Optimization of Structural Design Parameters in BIM Using Grid Search and Immune Genetic Algorithm

Abstract

As competition in the construction industry intensifies, construction companies must optimize structural design to enhance their competitiveness. However, traditional methods of structural optimization heavily rely on manual experience, which results in inefficiency and lack of accuracy. Therefore, this study proposes a structural design parameter optimization model based on Building Information Modeling, integrating Grid Search (GS) and Immune Genetic Algorithm (IGA). Accuracy is determined by calculating the average deviation rate, while convergence speed is assessed by the number of iterations required for the algorithm to reach converge. This model fully utilizes the global optimization capabilities of grid search method and the immune genetic algorithm to solve the problems of BIM technology lacking quantitative computing ability and low efficiency in processing large amounts of data. The results indicate that the GS-IGA model achieves a curve area of 0.97 under the Zero-Conductivity Transition test function, an F1 score of 0.98, an accuracy of 95.6%, and fast convergence speed, outperforming the genetic algorithm, particle swarm optimization, and simulated annealing algorithm. In addition, in the structural optimization case study of a factory building, the GS-IGA model reduced the required steel reinforcement weight by 7.8%, concrete weight by 8.7%, and overall cost by 9.5% compared to the original structure. These results indicate that the GS-IGA model demonstrates excellent efficiency and applicability in structural design parameter optimization, effectively solving the problems of inefficiency and inaccuracy in traditional optimization methods. It offers an innovative approach for building structure optimization and contributes to the advancement of intelligence in the construction industry.