The application and study of a neural network model based on multivariate phase space reconstruction

For multi-variable nonlinear system evolution with time-varying, a neural network model based on multi-variable phase-space reconstruction has been proposed, and is used in civil engineering for synthesized deformation prediction of deep foundation pit. By the various time series time delay and embedding dimension determined respectively in this model, the multi-variable series of excavation deformation for deep foundation pit has been done in the first phase space reconstruction. The neural network input extraction by the use of partial least squares regression method can be the strongest impact components. Finally non-linear fitting between the various components has been completed via BP neural network model. With practical application for deformation prediction of deep foundation pit, the method´s effectiveness has been verified.