Optimizing harbor seal whisker morphology for developing 3D-printed flow sensor

University of Groningen

Optimizing harbor seal whisker morphology for developing 3D-printed flow sensor

Zheng, Xingwen; Kamat, Amar M; Harish, Vinayak Sagar; Cao, Ming; Kottapalli, Ajay Giri

Prakash

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Final author's version (accepted by publisher, after peer review)

Publication date: 2021

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Zheng, X., Kamat, A. M., Harish, V. S., Cao, M., & Kottapalli, A. G. P. (Accepted/In press). Optimizing harbor seal whisker morphology for developing 3D-printed flow sensor. Paper presented at Transducers 2021, .

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

OPTIMIZING HARBOR SEAL WHISKER MORPHOLOGY FOR DEVELOPING

3D-PRINTED FLOW SENSOR

Xingwen Zheng

, Amar M. Kamat

, Vinayak Sagar Harish

Ming Cao

,

and Ajay Giri Prakash Kottapalli

1

_{Engineering and Technology Institute Groningen, Faculty of Science and Engineering,}

University of Groningen, Nijenborgh 4, 9747AG Groningen, The Netherlands and

2

MIT Sea Grant College Program, Massachusetts Institute of Technology, 77 Massachusetts

Avenue, Cambridge, MA 02139, USA

ABSTRACT

This paper presents a flow-structure interaction simulation-aided morphology optimization of harbor seal whisker, for generating a whisker-like structure that could possibly perform better in minimizing vortex-induced vibrations (VIVs) when subjected to steady flows. We also propose a whisker-inspired flow sensor design that features a 3D printed polymer model of the optimized seal whisker mounted on a 3D-printed cantilevered sensor base consisting of graphene nanoplatelets piezoresistors at the hinges. The designed flow sensor’s performance in sensing the flow velocity and its sensitivity to the external force are demonstrated by computational fluid dynamics simulations and proof-of-concept experiments, respectively.

KEYWORDS

Seal whisker, vortex-induced vibration, morphology optimization, flow sensor.

INTRODUCTION

Most mammals, including pinnipeds (seals, sea lions, and walruses), rats, cats, and otters, have whiskers. These whiskers act as sensors performing various sensing tasks ranging from flow sensing to tactile sensing and thereby generate situational awareness of the surroundings. Amongst other mammals with whiskers, pinnipeds possess the largest whiskers. Also, it is interesting to note that no other mammals other than certain seal species (grey seal, harbor seal, etc.) possess whiskers with undulations in their structure, which makes the whiskers look like beads on a string (Figure 1a) [1].

Harbor seals rely on their undulated whiskers to accurately track the hydrodynamic trail of fish in the water. Through live experiments involving harbor seals (Phoca

vitulina) tracking robotic fishes, Dehnhardt. et al.,

identified that harbor seals can detect the wake of a fish up to 35 seconds after the fish has passed (Figure 1b) [2]. The high sensitivity of the whiskers in tracking flows within the wake streets left by swimming fishes was attributed to the unique undulating morphological features on the whisker [3]. In the presence of steady flows, at higher velocities, a cylindrical structure would vibrate in the direction perpendicular to the oncoming flow direction due to the reaction of the shedding vortices, which create VIVs [4]. However, previous works have shown that whiskers’ unique undulating surface can suppress VIV (Figure 1b) [3]. Therefore, the whiskers rarely vibrate due to the seal’s own motion, allowing them to detect the fish’s eddy currents without self-generated noise caused due to VIVs.

Figure 1: Seals and their whisker sensors. (a) A harbor seal and its whiskers. (b) A diagrammatic sketch showing the fish tracking ability of the seal and the difference in the performance of cylindrical body and seal whisker in VIV.

In the recent years, a significant amount of research attention has been dedicated towards the development of artificial sensory analogs of the lateral-lines in fishes and seal whiskers for nature-inspired passive flow sensing on autonomous underwater vehicles (AUVs). While a significant number of fish neuromast inspired artificial flow sensors were developed, characterized and tested [5,6,7], research on artificial seal whisker inspired sensors, which utilize the specialized undulatory geometry of the whiskers is still nascent. Almost all of the bioinspired miniaturized flow sensors developed in the literature (including inspirations taken from sensory cilia in arthropods, mechanosensory hair cell sensors in fishes, and whiskers in seals) featured cylindrical hair-like structures that protrude into the flow, and interact with the flow through a drag force based bending mechanism that elicits a response in the sensing element at the base of the whisker [8]. Most MEMS (microelectromechanical systems) sensors developed earlier utilize doped silicon piezoresistors, strain gauges or piezoelectric sensing elements that were embedded into silicon or polymer structural materials designed as suspended membranes or cantilever beams [8]. In all these sensors, a hair cell or a whisker geometry was mimicked using a high aspect ratio cylindrical pillar that was developed either through lithographic microfabrication processes or 3D printing technologies. However, all these cylindrical-shaped hair structures utilized in the current flow sensors are susceptible to VIV induced flow noise. Therefore, investigating the seal whisker morphology and designing

seal whisker-inspired flow sensors may help overcome the existing disadvantages in current flow sensors. Though there are a few works conducted in the literature that have developed seal whisker-inspired sensors featuring undulatory whisker geometries [9,10,11], the seal whisker structures used in these sensors are usually constructed by a generalized geometrical framework proposed by Hanke

et al. [3]. There is very little research attention dedicated

towards investigating how modified whisker-like models would perform in VIV suppression and in developing flow sensor with the least possible self-generated noise caused due to VIVs.

Based on the above analyses, in this paper we 1) investigate the morphological parameters of harbor seal whisker, 2) construct 3D CAD (computer-aided design) models highlighting each morphological parameter, 3) compare to what extent the morphological parameters influence the VIV suppression through flow-structure interaction simulations, 4) generate the optimized seal whisker model with the best VIV suppression ability, 5) and finally design a seal whisker inspired flow sensor and demonstrate its flow sensing ability.

THE GEOMETRICAL FRAMEWORK OF

HARBOR SEAL WHISKER

To obtain a novel geometrical framework of harbor seal whisker, we scanned five real harbor seal whiskers using computed tomography (CT) scanning technology (GOM ATOS III triple scanner) and thus obtained the 3D CAD models (Figure 2a) of the harbor seal whisker. The seal whisker has ellipse-like cross-sections that can be fit using a standard ellipse. Then we calculated five parameters of each fitted ellipse on each cross-section from the bottom to the tip of the seal whisker (Figure 2b) in SolidWorks software. The parameters include the major axis (a), minor axis (b), the rotational angle (θ), the X coordinate, and the Y coordinate of the centroids (x, y). Finally we obtained Equations (1)-(5) for describing the geometrical framework of harbor seal whisker. Equations (1)-(4) are used to fit a, b, x, and y with the cross-section number n, while the general regularity of θ on the nth _{cross-section} was characterized by the mean of the five values (θi,n, i=1,

2, ..., 5) of θ obtained from five harbor seal whiskers. The distance between the nth _{and (n+1)}th _{cross-section is 0.5mm.} The Z coordinate (z) in Figure 2b can be described using the parameter n, and z = 0.5*n.

a(n)=0.14*sin(0.92*n+1.5 )+1.0 (1)

b(n)=0.067*sin(0.91*n+1.0 )-0.0041*n+0.64 (2)

x(n)=0.0001*n2_{+0.0109*n-0.0414 (3)}

y(n)=-0.0009*n2_{+0.0109*n-0.0230 (4)}

θ(n)=(θ1,n+θ2,n+θ3,n+θ4,n+θ5,n)/5 (5)

Based on the investigated regularities of a, b, θ, x, and

y, a general seal whisker model (that is, base model in

Figure 3a) can be constructed by SolidWorks software. To explore the influence of the morphological parameters in suppressing VIVs, we also constructed other modified whisker-like models, as shown in Figure 3. For models in Figure 3b-3d, only the parameter θ, x, y, and a are defined using the geometrical model. For the model in Figure 3e,

θ=0, while x, y, a, and b are defined using the geometrical

model. For models in Figure 3f-3g, x or y is constant and

equals the coordinates of the centroid on the base of the seal whisker, while θ, a, and b are defined using the geometrical model. All of the whisker models have a length of 25 mm, consisting of 50 cross-sections.

Figure 2: Seal whisker morphology. (a) A CT scanned seal whisker. (b) The elliptical cross-sections with morphology parameters (a, b, θ, x, y).

Figure 3: Various seal whisker models. Top: The view along the minor of seal whisker. Bottom: The view that is perpendicular to the base of seal whisker. (a) Base model. (b) a=2b. (c) a=3b. (d) a=4b. (e) θ=0. (f) Constant x and y. (g) Constant x. (h) Constant y. (i) Optimized seal whisker.

OPTIMIZATION OF SEAL

WHISKER-LIKE STRUCTURE

Based on the constructed models, we conducted flow-structure interaction simulations using COMSOL Multiphysics for comparing the VIVs of the above-constructed models. As shown in Figure 4, the whisker models are located in an oncoming flow with a velocity v, and the bottom of the whisker models are fixed. The seal whisker models are located with their major axis paralleling the flow direction. For the calculation domain, free tetrahedral mesh was used for the space within which the seal whisker model lies, while a free triangular mesh with a sweep option was used for other domains.

Figure 4: Dimensions and mesh of the calculation domain in simulations in which the seal whiskers are situated in an oncoming flow. (a) Dimensions. (b) Mesh.

By comparing the VIVs of the whisker tip for various models (Figure 5), it can be concluded that: 1) the model in which a=4b (Figure 5d) vibrates much smaller than the base model (Figure 5a) and the models in which a=2b (Figure 5b) or a=3b (Figure 5c) ; 2) the model in which

θ=0 (Figure 5e) has smaller VIVs, compared with the base

model; (3) the model with constant x and y (Figure 5f) vibrates much smaller than the model with constant x (Figure 5g). Considering the above points, we optimized the seal whisker by making θ=0 and a=4b. Though the model with constant x and y (Figure 5f) vibrates a bit more than the model with constant y (Figure 5h), we still made the (x,y) constant and finally obtained the optimized seal whisker (Figure 3i) that has no noticeable vibrations (Figure 5i).

Figure 5: VIVs of various seal whiskers. (a) Base model

(v=1 m/s). (b) a=2b (v=1 m/s). (c) a=3b (v=1 m/s). (d)

a=4b (v=1 m/s). (e) θ=0 (v=1 m/s). (f) Constant x and y

(v=0.4 m/s). (g) Constant x (v=0.4 m/s). (h) Constant y (v=0.4 m/s). (i) Optimized seal whisker (v=1 m/s).

Figure 6: The conceptual design of seal whisker inspired 3D printed flow sensor. (a) The design of the flow sensor with cantilevered structure and graphene nanoplatelet piezoresistors. (b) The section view of the sensor. (c) Dimensions of graphene piezoresistor. (d) A The 3D printed seal whisker inspired sensor. (e) Side view.

3D PRINTED HARBOR SEAL WHISKER

INSPIRED FLOW SENSOR

A conceptual design of a harbor seal whisker inspired 3D printed flow sensor with the optimized seal whisker model and a 3D-printed sensing base is shown in Figure 6.

The entire sensor was fabricated by 3D printing using the photopolymer resin (Model number: FLGPCL04) of the Formlabs (Form 3) 3D printer. The resin has a Young’s modulus of 2.8E9 Pa, a Poisson’s ratio of 0.35, and a density of 1200 kg/m3_{. The sensor structure features the} optimized 3D printed whisker-like structure situated at the center of a cantilevered structure. Piezoresistive sensing elements are designed at the four hinges (numbered 1-4) of the cantilevers where maximal strain occurs in the presence of the whisker deflection. The piezoresistive sensing elements are formed through drop-casting graphene nanoplatelets into the serpentine shaped micro-grooves 3D printed on the surface of the cantilever hinges. Considering that the optimized seal whisker-like structure has no noticeable VIV when the flow is along the major axis, the flow sensor could then reduce self-generated noise and be rendered more sensitive to steady flow sensing. Specifically, the strain induced on the cantilevers caused by the deflection of whisker in the presence of oncoming flow can be used to quantify the frequency and amplitude of the vibration of the seal whisker. To corelate the oncoming flow velocity with respect to the strain induced on the graphene piezoresistors, we conducted COMSOL Multiphysics based simulations in which the seal whisker inspired flow sensor was situated in an oncoming flow (Figure 7a). A free triangular mesh was used for the whole calculation domain (Figure 7b).

Figure 7: (a) Dimensions and (b) mesh of the calculation domain in simulations for the designed flow sensor.

Figure 8: The strain on the cantilever and the deformation of seal whisker in an oncoming flow with various flow velocities. The flow velocity is (a) 0.02 m/s, (b) 0.04 m/s, (c) 0.06 m/s, (d) 0.08 m/s, (e) 0.1 m/s, (f) 0.2 m/s, (g) 0.3 m/s, (h) 0.4 m/s, and (i) 0.5 m/s. The colour on the flow sensor indicates the strain value.

Figure 8 shows the strain on the cantilever and the deformation of the seal whisker in an incoming flow with a flow velocity varying from 0.02 m/s to 0.5 m/s. Figure 9

shows the average strain of the serpentine shaped micro-grooves at Hinge 1 and 3 (Figure 6) with the flow velocity. The absolute strain increases with the flow velocity. Besides, the regularity between the flow velocity and the difference between the strain of the two parts can be fitted well using a quadratic function. In addition to testing the performance of the sensor in sensing the flow velocity through simulations, the sensitivity of the sensor to an external force stimuli was preliminarily tested through a simple experiment involving a physical displacement of the whisker along the minor axis of the whisker (Figure 10). The sensor responded well to the external force, characterized by the voltage peaks corresponding to each act of displacement of the whisker through a tactile stimuli.

Figure 9: The average strain of the micro-grooves at Hinge 1 and 3 with various flow velocities.

Figure 10: The output voltage of the 3D printed whisker sensor in response to tactile stimuli. The seal whisker structure of the sensor was touched by the external force at 10 s, 20 s, 30 s, 40 s, and 50 s.

CONCLUSION AND FUTURE WORK

A novel geometrical framework of harbor seal whisker was proposed. Various modified seal whisker-like structures highlighting each morphological parameter were constructed. Flow-structure interaction simulations were conducted to compare the influence of various morphological parameters in VIV suppression. An optimized seal whisker-like structure without noticeable VIVs was constructed. A seal whisker inspired flow velocity sensor was designed, and its sensitivity to flow velocity and external force was explored. In the future work, flow velocity calibration of the 3D printed flow sensor will be conducted together with the experimental quantification of suppression in VIV and the resultant enhancement in the sensitivity of the flow sensor.

ACKNOWLEDGEMENTS

This research was supported in part by the University of Groningen’s start-up grant awarded to Ajay Kottapalli, in part by ITEA (grant agreement ITEA-2018-17030-Daytime), in part by the European Research Council

(ERC-CoG-771697), and in part by the Netherlands Organization for Scientific Research (NWO-vidi-14134).

REFERENCES

[1] Ginter, Carly C., Thomas J. DeWitt, Frank E. Fish, and Christopher D. Marshall, “Fused traditional and geometric morphometrics demonstrate pinniped whisker diversity,” PloS one, Vol. 7, no. 4, pp. e34481, 2012.

[2] Dehnhardt, Guido, Björn Mauck, Wolf Hanke, and Horst Bleckmann, “Hydrodynamic trail-following in harbor seals (Phoca vitulina),” Science, vol. 293, no. 5527, pp. 102-104, 2001.

[3] Hanke, Wolf, Matthias Witte, Lars Miersch, Martin Brede, Johannes Oeffner, Mark Michael, Frederike Hanke, Alfred Leder, and Guido Dehnhardt, “Harbor seal vibrissa morphology suppresses vortex-induced vibrations,” Journal of Experimental Biology, vol. 213, no. 15, pp. 2665-2672, 2010.

[4] Luo, Gang, Xiao Jun Zhou, and Shuang Wang, “Numerical Simulation of Vortex-Induced Vibration of Submerged Floating Tunnel Cable,” Applied Mechanics and Materials, vol. 256, pp. 1352-1358, 2013.

[5] Yang, Yingchen, Jack Chen, Jonathan Engel, Saunvit Pandya, Nannan Chen, Craig Tucker, Sheryl Coombs, Douglas L. Jones, and Chang Liu, “Distant touch hydrodynamic imaging with an artificial lateral line,” Proceedings of the National Academy of Sciences, vol. 103, no. 50, pp. 18891-18895, 2006.

[6] Kottapalli, Ajay Giri Prakash, et al, “Touch at a distance sensing: lateral-line inspired MEMS flow sensors,” Bioinspiration & biomimetics, vol. 9, pp. 046011, 2014.

[7] Abdulsadda, Ahmad T., and Xiaobo Tan, “Nonlinear estimation-based dipole source localization for artificial lateral line systems,” Bioinspiration & biomimetics, vol. 8, pp. 026005, 2013.

[8] Triantafyllou, Michael S., Gabriel D. Weymouth, and Jianmin Miao, “Biomimetic survival hydrodynamics and flow sensing,” Annual Review of Fluid Mechanics, vol. 48, pp. 1-24, 2016.

[9] Kottapalli, A. G. P., My Asadnia, H. Hans, J. M. Miao, and M. S. Triantafyllou, “Harbor seal inspired MEMS artificial micro-whisker sensor,” in IEEE International Conference on Micro Electro Mechanical Systems (MEMS), pp. 741-744, 2014.

[10] Kottapalli, A. G. P., Mohsen Asadnia, J. M. Miao, and M. S. Triantafyllou, “Harbor seal whisker inspired flow sensors to reduce vortex-induced vibrations,” in IEEE International Conference on Micro Electro Mechanical Systems (MEMS), pp. 889-892, 2015. [11] Beem, Heather R., and Michael S. Triantafyllou,

“Wake-induced ‘slaloming’ response explains exquisite sensitivity of seal whisker-like sensors,” J. Fluid Mech, vol. 783, pp. 306-322, 2015.

CONTACT

*Ajay Kottapalli, tel: +31 50 36 35486; a.g.p.kottapalli@rug.nl