Development of Unsteady-PSP Data Processing and Analysis Tools for the NASA Ames Unitary 11ft Wind Tunnel

Powell, Jessica M.; Murman, Scott M.; Ngo, Christina; Roozeboom, Nettie; Murakami, David D.; Baerny, Jennifer; Lie, Ji

doi:10.2514/6.2020-0292

Cited by 14 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our PSP data have unique characteristics, such as the images being black and white and having only one or two connected domains after binarization. On the basis of these features, we design the overlapping rate as the metric for computing registration accuracy: the overlapping rate = area off ∩area on_transfomred max (area off , area on_transfomred ) (6) where area off is the area of the model in the wind-off image and area on_transformed is the area of the model in the wind-on image after transformation using the estimated homography. An overlapping rate closer to 1 indicates more accurate estimated homography and thus higher registration accuracy.…”

Section: Evaluation Using Experimental Datamentioning

confidence: 99%

“…The watershed algorithm [5] works by identifying local maxima in the grayscale values of an image and connecting these points to form watershed lines, thereby segmenting the image into different regions. Powell and his colleagues [6] used a blob detector based on the watershed algorithm to identify markers on the surface of paint in the model beforehand. Through feature description and adopting a similarity metric, the correspondence between two images is then established.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A deep-learning image registration method for pressure-sensitive paint measurements

Di,

Gu,

Liu

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Image registration is an essential step in the data processing of pressure-sensitive paint (PSP) measurements. As PSP technology is gradually expanded to increasingly harsh environments, it faces challenges such as severe image misalignment due to model deformations/motions, which pose difficulties to traditional featurebased registration algorithms. To improve registration accuracy and efficiency, we propose an end-to-end image registration method based on deep learning. Initially, a PSP dataset based on PSP images is constructed through data augmentation. Three types of residual network and three training strategies are then adopted to prepare the deep-learning model for automatic image registration. The optimal combination of the residual network and training strategy is selected for validation using fabricated PSP images and outperforms two traditional algorithms (i.e., the Sift and Watershed methods). Finally, the performance of the deep-learning method is compared with that of traditional algorithms adopting a new metric of the overlapping rate for assessing the registration accuracy on experimental PSP images. The results show that the deep learning method outperforms the traditional algorithms in terms of registration accuracy and robustness.

show abstract

Section: Evaluation Using Experimental Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A deep-learning image registration method for pressure-sensitive paint measurements

Di,

Gu,

Liu

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Notable examples include the NASA Ames Unitary 11ft. Wind Tunnel experiment on the Space Launch System [7][8][9][10], hypersonic boundary layer separation studies conducted by Running et al on a conical structure resembling HiFire-I at the University of Notre Dame's pulsed arc-heated hypersonic wind tunnel (ACT-1) [11], and the surface pressure measurements of an elliptic cone body at Mach 7.5 in the Calspan University of Buffalo Research Center's hypersonic shock tunnel [12].…”

Section: Introductionmentioning

confidence: 99%

Fast-Responding Pressure-Sensitive Paint Measurements of the IC3X at Mach 7.2

Delgado Elizondo,

Dhanagopal,

Combs

2023

Aerospace

View full text Add to dashboard Cite

Global surface pressure measurements of a 5.7% scale AFRL Initial Concept 3.X vehicle (IC3X) were obtained using a fast-responding ruthenium-based pressure-sensitive paint (PSP) at the UTSA Mach 7 Ludwieg Tube Wind Tunnel at two different angles of attack, 0° and 2.5°. Static calibration of the paint was performed over a range of 0.386 kPa to 82.7 kPa to relate luminescent intensity to pressure. Details on the facility, paint preparation, application, calibration, and image processing techniques are provided in the manuscript. The results from statistical, spectral, and proper orthogonal decomposition (POD) analyses are presented to characterize the pressure field observed on the model. The experimental results qualitatively follow the expected trends and correspond to the occurrence of shock waves and expansion fans, which were visualized via Schlieren imaging. The theoretical pressure range obtained from conical shock analysis for 0° agrees with the experimentally derived pressure range for the model, and the outliers are attributed to errors in image registration. This study presents preliminary pressure measurements that pave the way for obtaining time-resolved global PSP measurements to train and validate aerothermodynamic machine learning models.

show abstract

Space Launch System Unsteady Forces Developed from Unsteady-Pressure-Sensitive-Paint–Based Corcos Model Parameters

Soranna,

Heaney,

Sekula

et al. 2024

Journal of Spacecraft and Rockets

View full text Add to dashboard Cite

During atmospheric ascent, launch vehicles (LVs) experience large dynamic loads at transonic conditions where aerodynamic buffet is most critical. To estimate buffet loads, coupled loads analyses typically utilize suitable forcing functions, called buffet forcing functions (BFFs). One of the key buffet environment contributors is the turbulent boundary layer (TBL) acting on the LV outer skin. The cross-spectral density function of TBL-induced fluctuating pressures can be estimated using the widely accepted Corcos model. In the context of transonic buffet, the performance of this model is not well established, partly because of lack of data. To fill this gap, NASA recently acquired extremely high-spatial-density data for the Space Launch System vehicle using the unsteady-pressure-sensitive-paint (uPSP) optical measurement technique. A methodology is developed for extraction of the Corcos model parameters using these unique data, with a focus on the LV design application. The model hypotheses are verified, and the model parameters are empirically tuned. For selected panels on the vehicle, force coherence factors are derived based on the Corcos model, and the associated panel BFFs are compared to uPSP data. It is shown that the modeled BFFs are in agreement with direct integration of uPSP data, except for regions where pressure fluctuations are spatially nonuniform. In those regions, the Corcos-based BFFs exhibit inherent limitations of BFF estimation methods that rely on discrete pressure measurements. Lastly, extending the present implementation of the Corcos model to frequencies impacted by vortex-shedding phenomena can result in underconservative BFF estimates at the subharmonic of the vortex shedding.

show abstract

Development of Unsteady-PSP Data Processing and Analysis Tools for the NASA Ames Unitary 11ft Wind Tunnel

Cited by 14 publications

References 7 publications

A deep-learning image registration method for pressure-sensitive paint measurements

A deep-learning image registration method for pressure-sensitive paint measurements

Fast-Responding Pressure-Sensitive Paint Measurements of the IC3X at Mach 7.2

Space Launch System Unsteady Forces Developed from Unsteady-Pressure-Sensitive-Paint–Based Corcos Model Parameters

Contact Info

Product

Resources

About