A Critical Review of Physical Models in High Temperature Multiphase Fluid Dynamics: Turbulent Transport and Particle-Wall Interactions

Abstract

This review article examines the last decade of studies investigating solid, molten and liquid particle interactions with one another and with walls in heterogeneous multiphase flows. Such flows are experienced in state-of-the-art and future-concept gas turbine engines, where particles from the environment, including volcanic ash, runway debris, dust clouds, and sand, are transported by a fluid carrier phase and undergo high-speed collisions with high-temperature engine components. Sand or volcanic ash ingestion in gas turbine engines is known to lead to power-loss and/or complete engine failure. The particle-wall interactions that occur in high temperature sections of an engine involve physics and intrinsic conditions that are sufficiently complex that they result in highly disparate and transient outcomes. These particles, which often times are made up of glassy constituents called CMAS (calcium-magnesium-alumino-silicate), are susceptible to phase change at combustor temperatures (1650?), and can deposit on surfaces, undergo elastic and plastic deformation, rebound, and undergo breakup. Considerable research has been put into developing empirical and physics-based models and numerical strategies to address phase interactions. This article provides a detailed account of the conceptual foundation of physics-based models employed to understand the behavior of particle-wall interaction, the evolution of numerical methods utilized for modeling these interactions, and challenges associated with improving models of particle-particle and particle-wall interactions needed to better characterize multiphase flows. It also includes description of a testbed for acquiring canonical data for model validation studies.

Characterization of partial wetting by CMAS droplets using multiphase many-body dissipative particle dynamics and data-driven discovery based on PINNs

“…As the CMAS droplet spreads on the substrate, it loses its initial spherical shape and begins to wet the surface as depicted in figure 3, forming a liquid film between the droplet and the substrate. Understanding how droplets behave on surfaces is important for a wide range of applications, including in industrial processes, microfluidics, propulsion materials, and the design of self-cleaning surfaces (Pitois & François 1999;Chen et al 2016;Hassan et al 2019;Jain et al 2021;Nieto et al 2021).…”

“…Physics-informed neural networks form another promising class of approaches that leverages the flexibility and scalability of deep neural networks to discover governing equations, incorporating physical laws (PDEs, ODEs, integro-differential equations (IDEs), etc.) and data (initial conditions plus boundary conditions) into the loss function (Raissi et al 2019;Chen et al 2020;Mao, Jagtap & Karniadakis 2020;Mishra & Molinaro 2020;Shukla et al 2020;Karniadakis et al 2021). In contrast, EQL and SINDy focus on deriving equations from data without enforcing the governing laws, lacking a guarantee of adherence to these principles.…”

“…Spreading of liquids on solid surfaces is a classic problem (de Gennes 1985; Bonn et al 2009). Although the theoretical foundations were laid by Young and Laplace already in the early 1800s (de Laplace 1805; Young 1805), there are still many open questions, and it remains a highly active research field especially in the context of microfluidics (Nishimoto & Bhushan 2013) as well as in the design of propulsion materials (Jain et al 2021). As discussed in detail in the review of Popescu et al (2012), there are two fundamentally different cases: non-equilibrium spreading of the droplet, and the case of thermodynamic equilibrium when spreading has ceased and the system has reached its equilibrium state.…”

“…The ability of PINNs to discover equations makes them promising for applications in scientific discovery, engineering design and data-driven modelling of complex physical systems (Karniadakis et al 2021). Their integration of physics-based constraints into the learning process enhances their capacity to generalize and capture the underlying physics accurately.…”

Particle Deposition Characteristics on Turbine Blade Surface Based on Critical Velocity Model

“…Particle deposition on turbine blade surfaces is a complex process influenced by various factors such as collision surface properties, temperature, particle physical properties, particle diameter, impact velocity, and impact angle [6]. Currently, two widely recognized particle deposition models are the critical viscosity model [7] and the critical velocity model [8].…”