Performance Prediction and Design Method for Centrifugal Pump Based on Computational Fluid Dynamics

Using Reynolds average N-S equations closed by standard k- turbulence model, the steady and unsteady turbulent flow in centrifugal pump was simulated by using MRF model and SM model respectively. A method for predicting the performances of centrifugal pump was built on the basis of computational fluid dynamics. By the presented performance prediction method, not only the flow characteristics of centrifugal pump can be obtained, but also can its performances be evaluated quantitatively and qualitatively. The advantages and disadvantages of some traditional design methods for centrifugal pump were analyzed. On the basis of performance prediction and flow analysis, a new design method was put forward in consideration of the steady and unsteady performances of centrifugal pump. The proposed method can be used to design the centrifugal pump with high running stability, efficiency and cavitation resistance, and it is available for shortening development period and improving design quality of pump.

Download Full-text

Computed tomography angiography as an adjunct to computational fluid dynamics for prediction of oxygenator thrombus formation

Perfusion ◽

10.1177/0267659120944105 ◽

2020 ◽

pp. 026765912094410

Author(s):

Robert G Conway ◽

Jiafeng Zhang ◽

Jean Jeudy ◽

Charles Evans ◽

Tieluo Li ◽

...

Keyword(s):

Fluid Dynamics ◽

Computed Tomography ◽

Computational Fluid Dynamics ◽

Computed Tomography Angiography ◽

Thrombus Formation ◽

Flow Analysis ◽

Flow Characteristics ◽

Region Of Interest ◽

Computational Fluid Dynamics Analysis ◽

Clot Burden

Introduction: Extracorporeal membrane oxygenation circuit performance can be compromised by oxygenator thrombosis. Stagnant blood flow in the oxygenator can increase the risk of thrombus formation. To minimize thrombogenic potential, computational fluid dynamics is frequently applied for identification of stagnant flow conditions. We investigate the use of computed tomography angiography to identify flow patterns associated with thrombus formation. Methods: A computed tomography angiography was performed on a Quadrox D oxygenator, and video densitometric parameters associated with flow stagnation were measured from the acquired videos. Computational fluid dynamics analysis of the same oxygenator was performed to establish computational fluid dynamics–based flow characteristics. Forty-one Quadrox D oxygenators were sectioned following completion of clinical use. Section images were analyzed with software to determine oxygenator clot burden. Linear regression was used to correlate clot burden to computed tomography angiography and computational fluid dynamics–based flow characteristics. Results: Clot burden from the explanted oxygenators demonstrated a well-defined pattern, with the largest clot burden at the corner opposite the blood inlet and outlet. The regression model predicted clot burden by region of interest as a function of time to first opacification on computed tomography angiography (R2 = 0.55). The explanted oxygenator clot burden map agreed well with the computed tomography angiography predicted clot burden map. The computational fluid dynamics parameter of residence time, when summed in the Z-direction, was partially predictive of clot burden (R2 = 0.35). Conclusion: In the studied oxygenator, clot burden follows a pattern consistent with clinical observations. Computed tomography angiography–based flow analysis provides a useful adjunct to computational fluid dynamics–based flow analysis in understanding oxygenator thrombus formation.

Download Full-text