|
| 基于响应面法的高填方渠道运行期沉降多因子分析及预测研究 |
| Multi-factor analysis and prediction of settlement of high fill channel during operation based on response surface method |
| 投稿时间：2025-02-21 修订日期：2025-06-04 |
| DOI： |
| 中文关键词: 高填方渠道 沉降分析与预测 响应面法 白鲸优化算法(BWO) 极限学习机(ELM) |
| 英文关键词:high-fill channel settlement analysis and prediction response surface method beluga whale optimization (BWO) extreme learning machine (ELM) |
| 基金项目:国家自然科学基金企业创新发展联合基金重点项目, U23B20144, 王媛；国家自然科学基金长江水科学联合基金重点项目, U2240210, 王媛；国家自然科学基金青年,52209129,任杰 |
|
| 摘要点击次数: 32 |
| 全文下载次数: 0 |
| 中文摘要: |
| 高填方渠道沉降往往是由多因子综合作用造成的,但不同水文地质条件的渠段,其沉降因子的种类和敏感性也有所不同。针对南水北调中线总干渠某高填方沉降段,采用响应面法设计62组有限元数值试验,重点研究了地下水埋深、地下水位降幅、堆土高度、弹性模量、泊松比、粘聚力、内摩擦角等因素对高填方渠道沉降的敏感性程度,分析单因素及多因素交互作用对高填方渠道沉降的影响。将数值试验结果作为数据集,通过白鲸优化算法(BWO)对极限学习机(ELM)的输入层与隐含层之间的连接权重、隐含层初始权重进行优化,构建基于白鲸优化算法改进极限学习机(BWO-ELM)的高填方渠道沉降预测模型。结果表明,地下水位降幅对高填方渠道沉降有显著影响,且地下水位降幅与堆土高度交互作用对高填方渠道沉降影响显著；BWO-ELM预测模型的均方根误差()为0.015,决定系数()为0.971,平均绝对误差()为0.012,预测精度优于传统ELM模型,能精确地预测渠道沉降值。研究结果可为高填方渠道沉降成因分析及预测起到借鉴作用,对保障总干渠工程安全和输水安全有重要意义。 |
| 英文摘要: |
| The settlement of high fill channel is usually caused by the comprehensive action of multiple factors, but the types and sensitivities of settlement factors are different in different hydrogeological conditions. In this paper, 62 sets of finite element numerical tests are designed using response surface method for a high fill settlement section of the main channel of the middle route of the South-to-North Water Transfer Project. The sensitivity of groundwater depth, groundwater level drop, pile height, elastic modulus, Poisson"s ratio, cohesive force, internal friction Angle and other factors to the settlement of high fill channels is studied. The influence of single factor and multi-factor interaction on the settlement of high fill channel is analyzed. Taking the numerical test results as data set, the connection weights between the input layer and the hidden layer and the initial weights of the hidden layer of the extreme learning machine (ELM) were optimized by the Beluga optimization algorithm (BWO), and the settlement prediction model of the high fill channel based on the optimization algorithm of the extreme learning machine (BWO-ELM) was constructed. The results show that the decrease of groundwater level has a significant effect on the settlement of high fill channel, and the interaction between the decrease of groundwater level and the height of soil pile has a significant effect on the settlement of high fill channel. The root mean square error (), determination coefficient () and mean absolute error () of BWO-ELM model are 0.015, 0.971 and 0.012 respectively. The prediction accuracy of BWO-ELM model is better than that of the traditional ELM model, and it can accurately predict the settlement value of the channel. The research results can provide reference for analysis and prediction of settlement cause of high fill channel during operation period. |
| 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
