A Novel Optimization Approach for Revolutionizing Architectural Design in Chinese Cultural Heritage

Abstract

Chinese cultural heritage architecture, a blend of traditional and modern construction methods, showcases the country's technological, artistic, and cultural advancements, yet its preservation faces challenges due to structural decay and aesthetic degeneration. Using Microtrans Maryland 4-1000, Dazu Rock Carvings, Nanchan Temple and Foguang Temple smokescreen were captured; surface images and ground photos were captured with smokescreen resolution of 5192 x 4153 pixels. The goal of the research is to use an Ensemble Ant Colony Fused Convolutional Capsule Neural Network (EAC-CCNN) to enhance fault analysis in Chinese cultural heritage structures images, then the combination of Augmented Reality (AR), and Building Information Modeling (BIM) to enhance the designing model for the safety management and decision making. The process entails gathering and annotating a variety of information, creating a hybrid EAC-CCNN model to investigate the architectural building's problem, training it, integrating it with BIM, doing on-site inspections, and utilizing AR-enhanced BIM models to analyze the flaws found. The findings demonstrate how this integrated method improves the precision of flaw diagnosis, fosters teamwork, and aids in the upkeep and preservation of cultural heritage assets. Metrics including accuracy and F1 score are used to evaluate the machine learning model recognizes and categorizes flaws in Chinese cultural heritage architecture. The defect identifying and safety management model of architectural designing outcome of the accuracy is 93.29% and F1 Score is 95.47%. During the training, validation, and testing phases, performance is evaluated by project goals. With the help of this method, machine learning models can be trained to identify patterns, identify flaws, and generate well-informed predictions in a variety of circumstances.