Uncertainty-Aware Flow Field Reconstruction Using SVGP Kolmogorov-Arnold Networks
This addresses the problem of characterizing complex thermal-fluid systems for researchers and engineers, offering a robust, data-driven framework with practical guidance for experimental design in periodic flows, though it appears incremental as it extends classical methods.
The paper tackled reconstructing time-resolved flow fields from sparse velocimetry measurements by introducing an uncertainty-aware machine learning framework using SVGP-KAN, achieving reconstruction accuracy comparable to established methods while providing well-calibrated uncertainty estimates.
Reconstructing time-resolved flow fields from temporally sparse velocimetry measurements is critical for characterizing many complex thermal-fluid systems. We introduce a machine learning framework for uncertainty-aware flow reconstruction using sparse variational Gaussian processes in the Kolmogorov-Arnold network topology (SVGP-KAN). This approach extends the classical foundations of Linear Stochastic Estimation (LSE) and Spectral Analysis Modal Methods (SAMM) while enabling principled epistemic uncertainty quantification. We perform a systematic comparison of our framework with the classical reconstruction methods as well as Kalman filtering. Using synthetic data from pulsed impingement jet flows, we assess performance across fractional PIV sampling rates ranging from 0.5% to 10%. Evaluation metrics include reconstruction error, generalization gap, structure preservation, and uncertainty calibration. Our SVGP-KAN methods achieve reconstruction accuracy comparable to established methods, while also providing well-calibrated uncertainty estimates that reliably indicate when and where predictions degrade. The results demonstrate a robust, data-driven framework for flow field reconstruction with meaningful uncertainty quantification and offer practical guidance for experimental design in periodic flows.