Online monitoring system of slope based on the RBF neural network

Abstract

Monitoring the deformation and other useful physical parameters of slopes, either natural mountain slopes or those formed in open pit mining or construction, is one of the most effective methods to avoid or reduce the risk caused by landslides. Within this paper, a set of multi-point displacement sensors and anchor rod stress sensors was preinstalled in a slope. The information from the sensors was timely collected by an A/D converse module, and through a RS485 network the data in the collect module was then passed to the host computer, where the data about multi-point displacement, outside deformation and the load of anchors was analyzed. Applying RBF neural network, it can assess the stability of the slope and also can forecast landslides. Since monitoring system is a online one, all those processes can be done automatically, and the data can be taken anytime wanted. ©2009 IEEE.