In the final stage of production, External Gear Pumps (EGPs) must be run-in following a specific working cycle in which the pump body is subject to material removal resulting in a consequent increase in energy demand. The run-in cycle impacts the pump performance and the production costs. In this study, five new run-in cycles are defined with the objective to analyse the impact on the power absorbed during and after the cycle, the material removed from the body and the impact on pump performance. The work shows that the use of different run-in cycles can result in a higher pump volumetric efficiency, up to + 2%, combined with a cost reduction of 1.74% and an increase of productivity of 78%. Subsequently, experimental data were used to define a preliminary AI based algorithm to predict the run-in cycles parameter. The AI algorithm, which was trained and validated on the experimental data, demonstrated a high predictive capability with a maximum error of 6%.
Volumetric Efficiency and Productivity of External Gear Pumps through Optimized Run-In Strategies
Alfonso Rosario Apuzzo
;Emma Frosina;
In corso di stampa
Abstract
In the final stage of production, External Gear Pumps (EGPs) must be run-in following a specific working cycle in which the pump body is subject to material removal resulting in a consequent increase in energy demand. The run-in cycle impacts the pump performance and the production costs. In this study, five new run-in cycles are defined with the objective to analyse the impact on the power absorbed during and after the cycle, the material removed from the body and the impact on pump performance. The work shows that the use of different run-in cycles can result in a higher pump volumetric efficiency, up to + 2%, combined with a cost reduction of 1.74% and an increase of productivity of 78%. Subsequently, experimental data were used to define a preliminary AI based algorithm to predict the run-in cycles parameter. The AI algorithm, which was trained and validated on the experimental data, demonstrated a high predictive capability with a maximum error of 6%.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


