Deformation prediction for deep excavations based on genetic algorithms-GRNN

Affected by various factors, the deep excavation has become one of the key problems in geotechnical engineering. In practice, the deformation must be controlled rigorously according to the actual situation, surrounding environment and building safety grade. The intelligent construction has become one of the tendencies of deep excavation engineering, that is, it is to predict the deformation of retaining structures by neural network by collecting and analyzing monitoring data which record the deformation information. The generalized regression neural network (GRNN) is studied based on the genetic algorithm (GA). In this algorithm, GA is adopted to search the optimal smooth factor which is the only factor of GRNN, and then the GA-GRNN is used for prediction. The simulation experiment indicates that the proposed method is effective in time series prediction.