Acceleration of RANS Solver Convergence via Initialization with Wake Extension Models

TL;DR

This paper presents a neural network-based method to model wake development for RANS simulations, significantly accelerating convergence and reducing computational time by leveraging wake predictions for better initialization.

Contribution

A novel approach combining CNN-based wake modeling with domain segmentation to improve RANS solver convergence speed and efficiency.

Findings

26.3x fewer iterations needed for convergence

16.4x reduction in wall-clock simulation time

Effective acceleration with minimal additional data requirements

Abstract

Use of appropriate initialization to warm-start Reynolds-averaged Navier-Stokes (RANS) simulations of turbulent flow can facilitate convergence and lead to efficient use of computational resources. In this work, a method to model downstream wake development in external turbulent flow is proposed and used for RANS solver convergence acceleration. To balance the model accuracy and cost, the proposed method divides the analysis domain into three regions: near-body, wake and off-body. An approach based on a convolutional neural network is introduced as an efficient method to predict the downstream wake development. The model training only requires data from a single simulation, and its use is demonstrated to be effective in accelerating the RANS simulation when combined with an accurate flow prediction in the near-body region. The simulation using the proposed method took 26.3x fewer iterations, achieving 16.4x speedup in wall-clock time, compared to a baseline run using freestream initialization.