Ant Colony Optimization Method for Design of Piled-Raft Foundations-Article

Article

Pdf Summary

The paper develops an ant colony optimization (ACO) method to reduce the material volume of piled-raft foundations while meeting structural and serviceability requirements. Unlike conventional piled foundations, piled rafts rely on the raft to carry much of the building load, while strategically placed piles improve bearing performance and control total and differential settlement.<br /><br />The design variables include pile number and layout, pile diameter and length, and raft thickness. Constraints cover pile structural and geotechnical capacities, raft bearing pressure, and foundation movements. Soil–pile interaction is modeled in OpenSees using nonlinear p-y, t-z, and Q-z springs, while raft–soil interaction is represented with Winkler springs. MATLAB runs the ACO search and exchanges candidate designs with OpenSees for analysis.<br /><br />In a high-rise building example on medium-dense sand, the ACO method found an optimum design after about 25 cycles with linear soil analysis and 31 cycles with nonlinear analysis. Accounting for soil nonlinearity produced a foundation approximately 5% more economical than the linear-analysis design and a more uniform distribution of loads among the piles. The linear model, by contrast, concentrated loads on peripheral piles and tended to overdesign the foundation.<br /><br />The authors conclude that ACO is a promising tool for optimizing piled-raft foundations and that modeling soil nonlinearity can improve design economy. The study’s findings are based on a specific example and simplified soil conditions; comparisons with other optimization methods are identified as future work.

Keywords

ant colony optimization

piled-raft foundations

material volume reduction

soil nonlinearity

OpenSees

MATLAB optimization

pile-raft interaction

structural constraints

serviceability requirements

high-rise building foundation