A Combined Safety Monitoring Model for High Concrete Dams

When applying reliability analysis to the monitoring of structural health, it is very important that gross errors–which affect prediction accuracy–are included within information. An approach using errors identification and a dam safety model for deformation data concrete dams proposed in this paper. It can solve problems strong nonlinearity difficulty identifying eliminating dams. This new method combines advantages an incremental extreme learning machine (I-ELM) seek optimal network structure, Least Median Squares (LMS) with robustness multiple failure points, robust estimation IGG good outliers (gross errors) (ELM) high efficiency handling nonlinear problems. The eliminate be utilized predict behavior existing 305 m-high arch acquired by combining remote sensing technology other methods. LMS-IGG-ELM from sequence compared processing result DBSCAN clustering algorithm, Romanovsky criterion 3σ method. results show has highest rate, strongest generalization ability best effect.