ScholarMate
English简体中文繁體中文
客服热线:400-1616-289
登录注册
期刊论文

Enhanced variational mode decomposition with deep learning SVM kernels for river streamflow forecasting

Deepa, Subramaniam Nachimuthu; Natarajan, Narayanan*; Berlin, Mohanadhas
Environmental Earth Sciences2023-11DOI: 10.1007/s12665-023-11222-5
3被引4阅读
引用收藏
分享

摘要

The present scenario of global climatic change challenges the sustainability and existence of water bodies around the globe. Due to which, it is always important and necessary to forecast the streamflow of rivers with respect to natural precipitation process. In this research study, novel enhanced variational mode decomposition (EVMD) with deep support vector machine (DSVM) kernels is proposed to perform forecasting of river streamflow. The developed computational intelligent machine learning model is a hybrid combination of the new enhanced VMD and the novel deep SVM kernels that is trained suitably to forecast the streamflow with respect to their deep learning layers. Initially, singular spectrum analysis (SSA) is employed for noise removal and the enhanced VMD with its features of decomposing and extracting more prominent features from the data are hybridized with the deep SVM kernel models to predict the streamflow of the considered Cahaba River data sets. Deterministic grey wolf optimizer (DGWO) is modelled in this research paper to fine tune the parameters of the deep SVM model. Previous prediction techniques modelled had difficulties in respect of local and global minima occurrences, stagnation, delayed and premature convergence and so on. Hence, in this study, the hybrid deep learning model forecasted the streamflow for the considered data sets and its superiority was validated with the comparative analysis with the previous forecasting techniques adopted. The developed forecasting model shall be used by the hydrologists for predicting the daily streamflow with the highest prediction accuracy rate of 97.54% with respect to training process and in case of testing mechanism the prediction accuracy rate is 96.47% This 97.54% of training prediction accuracy rate confirms the effectiveness of modelled new deep SVM algorithm for the streamflow forecasting of hydrologists.

学科与关键词

关键词
Deep learningRiver streamflowEnhanced variational mode decompositionSupport vector machine
相关
加载中…
  • A novel time-varying stepwise decomposition ensemble framework for forecasting nonstationary and nonlinear streamflow
    Guo, Tianli; Song, Songbai*; Singh, Vijay P.; Wei, Ting; Zhang, Te; Liu, XinJournal of Hydrology, 2023, 617: 128836.被引 14
  • Ensemble streamflow forecasting based on variational mode decomposition and long short term memory
    Sun, Xiaomei; Zhang, Haiou; Wang, Jian; Shi, Chendi; Hua, Dongwen; Li, Juan*Scientific Reports, 2022, 12(1): 518.被引 15
  • Improving the prediction accuracy of river inflow using two data pre-processing techniques coupled with data-driven model
    Nazir, Hafiza Mamona; Hussain, Ijaz*; Faisal, Muhammad; Elashkar, Elsayed Elsherbini; Shoukry, Alaa Mohamd*PEErJ, 2019, 7: e8043.被引 4
查看全部 →

发表信息

论文状态
公开发表
期刊
Environmental Earth Sciences
发表日期
2023-11
卷期
第 82 卷, 第 22 期
页码
544
DOI
10.1007/s12665-023-11222-5
语言
ENGLISH
成果收录
SCIE

作者

  • D

    Deepa, Subramaniam Nachimuthu (第一作者)

  • N

    Natarajan, Narayanan (通讯)

查看全部 3 位作者收起作者
  • B

    Berlin, Mohanadhas

登录解锁更多功能

  • 收藏这篇,随时在“我的 · 收藏”找回
  • 认领你的成果,生成推广素材
  • 关注作者获取最新成果
  • 加入项目组、协同写作
注册 / 登录

产品服务

  • 科研之友
  • 创新城
  • 科创云

服务支持

  • 帮助中心
  • 隐私政策
  • 服务条款

联系方式

在线客服:【】
客服热线:400-1616-289
电子邮箱:support@scholarmate.com

关注或下载科研之友

微信二维码
微信公众号
客户端下载二维码
下载客户端
科研成果科研人员科研机构科研动态爱瑞思软件

©2026 深圳市科研之友网络服务有限公司

公安备案图标粤公网安备 44030502000213
粤ICP备 16046710 号粤B2-20110417