GPU-Driven Visualization Technology for Large-Scale BIM Models in Power Grid Engineering

This paper aims to present a study on efficient display techniques for large-scale Building Information Models (BIM) based on the autonomous and controllable BIMBase platform. The research findings can be applied to three-dimensional design in power grid engineering. The poor user experience caused by lagging in the display of large-scale BIM models during power grid engineering design significantly limits the design efficiency. To address these issues, this research optimizes the display efficiency of large-scale power grid engineering models by leveraging the autonomous and controllable BIM software BIMBase, combined with GPU-Driven technology. The proposed method fully utilizes the advantages of GPU pipeline calculation and implements a GPU-based geometry discretization system, significantly reducing the CPU load. Well-selected representative real-world power grid engineering projects are used to validate the optimized technique proposed in this paper. The validation results demonstrate that the adoption of this optimization technique significantly improves the display efficiency of BIMBase for large-scale power grid BIM models, solving the challenge of efficient display. The research findings are of significant value for improving the efficiency of three-dimensional design in power grid engineering and serve as a reference for enhancing the display efficiency of domestically developed autonomous BIM platforms.