Deep Foundation Pit Settlement Deformation Prediction Based on Improved Least Square Support Vector Machine Combined Model
To improve prediction accuracy of deep foundation pit settlement deformation and provide timely guidance for deep foundation pit supporting construction,an improved least square support vector machine combined model was proposed.Complete ensemble empirical mode decomposition with adaptive noise was introduced to decompose original deep foundation pit settlement deformation data.Particle swarm optimization algorithm and genetic algorithm were combined to optimize parameters of least square support vector machine.The decomposed data were trained,predicted,and then superposed to obtain final prediction results.The proposed model was used to predict the cumulative settlement of a deep foundation pit in Jinan city,and was compared with other models to verify practicability and superiority of the proposed model.The results show that the mean relative error,mean square error,and root-mean-square error of the proposed model for predicting the cumulative settlement are 0.035％,0.080 9 mm2,and 0.283 8 mm.The accuracy of the proposed model is far better than that of other models.The introduction of complete ensemble empirical mode decom-position with adaptive noise is more conducive to take advantage of least square support vector machine in deep foundation pit settlement deformation prediction.
deep foundation pit settlement deformationleast square support vector machineempirical mode decomposi-tionparticle swarm optimizationgenetic algorithm
NETL
NSTL
万方数据
维普
