Hybrid Artificial Neural Network Activation Function to Reduce Water Wastage in Agricultural Irrigation

Most nations rely heavily on agriculture to provide employment. In recent years, as a result of a growing global population’s increased demand for food and water, numerous nations are consuming excessive amounts of already scarce freshwater resources. A large amount of water is used for irrigation. The purpose of this study is to find a way to reduce the amount of water lost during the irrigation process. The only practical solutions to the above problems are precision farming and smart irrigation. The study proposes a hybrid activation function based on the Artificial Neural Network algorithm. This function is used to classify the need for irrigation in various crops and predict the best time of the day for watering. The achievement of this is done by utilizing a hybrid activation function known as TanElu, which combines the Tanh and ELU functions. This function is used to classify the need for irrigation and predict the best time to water crops. The overall accuracy achieved by the proposed model was 98.24% and 97.31% in irrigation classification and prediction of the best irrigation time, respectively, which is a significant improvement compared to previous studies. Besides, execution time was 7.88 and 8.28 seconds in irrigation classification and prediction of best irrigation time respectively. The results indicate that the proposed hybrid activation function outperforms the single activation functions. Thus, the proposed model can significantly reduce water loss during the irrigation process and represents an encouraging step toward sustainable agriculture and water conservation.