Landslide displacement prediction based on error correction and ensemble of online sequential extreme learning machine

A novel prediction approach based on error correction and ensemble of online sequential extreme learning machine (EOS-ELM) was proposed in this paper for the landslide displacement prediction. Following the proposed approach, landslide displacement time series was divided into the trend component displacement and the periodic component displacement to express the relations between landslide displacement and different affecting factors. An online sequential extreme learning machine (OS-ELM) algorithm was adopted to forecast the trend component and periodic component landslide displacements, respectively. The ensemble learning method was used to improve the generalization ability of OS-ELM. For further improving the forecasting accuracy, an error correction method was proposed. This method utilized error series to build a predictor which was used to correct the final outcomes. The effectiveness of the proposed method was evaluated by using real data from Baishuihe landslide in the Three Gorges Reservoir of China. © 2017, Editorial Board of Journal of Huazhong University of Science and Technology. All right reserved.