Optimization of Pump Electric Drives Using Artificial Neural Networks: A Predictive Control Approach

Abstract

This paper presents a comprehensive approach to optimizing the efficiency of pump electric drives using Artificial Neural Networks (ANN). The study utilized experimental data from various operational scenarios of centrifugal pumps, including power consumption and energy usage, to train the ANN model. Applying the ANN-based predictive control, the system could forecast optimal operating conditions, significantly reducing energy consumption and improving overall system performance. The experimental setup involved both individual and multi-pump systems, where power and energy consumption were monitored and analyzed. The results demonstrate that ANN-based control can effectively optimize pump operations, achieving substantial energy savings while ensuring system reliability. The implementation of this approach led to energy savings of over 15% compared to traditional control methods. This study highlights the potential of ANN in enhancing the energy efficiency of industrial pumping systems and provides insights into future applications of intelligent control in resource management.