基于极限学习机的面板堆石坝施工期沉降预测模型及应用

• 作 者: 朱创家 季昀 李同春 单彬彬 张哲
• 机 构: 中国华电额勒赛下游水电项目(柬埔寨)有限公司,河海大学水利水电学院,河海大学水资源高效利用与工程安全国家工程研究中心,
• 出 处: 水电能源科学
• 发表日期: 2014年06期
• 关键词: 面板堆石坝 极限学习机 施工期 沉降 预测

针对大坝安全监控预测中存在的不确定性与非线性问题,提出了基于极限学习机算法的面板堆石坝施工期沉降预测方法,并将其应用于柬埔寨某面板堆石坝沉降预测中。与传统的统计回归分析模型计算的沉降量相比,基于极限学习机算法的预测模型拟合学习速度快,精度高,适用性强,预测值与实测结果吻合较好。

Aiming at the problems of uncertainty and non-linearity in dam safety monitoring and prediction, this paper proposes a method of settlement prediction for the face rockfill dam based on extreme learning machine algorithm, and applies it to the settlement prediction of a concrete faced rockfill dam in Cambodia. Compared with the traditional settlement calculation model, the prediction model based on the extreme learning machine algorithm has the advantages of fast learning speed, high precision and applicability, and the predicted value is in good agreement with the measured data.