Computer vision-based recognition of rainwater rivulet morphology evolution during rain–wind-induced vibration of a 3D aeroelastic stay cable. Issue 172 (January 2018)

• Record Type: Journal Article
• Authors: Cheng, Peng; Li, Wen-Jie; Chen, Wen-Li; Gao, Dong-Lai; Xu, Yang; Li, Hui
• Is Part Of: Journal of wind engineering and industrial aerodynamics. Issue 172(2017)
• Page Start: 367
• Page End: 378
• Publication Date: 2018-01
• Subjects: Stay cable -- Rain–wind-induced vibration -- Rainwater rivulet -- Computer vision -- Image recognition; Wind-pressure -- Periodicals; Buildings -- Aerodynamics -- Periodicals; Pression du vent -- Périodiques; Constructions -- Aérodynamique -- Périodiques; Buildings -- Aerodynamics; Wind-pressure; Periodicals
• DOI: 10.1016/j.jweia.2017.11.017
• Languages: English
• ISSNs: 0167-6105

Abstract

Abstract: In this study, the morphology evolution of rainwater rivulets during the rain–wind-induced vibration (RWIV) of a three-dimensional (3D) aeroelastic stay cable model is identified using computer-vision technology. The formation and oscillation of rainwater rivulets on the cable surface are critical to the occurrence of the RWIV of the cable model. However, it is very difficult to obtain the water film and rainwater rivulets on the cable surface because no sensors can be installed on the cable surface. The computer vision technology can obtain information about the surface of an object. In this study, a wind tunnel test is performed to reproduce the RWIV along with rainwater rivulets and periodic oscillations along the cable circumference. To obtain the geometric and dynamic characteristics of the rainwater rivulets, a high-speed charge-coupled device (CCD) camera with a resolution of 1200 × 800 pixels and an acquisition frequency of 200 fps is employed to record the snapshots of the rivulet measurement regions. Then, an image recognition method based on comparison of color differences and least-square fitting techniques is employed to identify the rainwater rivulet morphology evolution on the cable surface. The rivulet width, film endpoints, and rivulet distribution characteristics in the spanwise direction are obtained; the oscillation frequency is subsequently calculated according to the rivulet time evolutions. Finally, the relationship between the oscillating characteristics of the rivulet and cable vibration is investigated; this relationship may help reveal the possible mechanics of the RWIV of stay cables. Highlights: A computer vision-based recognition method is proposed to identify the rainwater rivulet. Rivulet morphology evolutions are investigated during RWIV of a 3D aeroelastic stay cable model. Rivulet morphology switches between dispersion and convergence configurations during RWIV. The rivulet movement with consistent dominant frequencies presents a phase delay as compared to the cable vibrations.