Upstream flow geometries can be uniquely learnt from single-point turbulence signatures

TL;DR

This study demonstrates that the upstream flow geometry can be precisely identified from single-point turbulence signatures using machine learning, revealing unique flow features linked to orifice shape with high accuracy.

Contribution

The paper introduces a novel method employing random forest classifiers on turbulence invariants to accurately determine upstream obstacle shapes from downstream flow data.

Findings

Random forest models achieved 100% accuracy in classifying orifice shapes.

Temporal autocorrelation coefficients are highly sensitive to flow geometry.

Flow physics analysis links turbulence invariants to flow structure and shape identification.

Abstract

We test the hypothesis that the microscopic temporal structure of near-field turbulence downstream of a sudden contraction contains geometry-identifiable information pertaining to the shape of the upstream obstruction. We measure a set of spatially sparse velocity time-series data downstream of differently-shaped orifices. We then train random forest multiclass classifier models on a vector of invariants derived from this time-series. We test the above hypothesis with 25 somewhat similar orifice shapes to push the model to its extreme limits. Remarkably, the algorithm was able to identify the orifice shape with 100% accuracy and 100% precision. This outcome is enabled by the uniqueness in the downstream temporal evolution of turbulence structures in the flow past orifices, combined with the random forests' ability to learn subtle yet discerning features in the turbulence microstructure. We are also able to explain the underlying flow physics that enables such classification by listing the invariant measures in the order of increasing information entropy. We show that the temporal autocorrelation coefficients of the time-series are most sensitive to orifice shape and are therefore informative. The ability to identify changes in system geometry without the need for physical disassembly offers tremendous potential for flow control and system identification. Furthermore, the proposed approach could potentially have significant applications in other unrelated fields as well, by deploying the core methodology of training random forest classifiers on vectors of invariant measures obtained from time-series data.