IoT Based Aquarium Water Quality Monitoring and Predictive Analytics Using Parameter Optimized Stack LSTM

Betta fish is the most common species used for ornamental purposes in captivity. There is an issue of maintaining the water quality to provide a good environment to the fish and thereby increasing the captive life span of the fish. An IoT based deep learning system is developed and an optimized LSTM based architecture is proposed to predict the quality of water stored in the tank. The parameters in the stack LSTM are altered and identified with the best variant fit enough to manage the quality of water. The data of pH sensor is subjected to stack LSTM with optimized parameters is identified and the best variants are chosen. The performance of best variants of pH sensor data are subjected with the series of other sensor data. The performance evaluation reports that the best performing variants of pH sensor data is ideal for predicting the quality of water.