Shape optimization of a centrifugal blood pump by coupling CFD with metamodel-assisted genetic algorithm
- 256 Downloads
A centrifugal blood pump is a common type of pump used as a left ventricular assist device in the medical industries. Therefore, the improvement of the device bio-compatibility to reduce the blood damage and to increase the efficiency has become a major challenge. In the current work, a metamodel-assisted genetic algorithm is employed to simultaneously optimize the impeller and volute geometries of a typical centrifugal blood pump. The overall shape of the base design is inspired from HeartMate3 LVAD, and the main dimensions of the base design including inlet and outlet radius, blade angle distribution, volute cross-section area distribution, etc., are designed in our laboratory. Three different scenarios are investigated using three different objective functions, i.e., (1) hydraulic efficiency, (2) pressure head, and (3) hemolysis index (HI). The results showed that the shape optimized by pump efficiency has also nearly the same level of HI as the shape optimized by HI. Hence, to reduce computation time, one can use efficiency instead of HI as an objective function. However, one must check the HI level after such optimization to see whether it is within the acceptable range of HI for such bio application.
KeywordsHemolysis Centrifugal blood pump Optimization Metamodel Genetic algorithm
This research is sponsored by the Iran National Science Foundation (INSF) with the Project No. of 95837323.
Compliance with ethical standards
Conflict of interest
The authors declared that there are no conflicts of interest.
- 3.Bartoli CR, Kang J, Zhang D, Howard J, Acker M, Atluri P, et al. Left ventricular assist device design reduces von Willebrand factor degradation: a comparative study between the HeartMate II and the EVAHEART left ventricular assist system. Ann Thorac Surg. 2017;103:1239–44.CrossRefPubMedGoogle Scholar
- 9.Nourbakhsh SA, Jaumotte BA, Hirsch C, Parizi HB, editors. Turbopumps and Pumping Systems. Berlin: Springer; 2007.Google Scholar
- 11.Kyo S, editor. Ventricular assist devices in advanced-stage heart failure. Tokyo: Springer; 2014.Google Scholar
- 15.Haykin S. Neural networks: a comprehensive foundation. 2nd ed. London: Pearson Education Inc.; 1999.Google Scholar
- 16.Koza JR. Genetic programming: on the programming of computers by means of natural selection. Cambridge: Massachusetts Institute of Technology Press; 1992.Google Scholar