Work

Research on internal flow mechanism and energy loss characteristics of centrifugal pump based on data regression algorithm

Year: 2025

Type: article

Abstract: The eddy viscosity model (EVM) is a widely used and cost-effective technique for predicting turbulence in the context of solving the Reynolds-averaged Navier-Stokes (RANS) equations. Despite its great practicality, this approach has drawbacks. The inability of the SST k-ω model to adequately capture the rotational motion and curvature variations of streamlines is a major flaw. By contrasting the numerical simulation results of the original SST model, the updated SST model with constant eddy viscosity coefficient(SST-MS), and utilizing the SST turbulence model in conjunction with a data regression algorithm and a user-defined function (UDF) to modified the Spalart-Shur rotation-curvature and correction (SST-CC). This paper reveals the fundamental laws of turbulence model. According to the computation results, the relative errors for power, efficiency, and pump head are 1.33%, 1.82%, and 1.02%, respectively. Under the design conditions, the SST-CC model's accuracy is greater than the SST model's. This work also focuses on the relationship between turbulent kinetic energy, helicity, vorticity, and entropy generation in order to better define the internal flow conditions, particularly in the complicated and fluctuating flow within the centrifugal pump. The distribution of vortex structures and the kinetic energy dissipation process are easier to understand with the help of the updated model. The improved model enhances the interpretability of vortex structure distribution and kinetic energy dissipation processes. Consequently, this research illustrates the effectiveness of rotation and curvature compensation.

Source: Results in Engineering

DOI: 10.1016/j.rineng.2025.107434

Authors: Tao Xu, Baoling Cui, Lulu Zhai, zuchao Zhu

Institutions: Zhejiang Sci-Tech University, Zhejiang Industry Polytechnic College

Language: English

Topic: Turbomachinery Performance and Optimization

Subfield: Aerospace Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $1,940

Funder: National Natural Science Foundation of China