Graph neural network-accelerated Lagrangian fluid simulation

Nanofluids application in machining: a comprehensive review

Numerical simulations of non-Newtonian fluids are indispensable for optimization and monitoring of several industrial processes such as crude oil transportation, nuclear cooling, geothermal and fossil fuel production. The governing equations derived for non-Newtonian fluid models result in nonlinear differential equations. Thus, increasing the complexity even for simple geometries. The cumbersome numerical computation and rudimentary empirical solutions hinder faster analysis over a wide range of parameters. However, machine and deep learning methods have higher accuracy but rely heavily on the quality and amount of training data, and the solution may become inconclusive if data is sparse. In this research, a novel algorithm (Herschel Bulkley Network) is introduced to simulate the non-Newtonian fluid flow in a pipe using data redundant deep neural network (DNN) for fully developed, laminar, and incompressible flow conditions. The objective of this investigation is to develop a physics dominated DNN solely driven by minimizing residuals from the Navier-Stokes based governing equations, establishing benchmark research. Herschel-Bulkley model is used to approximate the complex rheological behavior of a non-Newtonian fluid. The proposed DNN algorithm is structured to incorporate initial/boundary conditions in cylindrical coordinates and approximate the solution without the aid of any simulated or training data. The simulated results and analysis demonstrate an excellent agreement between the proposed algorithm and non-Newtonian fluids flow attributes. The detailed parametric analysis exhibits the competency of the proposed algorithm to explain the rheological features. Monte-Carlo simulation is performed by propagating uncertainty to investigate the dominant parameters affecting simulated results. The uncertainty in fluid consistency index is responsible for higher variance in the calculated flow rate, while the least variation is observed due to fluid behavior index uncertainty. The performance of the algorithm is validated with experimental datasets. The statistical error estimation exhibits a mean absolute error of 11.5%, and root mean squared error of 0.87. A comprehensive analysis on training unsupervised DNN and adjusted hyperparameters is also highlighted to achieve expedite convergence.

Physics-guided deep neural network to characterize non-Newtonian fluid flow for optimal use of energy resources

Spatial Convolution Neural Network for Efficient Prediction of Aerodynamic Coefficients

We propose a novel turbulence refinement method based on the Rankine vortex model for SPH (smoothed particle hydrodynamics) simulations. Surface details are enhanced by recovering the energy lost in the rotational degrees of freedom of SPH particles. The Rankine vortex model is used to convert the diffused and stretched angular kinetic energy of particles to the linear kinetic energy of their neighbours. Our model naturally prevents the positive feedback effect between the velocity and vorticity fields since the vortex model is designed to alter the velocity without introducing external sources. Experimental results show that our method can recover missing high-frequency details realistically and maintain convergence in both static and highly dynamic scenarios.

/pdf/robust-turbulence-simulation-for-particle-based-fluids-using-xl2uwef8be.pdf

Robust turbulence simulation for particle-based fluids using the Rankine vortex model

Over the past few decades, fluid simulation has emerged as an important tool in computer animation. However, traditional physical-based fluid simulation systems are time consuming and requires large computational resources to generate large-scale fluid flows. Intelligent systems provide us a new method of data-driven to accelerate simulations. Previous intelligent system manually crafted feature vectors by using a context-based integral method and resulted in high computational and memory requirements. Unlike the existing techniques, we do not use any manually crafted feature, instead directly operate on Lagrangian fluid simulation data and use machine learned features. This paper presents a novel end-to-end deep learning neural network that can automatically generate a model for fluid animation based-on Lagrangian fluid simulation data. This approach synthesizes velocity fields with irregular Lagrangian data structure using neural network. Every fluid particle is treated independently and identically. We use symmetric functions to capture space structure and interactions among particles and design different network structures to learn various hierarchical features of fluid. We test this method using several data sets and applications in various scenes with different sizes. Our experiments show that the model is able to infer velocity field with realistic details such as splashes. In addition, compared with exiting simulation system, this method shows significant speed-ups, especially on large scene simulations.

FluidsNet: End-to-end learning for Lagrangian fluid simulation

Multiscale research of microstructure evolution during turning Ti-6Al-4V alloy based on FE and CA

Effect of workpiece microstructure on tool wear behavior and surface quality during machining Inconel 718 alloy

The infiltration and cooling effect of lubricant droplets in the cutting zone largely depend on setting minimum quantity lubricant (MQL) system parameters appropriately. The study determines the optimal spray angle and spray distance relying on the geometric position relationship between the nozzle and the overall cutting zone. A numerical model that combines computational fluid dynamics (CFD) and discrete phase model (DPM) is presented to simulate droplet transport process for dynamic milling and one-tooth milling by flow field distribution and droplet characteristics. The influences of different spray distances and spray angles on the infiltration process and cutting tool temperature are analyzed. The experiments aimed at verifying the results of the analysis are conducted by measuring cutting force, cutting temperature, surface roughness, surface microtopography and tool wear. The result indicates that the spray distance has a significant impact on the jet divergence, and insufficient lubrication or excessive lubrication is effectively avoided with the optimal cutting distance. Spraying in the direction of cutting-out promotes the accessibility of the lubricant and greater concentration of droplets in the cutting zone, especially later in the cutting process, which results in increased infiltration, reduced friction, and improved heat dissipation. Moreover, experiments demonstate that the optimal state of jetting in the direction of cutting-out is 46 °C lower than that of dry cutting, the surface roughness is reduced by 19.8%, and the wear amount is reduced by roughly 34%. This study has a better understanding of the setting of spray distance and spray angle in MQL milling and the development of equipment, which provides theoretical guidance for the optimization of MQL system parameters in practical applications. 

Feilong Du

Papers

FluidsNet: End-to-end learning for Lagrangian fluid simulation

Multiscale research of microstructure evolution during turning Ti-6Al-4V alloy based on FE and CA

Effect of workpiece microstructure on tool wear behavior and surface quality during machining Inconel 718 alloy

Analysis of droplet characteristics and cooling lubrication effects in MQL milling of 316L stainless steel

Experimental characterization of the performance of MQL-assisted turning of solution heat-treated and aged Inconel 718 alloy