基于粒子群算法的深基坑岩土力学参数反分析

Back-analysis for Soil Parameter of Deep Foundation Based on Particle Swarm Optimization

运用人工神经网络对深基坑岩土参数进行反分析时,将粒子群算法与BP算法融合,充分发挥了粒子群算法全局寻优的能力和BP算法局部细致搜索优势.实例证明,应用该方法可提高模糊优选人工神经网络的训练效率,预估的岩土力学参数合理.

A model is established, integrating particle swarm optimization and fuzzy artificial neural network to back-analysis for soil parameter of deep foundation. The method makes full use of the global optimization of particle swarm optimization and local accurate searching of BP. The ease of back-analysis shows that this method is more efficient and has good generalization. The prediction result of soil parameter is reasonable.