Hybrid deep learning framework for real-time DO prediction in aquaculture

Abstract Dissolved oxygen (DO) is a vital parameter in regulating water quality and sustaining the health of aquatic organisms in aquaculture environments. Therefore, estimation and control of DO levels are essential in aquaculture operations. However, traditional chemical and photochemical approach...

Full description

Saved in:
Bibliographic Details
Main Authors: Longqin Xu, Wenjun Liu, Cai Chengqing, Tonglai Liu, Xuekai Gao, Ferdous Sohel, Murtaza Hasan, Mansour Ghorbanpour, Shahbaz Gul Hassan, Shuangyin Liu
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-10786-5
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849333334652485632
author Longqin Xu
Wenjun Liu
Cai Chengqing
Tonglai Liu
Xuekai Gao
Ferdous Sohel
Murtaza Hasan
Mansour Ghorbanpour
Shahbaz Gul Hassan
Shuangyin Liu
author_facet Longqin Xu
Wenjun Liu
Cai Chengqing
Tonglai Liu
Xuekai Gao
Ferdous Sohel
Murtaza Hasan
Mansour Ghorbanpour
Shahbaz Gul Hassan
Shuangyin Liu
author_sort Longqin Xu
collection DOAJ
description Abstract Dissolved oxygen (DO) is a vital parameter in regulating water quality and sustaining the health of aquatic organisms in aquaculture environments. Therefore, estimation and control of DO levels are essential in aquaculture operations. However, traditional chemical and photochemical approaches are limited by inaccuracies, environmental interferences, time consumption, and the inability to provide real-time data. Recently, artificial intelligence techniques have been studied for DO estimation. However, off-the-shelf models, such as Random Forest (RF) and Back Propagation (BP) have demonstrated poor performance due to intricate interactions in aquatic ecosystems, which leads to complex data patterns. This study proposes a water quality estimation model by combining a convolutional neural network (CNN), self-attention (SA), and bidirectional simple recurrent unit (BiSRU). One-dimensional convolution in CNN was employed to extract effective features and input into the SA mechanism to assign weights and emphasise crucial information. The model’s accuracy is improved by incorporating BiSRU. This model evaluated the DO levels of the intensive aquaculture base in Nansha, Guangzhou City, Guangdong Province, China. The proposed CNN-SA-BiSRU achieved MSE, MAE, RMSE, and R2 of 0.0022, 0.0341, 0.0471, and 0.9765, respectively. The results of the experiments showed that the proposed model had a high level of accuracy in estimating the outcomes with minimal fluctuations in estimation errors. Moreover, accuracy for short-term prediction was significantly improved, surpassing the performance of existing methods. The highly accurate results indicate the potential of the proposed methodology for DO-level monitoring in aquaculture and its usage in the fishery industry.
format Article
id doaj-art-59102d9e54ec4e13bf4532e8ec0d21b4
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-59102d9e54ec4e13bf4532e8ec0d21b42025-08-20T03:45:53ZengNature PortfolioScientific Reports2045-23222025-07-0115111310.1038/s41598-025-10786-5Hybrid deep learning framework for real-time DO prediction in aquacultureLongqin Xu0Wenjun Liu1Cai Chengqing2Tonglai Liu3Xuekai Gao4Ferdous Sohel5Murtaza Hasan6Mansour Ghorbanpour7Shahbaz Gul Hassan8Shuangyin Liu9College of Artificial Intelligence, Zhongkai University of Agriculture and EngineeringCollege of Artificial Intelligence, Zhongkai University of Agriculture and EngineeringCollege of Artificial Intelligence, Zhongkai University of Agriculture and EngineeringCollege of Artificial Intelligence, Zhongkai University of Agriculture and EngineeringCollege of Artificial Intelligence, Zhongkai University of Agriculture and EngineeringSchool of Information Technology, Murdoch UniversityFaculty of Chemical and Biological Science, Department of Biotechnology, The Islamia University of BahawalpurDepartment of Medicinal Plants, Faculty of Agriculture and Natural Resources, Arak UniversityCollege of Artificial Intelligence, Zhongkai University of Agriculture and EngineeringCollege of Artificial Intelligence, Zhongkai University of Agriculture and EngineeringAbstract Dissolved oxygen (DO) is a vital parameter in regulating water quality and sustaining the health of aquatic organisms in aquaculture environments. Therefore, estimation and control of DO levels are essential in aquaculture operations. However, traditional chemical and photochemical approaches are limited by inaccuracies, environmental interferences, time consumption, and the inability to provide real-time data. Recently, artificial intelligence techniques have been studied for DO estimation. However, off-the-shelf models, such as Random Forest (RF) and Back Propagation (BP) have demonstrated poor performance due to intricate interactions in aquatic ecosystems, which leads to complex data patterns. This study proposes a water quality estimation model by combining a convolutional neural network (CNN), self-attention (SA), and bidirectional simple recurrent unit (BiSRU). One-dimensional convolution in CNN was employed to extract effective features and input into the SA mechanism to assign weights and emphasise crucial information. The model’s accuracy is improved by incorporating BiSRU. This model evaluated the DO levels of the intensive aquaculture base in Nansha, Guangzhou City, Guangdong Province, China. The proposed CNN-SA-BiSRU achieved MSE, MAE, RMSE, and R2 of 0.0022, 0.0341, 0.0471, and 0.9765, respectively. The results of the experiments showed that the proposed model had a high level of accuracy in estimating the outcomes with minimal fluctuations in estimation errors. Moreover, accuracy for short-term prediction was significantly improved, surpassing the performance of existing methods. The highly accurate results indicate the potential of the proposed methodology for DO-level monitoring in aquaculture and its usage in the fishery industry.https://doi.org/10.1038/s41598-025-10786-5Non-linearCNNBiSRUSelf-Attention mechanism
spellingShingle Longqin Xu
Wenjun Liu
Cai Chengqing
Tonglai Liu
Xuekai Gao
Ferdous Sohel
Murtaza Hasan
Mansour Ghorbanpour
Shahbaz Gul Hassan
Shuangyin Liu
Hybrid deep learning framework for real-time DO prediction in aquaculture
Scientific Reports
Non-linear
CNN
BiSRU
Self-Attention mechanism
title Hybrid deep learning framework for real-time DO prediction in aquaculture
title_full Hybrid deep learning framework for real-time DO prediction in aquaculture
title_fullStr Hybrid deep learning framework for real-time DO prediction in aquaculture
title_full_unstemmed Hybrid deep learning framework for real-time DO prediction in aquaculture
title_short Hybrid deep learning framework for real-time DO prediction in aquaculture
title_sort hybrid deep learning framework for real time do prediction in aquaculture
topic Non-linear
CNN
BiSRU
Self-Attention mechanism
url https://doi.org/10.1038/s41598-025-10786-5
work_keys_str_mv AT longqinxu hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT wenjunliu hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT caichengqing hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT tonglailiu hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT xuekaigao hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT ferdoussohel hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT murtazahasan hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT mansourghorbanpour hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT shahbazgulhassan hybriddeeplearningframeworkforrealtimedopredictioninaquaculture
AT shuangyinliu hybriddeeplearningframeworkforrealtimedopredictioninaquaculture