Research on Intelligent Algorithm for Rolling Prediction of Horizontal Displacement of Deep Foundation Pit Piles
How to accurately predict the horizontal displacement of the enclosure structure during the construction of the deep foundation pit is one of the important indicators to judge whether the deep foundation pit is in a safe state.In this paper,genetic algorithm,grid search method and particle swarm optimization algorithm are established to optimize the support vector machine model.The two models are trained to learn the sample data of horizontal displacement of the enclosure structure,and the applicability of the algorithm is verified by comparing the prediction results of different piles after learning for the same time with the prediction results of the same pile after implementing multi-step rolling prediction.The results of the study show that the genetic algorithm generally outperforms the other two algorithms in the actual situation of less on-site construction monitoring data,and proposes a highly accurate approach for the envelope deformation.The multi-step rolling prediction has a better fitting effect and improves the accuracy of the foundation pit displacement trend warning.
deep foundation pitalgorithm optimizationsupport vector machinerolling prediction
国家科技期刊平台
NSTL
万方数据
