Feature engineering and symbolic regression methods for detecting hidden physics from sparse sensors

Abstract

In this study we put forth a modular approach for distilling hidden flow physics in discrete and sparse observations. To address functional expressiblity, a key limitation of the black-box machine learning methods, we have exploited the use of symbolic regression as a principle for identifying relations and operators that are related to the underlying processes. This approach combines evolutionary computation with feature engineering to provide a tool to discover hidden parameterizations embedded in the trajectory of fluid flows in the Eulerian frame of reference. Our approach in this study mainly involves gene expression programming (GEP) and sequential threshold ridge regression (STRidge) algorithms. We demonstrate our results in three different applications: (i) equation discovery, (ii) truncation error analysis, and (iii) hidden physics discovery, for which we include both predicting unknown source terms from a set of sparse observations and discovering subgrid scale closure models. We illustrate that both GEP and STRidge algorithms are able to distill the Smagorinsky model from an array of tailored features in solving the Kraichnan turbulence problem. Our results demonstrate the huge potential of these techniques in complex physics problems, and reveal the importance of feature selection and feature engineering in model discovery approaches.

References in corpus (5)

• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Closed-loop separation control using machine learning
• High-order methods for decaying two-dimensional homogeneous isotropic turbulence
• Elite Bases Regression: A Real-time Algorithm for Symbolic Regression
• Feedback Control of Turbulent Shear Flows by Genetic Programming

Cited by in corpus (19)

• Discovering Symbolic Models from Deep Learning with Inductive Biases
• On closures for reduced order models $-$ A spectrum of first-principle to machine-learned avenues
• Physics-guided deep learning framework for predictive modeling of the Reynolds stress anisotropy
• A computational framework for physics-informed symbolic regression with straightforward integration of domain knowledge
• Digital Twins in Wind Energy: Emerging Technologies and Industry-Informed Future Directions
• Evolving symbolic density functionals
• Machine learning constraints on deviations from general relativity from the large scale structure of the Universe
• A Framework Based on Symbolic Regression Coupled with eXtended Physics-Informed Neural Networks for Gray-Box Learning of Equations of Motion from Data
• Deep neural network enabled corrective source term approach to hybrid analysis and modeling
• Discovering explicit Reynolds-averaged turbulence closures for turbulent separated flows through deep learning-based symbolic regression with non-linear corrections
• Discovery of interpretable structural model errors by combining Bayesian sparse regression and data assimilation: A chaotic Kuramoto-Sivashinsky test case
• Symbolic Regression as Feature Engineering Method for Machine and Deep Learning Regression Tasks
• Iterative Symbolic Regression for Learning Transport Equations
• Machine learning cosmic inflation
• Complementary consistency test of the Copernican principle via Noether's theorem and machine learning forecasts
• Information Fusion via Symbolic Regression: A Tutorial in the Context of Human Health
• Explainable Artificial Intelligence for Exhaust Gas Temperature of Turbofan Engines
• Robust physics discovery via supervised and unsupervised pattern recognition using the Euler characteristic
• Optimal Prediction using Learning and Shape Optimization