Volume 117
您当前的位置:首页 > 期刊文章 > 过刊浏览 > Volumes 108-119 (2025) > Volume 117
Experimental investigation of suspension performance and chaotic characteristics of non-Newtonian liquid-solid mixing by the Paravisc impeller
Huibo Meng *, Jiayue Liu, Wenchun Jiang, Rupeng Zhu, Deao Li, Yingzheng Meng
College of New Energy, State Key Laboratory of Heavy Oil Processing, China University of Petroleum (East China), Qingdao, 266580, China
10.1016/j.partic.2026.07.017
Volume 117, October 2026, Pages 207-218
Received 7 May 2026, Revised 2 July 2026, Accepted 18 July 2026, Available online 29 July 2026, Version of Record 4 August 2026.
E-mail: huibomeng@upc.edu.cn

Highlights

• Visualization experiments on non-Newtonian liquid-solid suspensions were conducted.

• The complete off-bottom suspension condition was identified through visualization experiments.

• The nonlinear dynamic characteristics of liquid-solid suspension were revealed by LLE and MSE.

• The empirical correlation was established by a Python symbolic regression model.


Abstract

The Paravisc impeller is extensively employed in industrial mixing operations. However, its performance concerning particle suspension and chaotic mixing dynamics of non-Newtonian liquid-solid systems is inadequately characterized. This study systematically investigates the particle suspension state through visualization experiments in an elliptical-bottom stirred tank. Experiments are conducted at various rotational speeds ranging from 10 to 100 rpm and at solid hold-up values of 0%, 3%, 4%, and 5%. The critical flow regime transition from laminar to transitional flow is obtained when the Reynolds number is 78, approximately. Mixing performance is assessed by power consumption, mixing efficiency, and shear efficiency to elucidate the underlying flow structures and particle-induced energy dissipation mechanism. The results demonstrate that the particle promotes chaotic mixing when rotational speed is 60 rpm, which is smaller by 10 rpm than that in the single-phase. Furthermore, an empirical correlation among power, rotational speed, and solid hold-up is established based on the Python symbolic regression model, which demonstrates high predictive accuracy with the coefficient of determination is 0.9998, the mean absolute error is 0.0419, and the root mean square error is 0.0547. The robustness of the empirical correlation is validated across extended experiments at the rotational speeds ranging from 7 to 120 rpm. These findings provide technical guidance for optimizing mixing performance and improving energy utilization in non-Newtonian liquid-solid mixing systems agitated by the Paravisc impeller.

Graphical abstract
Keywords
Paravisc impeller; Non-Newtonian fluids; Liquid-solid suspension; Chaotic characteristics; PySR symbolic regression