Ventricular assist devices or total artificial hearts can be used to save patients with heart failure when there are no donors available for heart transplantation. Blood pumps are integral parts of such devices, but traditional axial flow blood pumps have several shortcomings. In particular, they cause hemolysis and thrombosis due to the mechanical contact and wear of the bearings, and they cause blood stagnation due to the separation of the front and rear guide wheel hubs and the impeller hub. By contrast, the implantable axial flow, maglev blood pump has the characteristics of no mechanical contact, no lubrication, low temperature rise, low hemolysis, and less thrombosis. Extensive studies of axial flow, maglev blood pumps have shown that these pumps can function in laminar flow, transitional flow, and turbulent flow, and the working state and performance of such pumps are determined by their support mechanisms and flow channel. Computational fluid dynamics (CFD) is an effective tool for understanding the physical and mechanical characteristics of the blood pump by accurately and effectively revealing the internal flow field, pressure–flow curve, and shear force distribution of the blood pump. In this study, magnetic levitation supports were used to reduce damages to the blood and increase the service life of the blood pump, and a conical impeller hub was used to reduce the speed, volume, and power consumption of the blood pump, thereby facilitating implantation. CFD numerical simulation was then carried out to optimize the structural parameters of the conical axial maglev blood pump, predict the hemolysis performance of the blood pump, and match the flow channel and impeller structure. An extracorporeal circulation simulation platform was designed to test whether the hydraulic characteristics of the blood pump met the physiological requirements. The results showed that the total pressure distribution in the blood pump was reasonable after optimization, with a uniform pressure gradient, and the hemolysis performance was improved.

For a blood pump, the blood flow channel and impeller parameters directly affect the performance of the pump and the resulting blood circulation. The flow channel in particular has a great impact on the hydraulic performance of the pump (e.g., flow and pressure), which directly determines the overall performance of the blood pump. Traditional bearing-supported blood pumps can cause mechanical damage to blood cells, leading to hemolysis and thrombosis. In this study, therefore, we designed a conical spiral axial blood pump with magnetic levitation. The blood pump was supported by electrodynamic bearings in the radial direction and electromagnetic bearings in the axial direction. The impeller and the front and rear hubs were integrated to minimize blood stagnation and reduce the formation of thrombosis. The hub had a conical spiral flow channel design, which not only reduced the size of the impeller but also increased blood flow and pressure while meeting the design requirements. Computational fluid dynamics (CFD) analysis was used to analyze the flow field of the axial blood pump, a power function model was used to establish a hemolysis prediction model, and the particle tracking method was used to obtain the flow trajectories of individual blood cells, thereby predicting hemolysis-related performance of the blood pump. The simulation results showed that the main high shear stress area in the blood pump was located in the impeller inlet and the clearance between the top of the impeller and the inner chamber of the blood pump. When the hub taper angle of the blood pump was 0.72° and the clearance was 0.3 mm, the average hemolysis prediction value was 0.00216. This prediction value was smaller than that of traditional axial blood pumps. These findings can provide an important reference for the structural design of axial blood pumps and for reducing the hemolysis prediction value.