Flow velocity prediction method based on deep learning

Zihan Zhao; Zihao Yang

doi:10.1117/12.3015667

20 February 2024 Flow velocity prediction method based on deep learning

Zihan Zhao, Zihao Yang

Proceedings Volume 13064, Seventh International Conference on Traffic Engineering and Transportation System (ICTETS 2023); 130640P (2024) https://doi.org/10.1117/12.3015667
Event: 7th International Conference on Traffic Engineering and Transportation System (ICTETS 2023), 2023, Dalian, China

Abstract

The prediction of port channel flow velocity is very important for offshore operation, navigation safety and coastal engineering construction. This paper focuses on the prediction of short-term flow velocity in port waters. By considering the characteristics of time dependence and feature dependence in flow velocity prediction, a hybrid model of CNN, Bi _ LSTM and self-attention mechanism is integrated, and a SA-CNN-Bi _ LSTM model is proposed to improve the performance of traditional models in flow velocity prediction. The port flow velocity prediction model based on deep neural network established in this paper can take advantage of CNN and Bi _ LSTM to extract features

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Zihan Zhao and Zihao Yang "Flow velocity prediction method based on deep learning", Proc. SPIE 13064, Seventh International Conference on Traffic Engineering and Transportation System (ICTETS 2023), 130640P (20 February 2024); https://doi.org/10.1117/12.3015667

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available