Design optimization of wood-carved window grilles in historical architectures using stable diffusion model and intuitionistic Fuzzy VIKOR

Wood-carved window grilles in traditional Chinese historical architectures, as an important cultural heritage, bear profound historical value and exemplify the extraordinary craftsmanship of ancient artisans. However, due to natural weathering and human damage, protecting and preserving these artifacts has become a major challenge. This study proposes an optimization design method for wood-carved window grilles that combines the Stable Diffusion Model (SDM) with Intuitionistic Fuzzy VIKOR. First, 2476 window grille patterns were collected to create a diverse database to support algorithm training. Next, the SDM was employed to learn the design patterns of the window carvings, generating designs that blend traditional charm with innovative styles. After high-quality samples were selected by design experts, the Intuitionistic Fuzzy VIKOR was used to prioritize the esthetic appeal of the patterns, ensuring that the designs met contemporary esthetic standards. These innovative patterns were then integrated into creative lamp design, and optical simulation experiments were conducted using TracePro software to evaluate the best options based on translucency, surface illuminance, and uniformity. Through a hybrid approach combining artificial intelligence and multi-criteria decision-making, this paper provide traditional artisans with a steady stream of creative inspiration, shorten the R&D cycle for wood carving products, and offer new directions for the protection and modernisation of wood carving cultural heritage.