Autonomous design framework for deploying building integrated photovoltaics

Abstract

The advancements in perovskite solar cells present promising prospects for the widespread deployment of Building-Integrated Photovoltaics (BIPVs). Finding an efficient and accurate approach is essential to provide deployment strategies for decision support. This study develops an autonomous decision-making design framework for BIPV, including data collection, 3D modeling, and deployment strategy. For data collection, an open-source unmanned aerial vehicle platform is produced to execute an innovation explore-then-exploit algorithm for viewpoints generation and path planning. Subsequently, point cloud models of buildings are generated using a unique deep learning-based multi-view stereo network and then converted into polygonal surface models. Moreover, a novel Grasshopper plugin component is developed to assess the economic performance of various BIPV layouts by life cycle cost analysis. Based on the analysis results, potential BIPV deployment strategies are provided to support the decision-making process. The effectiveness of the framework is validated through its application in an industrial building in Hong Kong, demonstrating a 7.6 % discrepancy in the average annual solar radiation access value. Finally, two BIPV deployment strategies are proposed for the building. This study has significant implications for the design and deployment of PV systems in urban environments, representing an important step towards supporting the transition to sustainable and low-carbon cities.