Abstract:Thermal striping is a phenomenon characterized by oscillatory mixing of non-isothermal streams, which is commonly seen in industrial processes such as nuclear coolant piping, petrochemical plants and liquefied natural gas transportation. The oscillatory mixing of hot and cold fluid can produce thermal field fluctuations and pose a potential risk of high-cycle thermal fatigue failures. Predicting and evaluating spatiotemporal fluctuations in thermal striping often requires high resolution and massive computational power. Although there have been extensive studies using machine learning algorithms on surrogate modeling, research focused on spatiotemporal fluctuation predictions is very limited. Due to the high dimensionality, it often requires complex algorithms with a large amount of high-fidelity training data, which limits the adoption of such methods for industrial applications. In this research, a two-level machine learning framework based on turbulence coherent structures is proposed and its application to a practical problem is demonstrated. The two-level design leverages vortex identification and local bias correction techniques, efficiently reducing the number of full-order simulations required for training. In the first level, well-organized coherent structures are extracted by performing Proper Orthogonal Decomposition on local parameters and then a tree-based machine-learning model is used to down-select the reference structures for the field reconstruction. In the second level, a parameterized convolution neural network is trained to predict the bias introduced by reference structures approximation. The demonstration of the methodology shows that the method can accurately capture the fluctuation frequencies and amplitudes of the spatiotemporal fields in a highly variational setting. Based on the vortex identification method, the methodology is expected to be applicable to general phenomenon driven by large coherent structures.

Recurrence Networks from Multivariate Signals for Uncovering Dynamic Transitions of Horizontal Oil-Water Stratified Flows

Multivariate weighted recurrence network inference for uncovering oil-water transitional flow behavior in a vertical pipe

Multiplex Multivariate Recurrence Network from Multi-Channel Signals for Revealing Oil-Water Spatial Flow Behavior

Multilayer Network from Multivariate Time Series for Characterizing Nonlinear Flow Behavior.

Visual identification of oscillatory two-phase flow with complex flow patterns

Multi-frequency complex network from time series for uncovering oil-water flow structure

Complex Network Analysis of Phase Dynamics Underlying Oil-Water Two-Phase Flows.

The Nonlinear Analysis of Horizontal Oil-Water Two-Phase Flow in a Small Diameter Pipe

Recurrence Network Analysis for Uncovering Dynamic Transition of Thermo-Acoustic Instability of Supercritical Hydrocarbon Fuel Flow

Three-phase Flow Displacement Dynamics and Haines Jumps in a Hydrophobic Porous Medium

Flow Sensing Method for Fluid-Structure Interaction Systems Via Multilayer Proper Orthogonal Decomposition

Insight into flow pattern evolution of vertical oil–water flows with large-diameter pipe

Testing for Nonlinearity in Dynamic Characteristics of Vertical Upward Oil-Gas-Water Three-phase Bubble and Slug Flows

Multifractal Analysis of Inclined Oil-Water Countercurrent Flow

Nonlinear feature analysis and automatic pattern recognition of oilgas water multiphase flow

A two-level machine learning approach for predicting thermal striping in T-junctions with upstream elbow

Multiscale Complex Network for Analyzing Experimental Multivariate Time Series

Identification and reconstruction of high-frequency broadband fluctuations in a turbulent flow exhibiting low-frequency deterministic motions

Data-driven modeling of unsteady flow based on deep operator network

Identifying dominant flow features from very-sparse Lagrangian data: a multiscale recurrence network-based approach

Multi-Scale Long-Range Magnitude and Sign Correlations in Vertical Upward Oil–Gas–Water Three-Phase Flow