Displacement Prediction Model of Earth-Rock Dam Based on Outlier-robust Extreme Learning Machine

Abstract

Aiming at the problem of low precision and poor forecasting results of traditional statistical model and neural network model under the influence of abnormal discrete values. This paper construct a new dam displacement prediction model based on outlier-robust extreme learning machine and the characteristics of environmental factors affecting the deformation of earth-rock dam. Compared with other models, the outlier-robust extreme learning machine model is more robust, anti-noise and easy to generalize. The analysis of engineering example shows that the ORELM model has higher fitting precision and better forecasting results. It is advanced and reasonable in the prediction of earth-rock dam displacement, and it can be used as a reference for similar projects.