Partially-algebraic two-fluid model and its application in sediment-laden flow computation for hydraulic machinery
摘要
水利水电工程中水力机械的过流泥沙往往是浓度高、粒径小、颗粒数目巨大的"细颗粒水沙两相流".双流体模型(Two-Fluid Model,TFM)能完备描述其每相的动力学特性,但联立方程数目多,限制了大规模工程计算的预测效率.目前有两种典型简化算法,一是基于水沙交融假设的混相流算法,二是基于被动扩散假设的拟单相流算法.然而,这两种方法仍难以兼顾计算精度、效率及鲁棒性,无法同时实现速度场的精确计算和浓度场的按需耦合.针对这一问题,本文提出了一种折中的局部代数化双流体模型(Partially-Algebraic Two-Fluid Model,PATFM)来综合两者的优势,其核心特征为:(1)基于达朗贝尔原理,利用伪瞬态代数表达式计算颗粒速度,避免求解沙相偏微分方程的高成本;(2)根据工程需要,利用附加对流-扩散方程实现速度场与浓度场的灵活耦合;(3)通过动量源项反馈相间作用对水流主相的影响.算例测试表明,PATFM能在基本准确预测水、沙每相流动特性的基础上显著降低计算成本,可以压缩70%以上的计算时长,从而为水力机械及其系统中细颗粒水沙两相流的大规模工程计算提供了一种有效的方案.
In water conservancy and hydropower engineering,the sediment-laden flow passing through hydraulic machinery often features high concentration,small particle size,and an enormous number of particles,constituting a"fine-particle water-sediment two-phase flow".The Two-Fluid Model(TFM)can fully describe the dynamic charac-teristics of each phase;however,the large number of coupled equations limits the predictive efficiency for large-scale engineering computations.Currently,two typical simplified approaches exist:one is the mixture model based on the water-sediment intermingling concept,and the other is the quasi-single-phase algorithm based on the passive diffusion assumption.However,both methods still struggle to balance computational accuracy,efficiency,and robustness,and cannot simultaneously achieve accurate velocity field calculations and on-demand coupling of the concentration field.To address this issue,this paper proposes a compromise approach—the Partially-Algebraic Two-Fluid Model(PATFM)—which integrates the advantages of both methods.Its core features are as follows:(1)Based on D'Alembert's principle,it employs a pseudo-transient algebraic expression to calculate the particle veloc-ity,avoiding the high cost of solving partial differential equations for the sediment phase;(2)According to engineer-ing needs,it uses an additional convection-diffusion equation to achieve flexible coupling between the velocity field and concentration field;(3)It feeds back the effect of interphase interactions on the primary water phase via a momentum source term.Case studies demonstrate that the PATFM can significantly reduce computational cost while maintaining essentially accurate predictions of the flow characteristics of both water and sediment phases,reducing computational time by over 70%.This new model provides an effective solution for large-scale engineering computa-tions of fine-particle sediment-laden flows in hydraulic machinery and their systems.