Water Level Prediction of Community Secondary Water Supply Tank Based on Deep Learning

Abstract

Compared to the LSTM model, the GRU model can be used to reduce significantly the computational time and the model size without losing the accuracy. For the sake of reduction risk of the insufficient water supply in the community, this paper establishes a GRU neural network model to predict the water level of the water supply tank. This GRU model solves the short-term memory problem in the recurrent neural network , at the same time, in order to speed up the training of the model, the GRU model is trained by the characteristics of the computational cluster based on the distributed framework. The GRU model is validated by a simulated water supply tank system in the laboratory, the example calculation results show that the GRU neural network model has good feasibility and accuracy in the prediction of the secondary water supply tank level in the community water supply system.