Session: 05-12 Instrumentation II: Pressure Probes

Paper Number: 127876

127876 - A Numerical and Experimental Study of Five-Hole Probe Calibrations in Low-Speed Flows 

Pressure probes are amongst the most traditionally used instruments for measuring pressure and velocities in fluid flows. They are easy to manufacture and do not require drastic modifications to the machines to allow for access to the flowpath. Five-hole probes have five channels at the tip aligned with the flow, which are calibrated for various speeds and angles in both the pitch and yaw directions in order to generate a calibration map. It is well known that the probe calibration process requires tremendous effort and cost to use the measurement and instrumentation correctly, this study would be the solution for the chronic research problems. In the present study, the four non-dimensional pressure coefficients are used to present the flow characteristics in the pre-determined flow depending on the different angle positions. An existing calibration facility is automated for data acquisition. Not only the experiment facility is calibrated well in the physical structure, but also the data acquisition setting is determined based on sufficient empirical tests to guarantee the data reliability. Computational fluid dynamics (CFD) has been carried out to study two main fronts: First, for securing the stable flow to read the pressure around the small-scaled jet, an optimal distance between the jet nozzle and the probe has been found to place the probe into the most stable location in the jet core. Second, a numerical calibration map is generated by investigating the aerodynamics of the calibration jet and the internal channels of the probe. A commercial software (Star CCM+) is used to solve the governing equation and analyze the stability observing the entrained flow around the probe surface and low turbulence at the sharp edges of the probe head. The main characteristics of the CFD work are summarized in this paper. The experimental analyses of the hemispherical-head straight probe in pre-determined flows are performed to investigate the pressure distribution inside the channels, and the post-processing is done to analyze the data and visualization. The robustness of an experimental calibration map is examined by investigating the different pitch and yaw angle increments for the subsonic flow having the same mass flow rate. To test the experiment repeatability of a new probe calibration method, the root-mean-square-error (RMSE) is presented to show the deviation by replicating multiple experiments under identically controlled conditions. The quality of the numerical calibration map is assessed by comparing it to the experimental research findings, which apparently show the physical behavior of the real flow. All relevant information will be presented in detail in the final manuscript. It is expected that these results will bring great insight on three fronts: proposing a series of calibration methods for multi-hole probes to achieve accurate data where the probes measure the flows in turbomachinery applications; improving the computational models to have a good agreement with the real flow in a well-known automated calibration facility by studying numerical and experimental comparison; and paving the way to develop a physics-informed neural network-based Machine Learning model to reduce the probe calibration time by utilizing an extensive database is being collected from the experimental data.

Presenting Author: Dahae Jeong Pennsylvania State University

Presenting Author Biography: Dahae Jeong is a Ph.D. candidate in the Department of Mechanical Engineering at Pennsylvania State University. She earned her B.S. in Mechanical Engineering from Incheon National University and M.S. in Robotics and Virtual Engineering from the University of Science and Technology. Her research centers on enhancing sensors, measurements, and instrumentation. Her work involves in-depth exploration of data analysis and post-processing methods leveraging a high-fidelity Computational Fluid Dynamics (CFD) model, as well as the development and implementation of an automated experimental embedded system.

Authors:

Dahae Jeong Pennsylvania State University
Tamara Guimarães Pennsylvania State University