REAL-TIME PREDICTION OF DISSOLVED OXYGEN LEVELS IN AUTOMATED AQUACULTURE FEEDING SYSTEMS USING A COMBINED TIME SERIES DECOMPOSITION
Keywords:
time series decomposition, TCN, deep learning.Abstract
In the aquaculture industry, improper feeding can significantly decrease dissolved oxygen concentration, posing a high risk of mortality. To mitigate this risk, it is crucial to monitor safety in real time before feeding. This study proposes a novel method to predict dissolved oxygen concentration in real time, even when feeding occurs at arbitrary times. Our approach integrates a time series data decomposition model with a pre-trained deep learning model. The effectiveness of the proposed method was validated by comparing its predictive performance for both regular and irregular feeding schedules against other deep learning models.
References
1. Relationship between changes in marine environment (dissolved oxygen) and survival of aquatic organisms.
https://www.busan.go.kr/comm/getFile?srvcId=BBSTY1&upperNo=151614&fileTy=ATT ACH&fileNo=1
2. O’.I.Begimov, T.M.Bo‘riboyev / Extracting tagging from exocardiographic images via machine learning algorithmics // Analysis of world scientific views International Scientific Journal Vol 2 Issue 1 IF(Impact Factor)8.2 / 2023
3. O’.I.Begimov, T.M.Bo‘riboyev / General Theory About the Traditional Methods and Algorithms of Machine Learning // AMERICAN Journal of Public Diplomacy and International Studies Volume 02, Issue 04, 2024 ISSN (E):2993-2157.
