Three dimensional finite element analysis and artificialintelligence assisted optimization of piled raft foundations

The increasing demand for high-rise building designs has led to the need for optimization of piled raft foundations, a widespread type of foundation, with novel methods for their economical and safe design. Numerous parametric three-dimensional finite element (3D FE) computations were performed in this study to optimize different piled raft foundation characteristics using high-capacity workstations. The 3D FE models were validated utilizing data from a structure in Frankfurt (Germany). Following validation, 90 distinct combinations of 3D FE examinations were conducted by varying the piles' diameter, number, and modulus of elasticity in the parametric piled raft foundation system. The raft foundation system's settlement was examined in relation to variations in the piles' diameter, number, and modulus of elasticity—all of which are independent variables in the designs utilizing the data from the three-dimensional finite element analysis of the piled raft system. An optimization study was carried out utilizing the novel artificial intelligence-supported Goal Attainment method, which can generate powerful regression functions with the data obtained from 3D FE analysis of the piled raft foundation system. Various methods appropriate for the problem type were automatically chosen in the associated optimization process using coding supported by artificial intelligence, and the highest possible desirability value was achieved in the optimization analysis. For the economic and safe design of the model, a balance between the foundation system parameters and reliable settlements was established in the multi-objective optimization study. As a result, an artificial intelligence-supported method that provides meaningful results for general piled raft foundation systems and optimum design parameters that can guide future research and project implementers has been reached.