An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning
Elastic metasurfaces have become one of the most promising platforms for manipulating mechanical wavefronts with the striking feature of ultra-thin geometry. The conventional design of mechanical metasurfaces significantly relies on numerical, trial-and-error methods to identify structural parameter...
| Main Authors: | , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/169420 |
| _version_ | 1826119075949969408 |
|---|---|
| author | Zhou, Weijian Wang, Shuoyuan Wu, Qian Xu, Xianchen Huang, Xinjing Huang, Guoliang Liu, Yang Fan, Zheng |
| author2 | School of Mechanical and Aerospace Engineering |
| author_facet | School of Mechanical and Aerospace Engineering Zhou, Weijian Wang, Shuoyuan Wu, Qian Xu, Xianchen Huang, Xinjing Huang, Guoliang Liu, Yang Fan, Zheng |
| author_sort | Zhou, Weijian |
| collection | NTU |
| description | Elastic metasurfaces have become one of the most promising platforms for manipulating mechanical wavefronts with the striking feature of ultra-thin geometry. The conventional design of mechanical metasurfaces significantly relies on numerical, trial-and-error methods to identify structural parameters of the unit cells, which requires huge computational resources and could be extremely challenging if the metasurface is multi-functional. Machine learning technique provides another powerful tool for the design of multi-functional elastic metasurfaces because of its excellent capability in building nonlinear mapping relation between high-dimensional input data and output data. In this paper, a machine learning network is introduced to extract the complex relation between high-dimensional geometrical parameters of the metasurface unit and its high-dimensional dynamic properties. Based on a big dataset, the well-trained network can play the role of a surrogate model in the inverse design of a multi-functional elastic metasurface to significantly shorten the time for the design. Such method can be conveniently extended to design other multi-functional metasurfaces for the manipulation of optical, acoustical or mechanical waves. |
| first_indexed | 2024-10-01T04:54:08Z |
| format | Journal Article |
| id | ntu-10356/169420 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:54:08Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | ntu-10356/1694202023-07-22T16:48:10Z An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning Zhou, Weijian Wang, Shuoyuan Wu, Qian Xu, Xianchen Huang, Xinjing Huang, Guoliang Liu, Yang Fan, Zheng School of Mechanical and Aerospace Engineering Engineering::Mechanical engineering Acoustic Metasurface Inverse Design Elastic metasurfaces have become one of the most promising platforms for manipulating mechanical wavefronts with the striking feature of ultra-thin geometry. The conventional design of mechanical metasurfaces significantly relies on numerical, trial-and-error methods to identify structural parameters of the unit cells, which requires huge computational resources and could be extremely challenging if the metasurface is multi-functional. Machine learning technique provides another powerful tool for the design of multi-functional elastic metasurfaces because of its excellent capability in building nonlinear mapping relation between high-dimensional input data and output data. In this paper, a machine learning network is introduced to extract the complex relation between high-dimensional geometrical parameters of the metasurface unit and its high-dimensional dynamic properties. Based on a big dataset, the well-trained network can play the role of a surrogate model in the inverse design of a multi-functional elastic metasurface to significantly shorten the time for the design. Such method can be conveniently extended to design other multi-functional metasurfaces for the manipulation of optical, acoustical or mechanical waves. Agency for Science, Technology and Research (A*STAR) Ministry of Education (MOE) Published version The authors would like to acknowledge funding support from the Ministry of Education Singapore under Grant No. MOE2019- T2-2-068, A*STAR Singapore Science and Engineering Research Council under AME Individual Research Grant (IRG) 2018 Grant Call (Project No. A1983c0030), and the funding support from National Natural Science Foundation of China (No. 12202117). 2023-07-18T04:40:27Z 2023-07-18T04:40:27Z 2023 Journal Article Zhou, W., Wang, S., Wu, Q., Xu, X., Huang, X., Huang, G., Liu, Y. & Fan, Z. (2023). An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning. Materials & Design, 226, 111560-. https://dx.doi.org/10.1016/j.matdes.2022.111560 0264-1275 https://hdl.handle.net/10356/169420 10.1016/j.matdes.2022.111560 2-s2.0-85146050871 226 111560 en MOE2019-T2-2-068 A1983c0030 Materials & Design © 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). application/pdf |
| spellingShingle | Engineering::Mechanical engineering Acoustic Metasurface Inverse Design Zhou, Weijian Wang, Shuoyuan Wu, Qian Xu, Xianchen Huang, Xinjing Huang, Guoliang Liu, Yang Fan, Zheng An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning |
| title | An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning |
| title_full | An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning |
| title_fullStr | An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning |
| title_full_unstemmed | An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning |
| title_short | An inverse design paradigm of multi-functional elastic metasurface via data-driven machine learning |
| title_sort | inverse design paradigm of multi functional elastic metasurface via data driven machine learning |
| topic | Engineering::Mechanical engineering Acoustic Metasurface Inverse Design |
| url | https://hdl.handle.net/10356/169420 |
| work_keys_str_mv | AT zhouweijian aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT wangshuoyuan aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT wuqian aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT xuxianchen aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT huangxinjing aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT huangguoliang aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT liuyang aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT fanzheng aninversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT zhouweijian inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT wangshuoyuan inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT wuqian inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT xuxianchen inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT huangxinjing inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT huangguoliang inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT liuyang inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning AT fanzheng inversedesignparadigmofmultifunctionalelasticmetasurfaceviadatadrivenmachinelearning |