Digital twin for renewable energy landscape planning and design

Abstract To address the uncertainties and the integration of multisource data during landscape planning, we proposed a digital twin-based renewable energy landscape planning and design method. First, a probabilistic graph model combined with decision-coupled network was introduced to optimize modeling uncertainties and interrelationships. Then, a landscape planning recognition model was proposed to achieve feature matching between real and simulated scenes. And heterogeneous data assimilation was used to optimize the discrepancies between multimodal data. Experimental results proved that the proposed method outperformed existing methods in several key performance indicators, particularly in model accuracy and reliability, and it significantly reduced data assimilation errors.