Optimization of a centrifugal blood pump in terms of hemolysis index and hydraulic efficiency

Abstract

In the development of blood pumps, clinical and hydraulic performance requirements must be met. Optimization studies addressing biocompatibility and efficiency issues in the design of centrifugal blood pumps are increasing rapidly. This study aims to increase hydraulic efficiency and decrease the hemolysis index which is an indicator of the increase in the amount of free hemoglobin in blood plasma by optimizing the centrifugal blood pump. A centrifugal blood pump, whose dimensions were calculated using classical formulas, was optimized with the genetic algorithm by changing seven dimensions that were likely to have a significant impact on efficiency and hemolysis index. These dimensions are blade inlet angle, blade outlet angle, blade inlet height, blade outlet height, the gap between the blade tip and volute, the gap between disk and volute, and the gap between shroud and volute. By determining the lower and upper limits of these seven dimensions, 183 different pump geometries were generated, and computational fluid dynamics simulations were performed. Using simulation results, two adaptive neuro-fuzzy inference systems for the hemolysis index and hydraulic efficiency were generated. Using these models, optimization was made with the genetic algorithm. The optimum pump found by the genetic algorithm was simulated and compared with the base pump. The results showed that there is generally a direct relationship between hydraulic efficiency and hemolysis index. It was observed that the hemolysis index of the optimum pump decreased from 2.55E−05 of the base pump to 2.45E−05, while the hydraulic efficiency increased from 42.24 to 45.92%.