水力发电2025，Vol.51Issue(8)：32-37,6.
基于极限学习机的齿形丁坝水毁等级评价方法
Damage Level Assessment Model for the Chevron-shaped Spur Dikes Based on Extreme Learning Machine
摘要
Abstract
The chevron-shaped spur dike is a widely used structure in the navigation channels of the tidal river sections of the lower Yangtze River,and primarily employed to control and guide water flow to prevent bank erosion.Its stability plays a crucial role in ensuring the smooth flow of the navigation channel.Currently,there is limited research on methods to assess the water damage levels of chevron-shaped spur dikes.This paper constructs an evaluation model for assessing the water damage levels of chevron-shaped spur dikes based on Extreme Learning Machine(ELM)algorithm,and the impact of the model parameters is investigated.The results show that,(a)based on a small sample of training data,the constructed ELM model exhibits excellent prediction accuracy for validation data;(b)compared to the type of activation function,the number of hidden neurons has a greater influence on the accuracy of prediction results;and(c)compared to assessment models based on SVM and KNN algorithms,the ELM model performs better in terms of prediction accuracy and evaluation time for assessing the water damage levels of chevron-shaped spur dikes.关键词
齿形丁坝/航道整治/水毁等级/评价模型/极限学习机Key words
chevron-shaped spur dike/waterway regulation/damage level/assessment model/Extreme Learning Machine(ELM)分类
建筑与水利引用本文复制引用
高攀,刘菁,黄茗,魏祥龙..基于极限学习机的齿形丁坝水毁等级评价方法[J].水力发电,2025,51(8):32-37,6.基金项目
国家自然科学基金资助项目(52309098,52309085) （52309098,52309085）
江苏省自然科学基金资助项目(BK20230960) （BK20230960）
