基于误差修正EOS-ELM的滑坡位移预测

发布时间：2018-03-17 04:00

  本文选题：滑坡位移预测　切入点：集成学习　出处：《华中科技大学学报(自然科学版)》2017年09期 　论文类型：期刊论文

【摘要】：提出了一种基于误差修正在线贯序超限学习机集成(EOS-ELM)的滑坡位移预测模型.预测过程中对滑坡位移时间序列进行了趋势项和周期项分解,分别考虑了不同的影响因子对滑坡趋势项位移和周期项位移的影响.利用在线贯序超限学习机(OS-ELM)算法分别对趋势项位移和周期项位移建模预测.采用集成预测的思想提高OS-ELM模型的泛化能力,同时为了进一步提高预测精度,提出了一种在线误差修正预测方法.该方法通过对误差序列进行建模预测,修正最终的预测结果.以三峡库区白水河滑坡为例,实验验证了提出方法的有效性.
[Abstract]:This paper presents a landslide displacement prediction model based on error correction online sequential overrun learning machine integrated with EOS-ELM. In the process of prediction, the trend term and period term of the time series of landslide displacement are decomposed. The influence of different influence factors on the displacement of the trend term and the periodic term of landslide is considered respectively. The method of on-line sequential over-limit learning machine (OS-ELM) is used to model and predict the displacement of the trend term and the periodic term, respectively. The integrated method is used to predict the displacement of the trend term and the periodic term. To improve the generalization ability of OS-ELM model, At the same time, in order to further improve the prediction accuracy, an online error correction prediction method is proposed. By modeling and forecasting the error series, the final prediction results are corrected. Taking the Baishui River landslide in the three Gorges Reservoir area as an example, The effectiveness of the proposed method is verified by experiments.
【作者单位】： 武汉理工大学自动化学院;华中科技大学自动化学院;中南民族大学计算机科学学院;
【基金】：国家自然科学基金资助项目(61503144) 中央高校基本科研业务费专项资金资助项目(2017IVA058)
【分类号】：P642.22
，

本文编号：1623038