Optimal Processing Parameters of Transmission Parts of a Flapping-Wing Micro-Aerial Vehicle Using Precision Injection Molding
Abstract
:1. Introduction
2. Experimental Methods
3. Results and Discussion
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Ashley, S. Palm-size spy planes. Mech. Eng. 1998, 120, 74–78. [Google Scholar] [CrossRef] [Green Version]
- Pornsin-Sirirak, T.N.; Tai, Y.C.; Nassef, H.; Ho, C.M. Titanium-alloy MEMS wing technology for a micro aerial vehicle application. Sens. Actuators A Phys. 2001, 89, 95–103. [Google Scholar] [CrossRef]
- Yang, L.J.; Hsua, C.K.; Hoa, J.Y.; Feng, C.K. Flapping wings with PVDF sensors to modify the aerodynamic forces of a micro aerial vehicle. Sens. Actuators A: Phys. 2007, 139, 95–103. [Google Scholar] [CrossRef]
- Ramasamy, M.; Leishman, J.G.; Lee, T.E. Flow field of a rotating-wing micro air vehicle. J. Aircr. 2007, 44, 1236–1244. [Google Scholar] [CrossRef]
- Zhang, C.; Rossi, C. A review of compliant transmission mechanisms for bio-inspired flapping-wing micro air vehicles. Bioinspiration Biomim. 2017, 12, 025005. [Google Scholar] [CrossRef] [PubMed]
- Sai, K.P.P.M.; Bharadwaj, K.; Teja, K.R.; Dagamoori, K.; Tarun, K.V.S.; Vijayan, V. Design, fabrication and testing of flapping wing micro air vehicle. Int. J. Eng. Res. Appl. 2016, 6, 133–150. [Google Scholar]
- Deng, S.; Oudheusden, B.V. Wake structure visualization of a flapping-wing Micro-Air-Vehicle in forward flight. Aerosp. Sci. Technol. 2016, 50, 204–211. [Google Scholar] [CrossRef]
- Phan, H.V.; Park, H.C. Generation of Control Moments in an Insect-like Tailless Flapping-wing Micro Air Vehicle by Changing the Stroke-plane Angle. J. Bionic Eng. 2016, 13, 449–457. [Google Scholar] [CrossRef]
- Phan, H.V.; Au, T.K.L.; Park, H.C. Clap-and-fling mechanism in a hovering insect-like two-winged flapping-wing micro air vehicle. R. Soc. Open Sci. 2016, 3, 160746. [Google Scholar] [CrossRef] [Green Version]
- Phan, H.V.; Kang, T.; Park, H.C. Design and stable flight of a 21 g insect-like tailless flapping wing micro air vehicle with angular rates feedback control. Bioinspiration Biomim. 2017, 12, 036006. [Google Scholar] [CrossRef]
- Van Truong, T.; Kureemun, U.; Tan, V.B.C.; Lee, H.P. Study on the structural optimization of a flapping wing micro air vehicle. Struct. Multidiscip. Optim. 2017, 57, 653–664. [Google Scholar] [CrossRef]
- Badrya, C.; Sridharan, A.; Baeder, J.D.; Kroninger, C.M. Multi-Fidelity Coupled Trim Analysis of a Flapping-Wing Micro Air Vehicle Flight. J. Aircr. 2017, 54, 1614–1630. [Google Scholar] [CrossRef]
- Hassanalian, M.; Abdelkefi, A.; Wei, M.; Ziaei-Rad, S. A novel methodology for wing sizing of bio-inspired flapping wing micro air vehicles: Theory and prototype. Acta Mech. 2016, 228, 1097–1113. [Google Scholar] [CrossRef]
- Nguyen, Q.V.; Chan, W.L.; Debiasi, M. Experimental investigation of wing flexibility on force generation of a hovering flap-ping wing micro air vehicle with double wing clap-and-fling effects. Int. J. Micro Air Vehicles 2017, 9, 187–197. [Google Scholar] [CrossRef]
- Bluman, J.E.; Kang, C.-K.; Shtessel, Y. Control of a Flapping-Wing Micro Air Vehicle: Sliding-Mode Approach. J. Guid. Control. Dyn. 2018, 41, 1223–1226. [Google Scholar] [CrossRef]
- Lu, Z.; Debiasi, M.; Nguyen, Q.V.; Cha, W.L. Bio-inspired low-noise wing design for a two-winged flapping wing micro air vehicle. AIAA J. 2018, 56, 4697–4705. [Google Scholar] [CrossRef]
- Del Estal Herrero, A.; Percin, M.; Karasek, M.; Van Oudheusden, B. Flow Visualization around a Flapping-Wing Micro Air Vehicle in Free Flight Using Large-Scale PIV. Aerospace 2018, 5, 99. [Google Scholar] [CrossRef] [Green Version]
- Yang, W.; Wang, L.; Song, B. Dove: A biomimetic flapping-wing micro air vehicle. Int. J. Micro Air Veh. 2017, 10, 70–84. [Google Scholar] [CrossRef]
- Nan, Y.; Peng, B.; Chen, Y.; Feng, Z.; McGlinchey, D. Can Scalable Design of Wings for Flapping Wing Micro Air Vehicle Be Inspired by Natural Flyers? Int. J. Aerosp. Eng. 2018, 1–24. [Google Scholar] [CrossRef]
- Nguyen, A.T.; Han, J.-H. Wing flexibility effects on the flight performance of an insect-like flapping-wing micro-air vehicle. Aerosp. Sci. Technol. 2018, 79, 468–481. [Google Scholar] [CrossRef]
- Jankauski, M.; Guo, Z.; Shen, I. The effect of structural deformation on flapping wing energetics. J. Sound Vib. 2018, 429, 176–192. [Google Scholar] [CrossRef]
- Badrya, C.; Govindarajan, B.; Baeder, J.D.; Harrington, A.; Kroninger, C.M. Computational and Experimental Investigation of a Flapping-Wing Micro Air Vehicle in Hover. J. Aircr. 2019, 56, 1610–1625. [Google Scholar] [CrossRef]
- Cao, C.; Burgess, S.; Conn, A.T. Toward a Dielectric Elastomer Resonator Driven Flapping Wing Micro Air Vehicle. Front. Robot. AI 2019, 5. [Google Scholar] [CrossRef] [PubMed]
- Gallar, B.M.; Van Oudheusden, B.W.; Sciacchitano, A.; Karásek, M. Large-scale volumetric flow visualization of the unsteady wake of a flapping-wing micro air vehicle. Exp. Fluids 2020, 61, 16. [Google Scholar] [CrossRef] [Green Version]
- Lane, P.; Throneberry, G.; Fernandez, I.; Hassanalian, M.; Vasconcellos, R.; Abdelkefi, A. Towards Bio-Inspiration, Development, and Manufacturing of a Flapping-Wing Micro Air Vehicle. Drones 2020, 4, 39. [Google Scholar] [CrossRef]
- Lee, J.G.; Ryu, S.G.; Kim, H.J. Stable flight of a flapping-wing micro air vehicle under wind disturbance. In Proceedings of the 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Las Vegas, NV, USA, 25–29 October 2020. [Google Scholar]
- Dong, X.; Li, D.; Xiang, J.; Wang, Z. Design and experimental study of a new flapping wing rotor micro aerial vehicle. Chin. J. Aeronaut. 2020, 33, 3092–3099. [Google Scholar] [CrossRef]
- Yoon, S.-H.; Cho, H.; Lee, J.; Kim, C.; Shin, S.-J. Effects of camber angle on aerodynamic performance of flapping-wing micro air vehicle. J. Fluids Struct. 2020, 97, 103101. [Google Scholar] [CrossRef]
- Nguyen, K.; Au, L.T.K.; Phan, H.-V.; Park, S.H.; Park, H.C. Effects of wing kinematics, corrugation, and clap-and-fling on aerodynamic efficiency of a hovering insect-inspired flapping-wing micro air vehicle. Aerosp. Sci. Technol. 2021, 118, 106990. [Google Scholar] [CrossRef]
- Wang, S.; Song, B.; Chen, A.; Fu, Q.; Cui, J. Modeling and flapping vibration suppression of a novel tailless flapping wing micro air vehicle. Chin. J. Aeronaut. 2022, 35, 309–328. [Google Scholar] [CrossRef]
- Su, Y.-C.; Shah, J.; Lin, L. Implementation and analysis of polymeric microstructure replication by micro injection molding. J. Micromech. Microeng. 2003, 14, 415–422. [Google Scholar] [CrossRef]
- Shen, Y.K. Comparison of height replication properties of micro-injection moulding and micro-injection–compression moulding for production of microstructures on lightguiding plates. Plast. Rubber Compos. 2007, 36, 77–84. [Google Scholar] [CrossRef]
- Hill, S.D.J. LIGA News Third Issue; 1995. [Google Scholar]
- Wan, A.W.A.R.; Lee, T.S.; Rahmat, A.R. Injection moulding simulation analysis of natural fiber composite window frame. J. Mater. Process. Technol. 2008, 197, 22. [Google Scholar]
- Jansen, K.M.B.; Van Dijk, D.J.; Husselman, M.H. Effect of processing conditions on shrinkage in injection molding. Polym. Eng. Sci. 1998, 38, 838–846. [Google Scholar] [CrossRef] [Green Version]
- Gao, Y.; Turng, L.-S.; Wang, X. Process optimization and effects of material properties on numerical prediction of warpage for injection molding. Adv. Polym. Technol. 2008, 27, 199–216. [Google Scholar] [CrossRef]
- Hakimian, E.; Sulong, A.B. Analysis of warpage and shrinkage properties of injection-molded micro gears polymer composites using numerical simulations assisted by the Taguchi method. Mater. Des. 2012, 42, 62–71. [Google Scholar] [CrossRef]
- Oktem, H.; Erzurumlu, T.; Uzman, I. Application of Taguchi optimization technique in determining plastic injection molding process parameters for a thin-shell part. Mater. Des. 2007, 28, 1271–1278. [Google Scholar] [CrossRef]
- Su, T.-L.; Chen, H.-W.; Lu, C.-F. Systematic optimization for the evaluation of the microinjection molding parameters of light guide plate with TOPSIS-based Taguchi method. Adv. Polym. Technol. 2010, 29, 54–63. [Google Scholar] [CrossRef]
- Chiang, Y.C.; Cheng, H.C.; Huang, C.F.; Lee, J.J.; Lin, Y.; Shen, Y.K. Warpage phenomenon of thin-wall injection molding. Int. J. Adv. Manuf. Technol. 2011, 55, 517–526. [Google Scholar] [CrossRef]
- Taguchi, G. Introduction to Quality Engineering; McGraw-Hill: New York, NY, USA, 1990. [Google Scholar]
- Yang, L.J.; Hsu, C.K.; Hsiao, F.Y.; Feng, C.K.; Shen, Y.K. A micro-aerial-vehicle (MAV) with figure-of-eight flapping induced by flexible wing frames. In Proceedings of the 47th AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition, Orlando, FL, USA, 5–8 January 2009; p. 092407. [Google Scholar]
Author | Material | Flapping-Wing Type | Fabrication | Flight Time |
---|---|---|---|---|
Ashley [1] | Polymer | Fixed wing | Unknown | 960 s |
Sirirak [2] | Titanium alloy /carbon fiber | Bird | Microelectromechinal system (MEMS) | 80 s |
Yang [3] | Polyvinylidene dufluonicle (PVDF) | Bird | MEMS | 90 s |
Ramasamy [4] | Composite carbon fiber | Bird | MEMS | Angle degree |
Zhang [5] | Carbon fiber reinforced polymer | Bird | MEMS | 100 s |
Sai [6] | Non-woven fabric plastic | Bird | MEMS | Angle degree |
Deng [7] | Polymer | Bird | Performed by author’s Lab | Angle degree |
Phan [8] | Carbon/epoxy | Bee | Performed by author’s Lab | Angle degree |
Phan [9] | Carbo/epoxy | Bee | Performed by author’s Lab | Flapping Moment |
Phan [10] | Polyethylene terephthalate | Beetle | MEMS | 1.5 m height on 3 s |
Truong [11] | POM | Bird | CNC processing | Optimization for gear |
Badrya [12] | Simulation | Dipteral insert | Simulation | Flapping force |
Hassanalian [13] | Polymer | Bird | Performed by author’s Lab | 50 m height |
Nguyen [14] | POM/carbon fiber | Bird | CNC | Thrust |
Bluman [15] | Model/simulation | Bumblebee | Model/simulation | Angle degree |
Lu [16] | Carbon fiber | Bat | Performed by author’s Lab | Nosie |
Herrero [17] | Polystyreme | Bird | CNC | Angle degree |
Yang [18] | POM/Al/Ti | Bird (dove) | Bird | CNC |
Nan [19] (review) | Polymer Cabon fiber Ti | Insect Bird | MEMS CNC Mold | Thrust Angle degree |
Nguyen [20] | POM/Carbon fiber | Insect | CNC | Flight speed |
Jankauski [21] | Model/simulation | Insect | Model/simulation | Frequency |
Badrya [22] | Polymer | Insect | CNC | Lift force drag force |
Cao [23] | Polyacrylate | Insect | CNC | Flight speed |
Gallar [24] | Carbon fiber | Fiber | CNC | Flight speed |
Lane [25] | PLA/carbon spars/epoxy/Malar | Insect (bumblebee) | 3D printing | Thrust |
Lee [26] | Model/simulation | Bird | Model/simulation | Angle degree |
Dong [27] | Polymer | Bird | Performed by author’s Lab | Angle degree |
Yoon [28] | Carbon/epoxy | Bird | CNC | Camber angle |
Nauyen [29] | Model/simulation | Insect (beetle) | Model/simulation | Angle of attack |
Wang [30] | Model/simulation | Insect | Model/simulation | Angle degree |
This study | POM | Bird | CNC/mold | 180 s |
Parameters | Levels | ||
---|---|---|---|
Level 1 | Level 2 | Level 3 | |
A. Mold temperature (°C) | 80/80/80 | 90/90/90 | 100/100/100 |
B. Melt temperature (°C) | 200/200/200 | 210/210/210 | 220/220/220 |
C. Injection pressure (bar) | 300/300/300 | 400/400/400 | 500/500/500 |
D. Packing time (s) | 1/1/1 | 1.5/1.5/1.5 | 2/2/2 |
Exp. | y1 | y2 | y3 | y4 | y5 | Ave. | S | S/N |
---|---|---|---|---|---|---|---|---|
1 | 0.0183 | 0.0182 | 0.0178 | 0.0179 | 0.0175 | 0.0179 | 0.0003 | 35.0238 |
2 | 0.0172 | 0.0172 | 0.0162 | 0.0165 | 0.0160 | 0.0166 | 0.0006 | 35.7912 |
3 | 0.0118 | 0.0120 | 0.0110 | 0.0120 | 0.0100 | 0.0114 | 0.0009 | 39.1483 |
4 | 0.0124 | 0.0124 | 0.0126 | 0.0124 | 0.0127 | 0.0125 | 0.0001 | 38.0152 |
5 | 0.0132 | 0.0132 | 0.0136 | 0.0137 | 0.0138 | 0.0135 | 0.0003 | 37.2654 |
6 | 0.0185 | 0.0185 | 0.0190 | 0.0192 | 0.0189 | 0.0188 | 0.0003 | 34.4095 |
7 | 0.0320 | 0.0320 | 0.0320 | 0.0320 | 0.0320 | 0.0320 | 0.0000 | 29.8970 |
8 | 0.0400 | 0.0450 | 0.0420 | 0.0430 | 0.0440 | 0.0428 | 0.0019 | 27.3291 |
9 | 0.0370 | 0.0370 | 0.0360 | 0.0380 | 0.0400 | 0.0376 | 0.0015 | 28.3963 |
Optimum | 0.001 | 0.002 | 0.001 | 0.003 | 0.001 | 0.0022 | 0.0000 | 53.1515 |
Average | 0.0226 | 0.0007 | 33.9195 |
Exp. | y1 | y2 | y3 | y4 | y5 | Ave. | S | S/N |
---|---|---|---|---|---|---|---|---|
1 | 0.017 | 0.016 | 0.015 | 0.014 | 0.016 | 0.0156 | 0.0011 | 36.4653 |
2 | 0.012 | 0.010 | 0.014 | 0.016 | 0.014 | 0.0132 | 0.0023 | 36.6555 |
3 | 0.010 | 0.005 | 0.007 | 0.006 | 0.009 | 0.0074 | 0.0021 | 42.5701 |
4 | 0.012 | 0.014 | 0.011 | 0.010 | 0.009 | 0.0112 | 0.0019 | 39.9711 |
5 | 0.014 | 0.016 | 0.012 | 0.012 | 0.010 | 0.0128 | 0.0023 | 38.8829 |
6 | 0.019 | 0.019 | 0.019 | 0.019 | 0.019 | 0.0194 | 0.0000 | 34.2364 |
7 | 0.021 | 0.019 | 0.019 | 0.022 | 0.020 | 0.0202 | 0.0013 | 33.8195 |
8 | 0.024 | 0.028 | 0.025 | 0.021 | 0.027 | 0.0250 | 0.0027 | 32.2306 |
9 | 0.020 | 0.019 | 0.022 | 0.024 | 0.023 | 0.0216 | 0.0021 | 32.7600 |
Optimum | 0.007 | 0.006 | 0.006 | 0.005 | 0.004 | 0.0056 | 0.0011 | 45.9063 |
Average | 0.0163 | 0.0018 | 36.3990 |
Exp. | y1 | y2 | y3 | y4 | y5 | Ave. | S | S/N |
---|---|---|---|---|---|---|---|---|
1 | 0.036 | 0.037 | 0.035 | 0.036 | 0.037 | 0.0362 | 0.0008 | 28.8717 |
2 | 0.026 | 0.028 | 0.032 | 0.029 | 0.030 | 0.0290 | 0.0022 | 30.3543 |
3 | 0.013 | 0.017 | 0.014 | 0.016 | 0.015 | 0.0150 | 0.0016 | 36.4653 |
4 | 0.019 | 0.020 | 0.019 | 0.020 | 0.021 | 0.0198 | 0.0008 | 33.9722 |
5 | 0.023 | 0.022 | 0.025 | 0.026 | 0.028 | 0.0248 | 0.0024 | 31.5802 |
6 | 0.049 | 0.043 | 0.040 | 0.041 | 0.039 | 0.0424 | 0.0040 | 27.9570 |
7 | 0.048 | 0.052 | 0.050 | 0.048 | 0.049 | 0.0494 | 0.0017 | 26.1949 |
8 | 0.055 | 0.052 | 0.058 | 0.063 | 0.063 | 0.0582 | 0.0049 | 24.2397 |
9 | 0.054 | 0.052 | 0.059 | 0.061 | 0.060 | 0.0572 | 0.0040 | 24.4362 |
Optimum | 0.010 | 0.011 | 0.008 | 0.009 | 0.007 | 0.0090 | 0.0016 | 41.8932 |
Average | 0.0369 | 0.0025 | 29.3413 |
Factor | Square Sum | Degree of Freedom | Variance | F Distribution | Confidence |
---|---|---|---|---|---|
A | 130.20 (b) 55.07 (g) 87.30 (l) | 2 (b) 2 (g) 2 (l) | 65.100 (b) 27.536 (g) 43.648(l) | 3.4630 (b) 2.2244 (g) 2.4878 (l) | 91.75% (b) 82.95% (g) 85.55% (l) |
B | 1.10 (b) 1.10 (g) 1.72 (l) | 2 (b) 2 (g) 2 (l) | 0.552 (b) 0.550 (g) 0.858 (l) | 0.0293 (b) 0.0444 (g) 0.0489 (l) | 2.88% (b) 4.32% (g) 4.74% (l) |
C | 15.29 (b) 25.40 (g) 29.19 (l) | 2 (b) 2 (g) 2 (l) | 7.647 (b) 12.699 (g) 14.595 (l) | 0.4068 (b) 1.0259 (g) 0.8319 (l) | 32.12% (b) 59.88% (g) 53.03% (l) |
D | 3.79 (b) 17.46 (g) 22.16 (l) | 2 (b) 2 (g) 2 (l) | 1.897 (b) 8.731 (g) 11.079 (l) | 0.1009 (b) 0.7053 (g) 0.6315 (l) | 9.49% (b) 47.77% (g) 44.36% (l) |
Total | 150.39 (b) 99.03 (g) 140.36 (l) | 8 (b) 8 (g) 8 (l) |
Factor | Square Sum | Degree of Freedom | Variance | F Distribution | Probability | Confidence |
---|---|---|---|---|---|---|
A | 130.20 (b) 55.07 (g) 87.295 (l) | 2 (b) 2 (g) 2 (l) | 65.100 (b) 27.536 (g) 43.648 (l) | 19.3447 (b) 3.7584 (g) 4.935 (l) | 0.24% (b) 8.75% (g) 5.40% (l) | 99.76% (b) 91.25% (g) 94.60% (l) |
B | Pooling of errors | |||||
C | Pooling of errors | |||||
D | Pooling of errors | |||||
Error | 20.1917 (b) 43.9594 (g) 53.064 (l) | 6 (b) 6 (g) 6 (l) | 3.365 (b) 7.327 (g) 8.844 (l) | S exp. Error = 1.8345 dB (b) S exp. Error = 2.7068 dB (g) S exp. Error = 2.974 dB (l) | ||
Total | 150.3925 (b) 99.0321 (g) 140.359 (l) | 8 (b) 8 (g) 8 (l) |
Material Name | ||
---|---|---|
Aluminum (6061) | Plastic (POM) | |
Total weight (g) | 2.528 | 2.279 |
Fabrication method | WCNC | Precision injection molding |
Processing | Half automation | Automation |
Processing flowchart | Complex | Simple |
Molding time (min/group) | 30 | 2 |
Reproduction | Low | High |
Percent defectives | High | Low |
Costs (NT dollar) | 620 | 12.2 |
Assembly learning time | Long | Short |
Flying time (s) | 47 | 106 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Huang, H.-Y.; Fan, F.-Y.; Lin, W.-C.; Huang, C.-F.; Shen, Y.-K.; Lin, Y.; Ruslin, M. Optimal Processing Parameters of Transmission Parts of a Flapping-Wing Micro-Aerial Vehicle Using Precision Injection Molding. Polymers 2022, 14, 1467. https://doi.org/10.3390/polym14071467
Huang H-Y, Fan F-Y, Lin W-C, Huang C-F, Shen Y-K, Lin Y, Ruslin M. Optimal Processing Parameters of Transmission Parts of a Flapping-Wing Micro-Aerial Vehicle Using Precision Injection Molding. Polymers. 2022; 14(7):1467. https://doi.org/10.3390/polym14071467
Chicago/Turabian StyleHuang, Huei-Yu, Fang-Yu Fan, Wei-Chun Lin, Chiung-Fang Huang, Yung-Kang Shen, Yi Lin, and Muhammad Ruslin. 2022. "Optimal Processing Parameters of Transmission Parts of a Flapping-Wing Micro-Aerial Vehicle Using Precision Injection Molding" Polymers 14, no. 7: 1467. https://doi.org/10.3390/polym14071467
APA StyleHuang, H. -Y., Fan, F. -Y., Lin, W. -C., Huang, C. -F., Shen, Y. -K., Lin, Y., & Ruslin, M. (2022). Optimal Processing Parameters of Transmission Parts of a Flapping-Wing Micro-Aerial Vehicle Using Precision Injection Molding. Polymers, 14(7), 1467. https://doi.org/10.3390/polym14071467