Optimized Design of a Squeezed-Branch Pile Group Based on an Improved Particle Swarm Algorithm

To reduce the differential settlement of pile group foundations, a squeezed-branch optimization method based on an improved particle swarm algorithm is proposed in this paper. This translates problems design into pile-bearing-plate distribution using theory variable-stiffness leveling. In process, divided groups according to top axial force pile. The finite element analysis software used solve under control algorithm, with objective functions bearing-plate number, vertical bearing capacity, value and difference as constraint conditions. An engineering example verify method. results show that optimized can by 39%, while number plate reduced 56%, which makes deformation more uniform conducive normal use structure.