LGA-NET: A LSTM-GRU-ATTENTION LAYER HYBRID NETWORK FOR WATER QUALITY ASSESSMENT IN SMART AND SUSTAINABLE AQUACULTURE

Abstract

Water quality management in aquaculture is important as it directly impacts aquatic animals and plant's health, growth, and productivity. Management of water conditions at their optimum promotes innovative and sustainable aquaculture production at maximum yield. Traditional approaches to water quality measurement are by intermittent sampling, which only gives a snapshot of environmental conditions at a particular moment. The approaches are slow, labor-consuming, and incapable of feeding back real-time data on key parameters like pH, dissolved oxygen, and temperature. As a result, critical water quality issues may go undetected until they reach harmful levels, leading to significant losses in aquaculture systems. This study explores using deep learning models to improve the accuracy and efficiency of water quality forecasting. Leveraging large datasets and advanced neural network architectures, the proposed LGA-Net: A LSTM-GRU-Attention Layer Hybrid Network for Water Quality Assessment in Smart and Sustainable Aquaculture aims to provide real-time, data-driven intelligence for proactive decision-making in aquaculture management. The model is educated on past water quality data and identifies complex patterns and relationships between parameters to produce accurate forecasts. In contrast to conventional monitoring, real-time forecasting based on deep learning avoids sudden environmental shifts, allowing real-time response and risk minimization. The findings of this study highlight the capability of deep learning to transform the process of innovative and sustainable water quality management. With the integration of artificial intelligence into aquaculture systems, experts can optimize environmental conditions, reduce the need for reactive measures, and optimize overall sustainability. This study highlights the ability of technological innovation to ensure robust and efficient aquaculture operations, contributing to food security globally and smart and sustainable fisheries management.