Prediction of centrifugal slurry pump head reduction: An artificial neural networks approach

The feasibility of using artificial neural networks (ANN) in the prediction of head reduction of centrifugal pumps handling slurries is examined. An ANN model is proposed and compared with the empirical correlation given by the present authors earlier. The comparison showed that the ANN could successfully be used for the prediction of head reductions of centrifugal slurry pumps. The mean deviation between predicted and experimental values is 5.86% which is reasonable for slurry handling processes.