EngineBench: Flow Reconstruction in the Transparent Combustion Chamber III Optical Engine

TL;DR

EngineBench provides a high-quality, real-world PIV dataset for turbulent flow analysis in combustion engines, introducing a novel inpainting task with random edge gaps to improve ML-based flow reconstruction.

Contribution

This paper introduces EngineBench, a unique dataset for ML in fluid mechanics, and proposes a new inpainting method using random edge gaps for turbulent flow data reconstruction.

Findings

UNet models outperform industry-standard approaches in inpainting accuracy.

The dataset enables development of general-purpose ML models for engine flow analysis.

The inpainting method effectively handles occlusions of various sizes and orientations.

Abstract

We present EngineBench, the first machine learning (ML) oriented database to use high quality experimental data for the study of turbulent flows inside combustion machinery. Prior datasets for ML in fluid mechanics are synthetic or use overly simplistic geometries. EngineBench is comprised of real-world particle image velocimetry (PIV) data that captures the turbulent airflow patterns in a specially-designed optical engine. However, in PIV data from internal flows, such as from engines, it is often challenging to achieve a full field of view and large occlusions can be present. In order to design optimal combustion systems, insight into the turbulent flows in these obscured areas is needed, which can be provided via inpainting models. Here we propose a novel inpainting task using random edge gaps, a technique that emphasises realism by introducing occlusions at random sizes and orientations at the edges of the PIV images. We test five ML methods on random edge gaps using pixel-wise, vector-based, and multi-scale performance metrics. We find that UNet-based models are more accurate than the industry-norm non-parametric approach and the context encoder at this task on both small and large gap sizes. The dataset and inpainting task presented in this paper support the development of more general-purpose pre-trained ML models for engine design problems. The method comparisons allow for more informed selection of ML models for problems in experimental flow diagnostics. All data and code are publicly available at https://eng.ox.ac.uk/tpsrg/research/enginebench/.