Dam water level prediction system utilizing Artificial Neural Network Back Propagation

Abstract

For flood and drought disasters, Reservoir dams are the one of the best protection mechanism. While in time of flood like conditions, the gate of the dam must be open sufficient to make sure that the reservoir potential will now not cross its limits and the discharges no longer reason for the overflow downstream. While at some point of drought the reservoir wants to impound water and launch properly to fulfill its purposes. Modeling of the reservoir water launch is fundamental to help the reservoir operator to make quickly and correct choices when dealing with each disasters. In this paper, shrewd selection help mannequin primarily based on Artificial Neural Network (ANN) Back Propagation is proposed. The proposed mannequin consists of state of affairs assessment, forecasting, and selection models. Situation evaluation utilized the temporal records mining approach to extract applicable statistics and attribute from the reservoir operation record. The forecasting mannequin makes use of ANN to operate forecasting of the reservoir water level, whilst in the choice model, ANN is utilized to operate the classification of the cutting-edge and adjustments of reservoir water level.