An IoT framework for quality analysis of aquatic water data using time-series convolutional neural network

An IoT-based water contamination analysis for aquaculture using lightweight multi-headed GRU model

Maintaining the quality of water is essential for the health and productivity of aquatic organisms, including fish in aquaculture ponds. However, water contamination can severely impact fish health and survival, making it necessary to develop monitoring systems that can detect early signs of water contamination. Initial deep learning models had limitations in capturing the temporal and spatial dependencies of time-series data, which can lead to inaccurate predictions. In this paper, we propose a smart monitoring system that uses IoT devices to collect water quality data and segment it into contaminated and non-contaminated categories based on a water toxic index (WTI), a measure of water contamination levels. To address the limitations of early deep learning models for classification of toxic and non-toxic water quality, an enhanced light-weight multi-headed gated recurrent unit (MHGRU) model that captures the spatial and temporal dependencies of water quality parameters. Our study demonstrates that the proposed model outperforms existing models, achieving an impressive accuracy of 99.7% when evaluated on real-time data. Notably, our model also excels when tested on a public dataset, achieving an accuracy of 99.12%. In comparison, best performed existing ANN models achieve accuracies of 99.52% and 98.71% on the respective datasets.

Deep reinforcement learning for forecasting fish survival in open aquaculture ecosystem

Ensuring the classification of water bodies suitable for fish habitat is essential for animal preservation and commercial fish farming. However, existing supervised machine learning models for predicting water quality lack specificity regarding fish survival. This study addresses this limitation and presents a novel model for forecasting fish viability in open aquaculture ecosystems. The proposed model combines reinforcement learning through Q-learning and deep feed-forward neural networks, enabling it to capture intricate patterns and relationships in complex aquatic environments. Moreover, the model's reinforcement learning capability reduces the reliance on labeled data and offers potential for continuous improvement over time. By accurately classifying water bodies based on fish suitability, the proposed model provides valuable insights for sustainable aquaculture management and environmental conservation. Experimental results show a significantly improved accuracy of 96% for the proposed DQN-based model, outperforming existing Gaussian Naive Bayes (78%), Random Forest (86%), and K-Nearest Neighbors (92%) classifiers on the same dataset. These findings highlight the effectiveness of the proposed approach in forecasting fish viability and its potential to address the limitations of existing models.