BP Neural Network Prediction Model of Rock Surface Displacement Caused by Anchor-Hold Change

The prestressed loss of group anchor in rock slope increase with time, which leads to the compression belt of structure plane in group anchor area was weakened, deformation of rock surface toward the free surface direction increase gradually, as a result, the slope stability was drastically reduced. Based on the group anchor layout of the abutment rock slope of an arch dam, the anchor-hold monitoring series, the rock surface displacement monitoring series, and according to the adaptive BP neural network with variable steps, a prediction model for rock surface displacement caused by the prestressed change in group anchor area was established in this paper. By using the model, the displacement of rock surface for typical days is simulated, with the result perfectly consistent with the monitoring data, which verify that when the anchor prestressed loss in group anchor area is known, the rock surface displacement can be correctly predicted by the BP neural network model established in this paper.