Prediction foundation pit deformation based on least squar support vector machine regression

The least squares support vector machine(LSSVM) regression was proposed to predict the deformation of foundation pit.On the basis of measured foundation pit displacement data with real time serie,the model of foundation pit displacement with time was built by LSSVM regression.The results show that LSSVM regression is an effective method to achieve high accuracy for predicting the deformation of foundation pit.The proposed modeling approach is superior to classical back-propagation(BP) neural network in terms of computation speed,forecast errors and less data needed.