO) model to determine the location and capacity of vertiports and the fleet size of eVTOLs in the first stage, while solving the operational problem of the integrated AAM and power networks in the second stage, capturing the eVTOL flight profile. We model the eVTOL energy consumption uncertainty by using a distributional ambiguity set constructed from moment information. To address the computationally challenging two-stage DRO problem, we develop a tailored decomposition framework that integrates a modified delayed constraint generation algorithm with the K-adaptability reformulation and the linear programming (LP) relaxation, yielding valid lower and upper bounds. Numerical case studies demonstrate the effectiveness of the proposed model and solution methods. Our approach provides a robust and practical decision-making tool for the scalable development of AAM infrastructure.