Design Optimization of a Centrifugal Pump Using Particle Swarm Optimization Algorithm

Abstract

In the many industries centrifugal pumps consume a lot of energy. So the efficiency optimization of these pumps is so important. In this study shape of impeller of centrifugal pump was optimized to increase efficiency and head. The evolutionary algorithm based on modified artificial neural network (ANN), particle swarm optimization (PSO) and validated CFD data were used to optimized shape of centrifugal impeller. The pump experimentally investigated in test rig and measured data were used to verify the numerical results. Finally, to verify optimization results the complete numerical characteristic curves of the initial pump were compared with the optimized pump. The numerical results of optimized pump depict that efficiency improved 3.2% and head increase 5.52(m).