This content is not included in your SAE MOBILUS subscription, or you are not logged in.
Fuel Tank Dynamic Strain Measurement Using Computer Vision Analysis
ISSN: 0148-7191, e-ISSN: 2688-3627
Published April 14, 2020 by SAE International in United States
This content contains downloadable datasetsAnnotation ability available
Stress and strain measurement of high density polyethylene (HDPE) fuel tanks under dynamic loading is challenging. Motion tracking combined with computer vision was employed to evaluate the strain in an HDPE fuel tank being dynamically loaded with a crash pulse. Traditional testing methods such as strain gages are limited to the small strain elastic region and HDPE testing may exceed the range of the strain gage. In addition, strain gages are limited to a localized area and are not able to measure the deformation and strain across a discontinuity such as a pinch seam. Other methods such as shape tape may not have the response time needed for a dynamic event. Motion tracking data analysis was performed by tracking the motion of specified points on a fuel tank during a dynamic test. An HDPE fuel tank was mounted to a vehicle section and a sled test was performed using a Seattle sled to simulate a high deltaV crash. Multiple target markers were placed on the fuel tank. The motion of these markers was captured using high speed video cameras. The high speed videos were processed using the OpenCV computer vision library. Using OpenCV, the high speed videos were imported, and the position of the central location of each target marker was extracted frame by frame from the high speed videos. Once the position was known, the strain was computed using the change in relative position between two marker positions. Results of the testing showed that the acceleration-induced strain is low, generally less than the material yield strain. It was noted that reliable and accurate results require that the camera be placed normal to, or at a shallow angle to, the points being tracked. In addition, curved surfaces lead to limited fidelity of strain data due to the varying focal length of the points being tracked and measurement increased sensitivity. This method is similar to a “typical” tensile test in which displacement is tracked between two pre-established points on a sample. As such, the methodology was replicated on a tensile specimen to validate the methodology.
CitationFleming, M., Krishnaswami, R., and Nakamoto, K., "Fuel Tank Dynamic Strain Measurement Using Computer Vision Analysis," SAE Technical Paper 2020-01-0924, 2020, https://doi.org/10.4271/2020-01-0924.
Data Sets - Support Documents
|[Unnamed Dataset 1]|
- Pan, B., Qian, K., Xie, H., and Asundi, A. , “Two-Dimensional Digital Image Correlation for in-Plane Displacement and Strain Measurement: A Review,” Meas. Sci. Technol. 20(6):1-17, April 2009.
- Pan, B. and Li, K. , “A Fast Digital Image Correlation Method for Deformation Measurement,” Optical Lasers in Engineering 49(7):841-847, March 2011.
- Zhu, C., Wang, H., Kaufmann, K., and Vecchio, K. , “Computer Vision Approach to Study Deformation of Materials”, arXiv.org, April 2019, http://arxiv.org/abs/1904.03321.
- Cao, G., Han, G., Liu, W., and Zhao, F. , “A Study on Analysis Method of Motion Characteristics in the Crash Test Based on Computer Vision,” in Proceedings of the FISITA 2012 World Automotive Congress, Lecture Notes in Electrical Engineering 196, November 2012, doi:10.1007/978-3-642-33738-3_65.
- Measurand, Inc. , “Product Catalogue - ShapeMRI: Motion Capture for MRI,” n.d., https://www.measurand.com.
- OpenCV , https://opencv.org.
- Lyvers, E.P., Mitchell, O.R., Akey, M.L., and Reeves, A.P. , “Subpixel Measurements Using a Moment-Based Edge Operator,” IEEE Trans. on Pattern Analysis and Machine Intelligence 11(12):1293-1309, December 1989.