Optimization of Interior Art Design Space Engineering Technology Based on Digital Twin Modeling

Accurate perception and dynamic optimization of intelligent indoor environments are increasingly important for digital engineering systems integrating sensing, communication, and physical space management. To overcome the separation between aesthetic design and engineering performance in conventional interior design, this study proposes a multiobjective collaborative optimization framework based on digital twin modeling. A dynamic Building Information Modeling (BIM) twin is established by integrating Internet of Things (IoT) sensing data, while convolutional neural networks are employed to extract visual feature representations of interior scenes. Computational Fluid Dynamics and Finite Element Analysis are incorporated to evaluate thermal comfort and structural safety, and an improved NSGA-III algorithm performs global optimization of aesthetic quality, energy efficiency, and engineering constraints. Experimental results demonstrate that the proposed framework achieves a hypervolume value of 0.8978 after 500 iterations, with aesthetic scores reaching 8.74 and compliance rates of 96.4% for thermal comfort, 98.2% for structural safety, and 95.8% for energy consumption. By coupling real-time sensing information with digital twin feedback and multi-physics simulation, the proposed approach establishes a closed-loop optimization paradigm for intelligent indoor environments, providing a scalable methodology for electromagnetic sensing-assisted spatial perception, networked digital twins, and engineering optimization in smart built environments.