About 3D Incompressible Flow Reconstruction from 2D Flow Field Measurements

Area of Science:

• Fluid Dynamics
• Computational Mechanics
• Experimental Methods

Background:

• Industrial fluid flow analysis often requires complex and expensive 3D measurements.
• Existing methods may not fully capture the intricacies of 3D flow fields from limited data.
• There is a need for cost-effective and accurate methods to reconstruct 3D flow fields.

Purpose of the Study:

• To assess the uncertainty of a hybrid experimental/numerical procedure for 3D incompressible flow field reconstruction.
• To validate this hybrid method for potential industrial applications.
• To demonstrate the feasibility of using 2D measurements to predict the third velocity component.

Main Methods:

• A hybrid approach combining 2D experimental measurements with numerical computation using the continuity equation.
• Utilizing a NACA 0012 airfoil in a water channel as a quasi-3D test case.
• Employing a 3D Acoustic Doppler Velocimetry (ADV) instrument, considering only two components as measured and the third as a reference.
• Evaluating uncertainties according to the Guide to the Expression of Uncertainty in Measurement (GUM).
• Applying supervised learning classification and machine learning algorithms for data processing and comparison.

Main Results:

• The hybrid procedure accurately predicts the third velocity component by leveraging the continuity equation.
• Uncertainty analysis, following GUM, confirms the agreement between experimental data and numerical predictions.
• Comparison of flow indicators derived from experimental data and predictions shows strong correlation.
• Machine learning algorithms successfully processed sensed data into a 3D representation with accuracy and epoch metrics.

Conclusions:

• The validated hybrid method provides a reliable and cost-effective alternative to full 3D measurements for industrial flow analysis.
• The uncertainty assessment is crucial for establishing the method's validity and practical applicability.
• This approach facilitates easier design and testing of devices in industrial settings by simplifying flow field characterization.