Machine-learning-driven on-demand design of phononic beams

He, Liangshu; Guo, Hongwei; Jin, Yabin; Zhuang, Xiaoying; Rabczuk, Timon; Li, Yan

doi:10.1007/s11433-021-1787-x

Cited by 42 publications

(9 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Li et al (2020b) came up with a data-driven approach to design phononic crystals, assisted by image-based finite element analysis and deep learning. He et al (2021) efficiently reversed the engineering structural parameters to maximize the band gap width with reinforcement learning algorithms. Li et al (2021) suggested an optimization method based on subset simulation and generative adversarial network (GAN) guidance to optimize the frequency band gap characteristic of periodic structures.…”

Section: Introductionmentioning

confidence: 99%

Vibration transmission characteristic prediction and structure inverse design of acoustic metamaterial beams based on deep learning

Chen

et al. 2023

Journal of Vibration and Control

View full text Add to dashboard Cite

In this paper, an intelligent prediction and design method of acoustic metamaterial beams based on deep learning is presented. A theoretical model of acoustic metamaterial beams is derived by using the spectral element method to calculate the band structure and transmission characteristic of beam-type acoustic metamaterial. A high-degree-of-freedom design space dataset of band structure and transmission characteristics is constructed. In addition, the vibration transmission characteristics of acoustic metamaterial beams are predicted and the on-demand inverse design of acoustic metamaterial beams is realized by constructing a fully connected deep learning neural network model. Furthermore, the forward prediction and reverse design network model is verified by using the autoencoder neural network model. Results show that the predicted value of the autoencoder neural network structure is in good agreement with the target value, which indicates the feasibility of the intelligent design method of acoustic metamaterials based on deep learning. The presented intelligent design method could be potentially utilized in the field of fast and efficient acoustic metamaterial design for low frequency vibration isolation in engineering structures.

show abstract

Section: Introductionmentioning

confidence: 99%

Vibration transmission characteristic prediction and structure inverse design of acoustic metamaterial beams based on deep learning

Chen

et al. 2023

Journal of Vibration and Control

View full text Add to dashboard Cite

show abstract

“…In this paper, we define a RL procedure for the optimisation of graded metamaterials for energy harvesting under general loading conditions, here exemplified for the common cases of magnetic loading or random vibrations. Other works treating design optimisation as a Markov decision process (MDP) solvable via RL were proposed for phononic crystals for matching 23 , 24 or maximising band-gaps 25 , and for acoustic metamaterials to minimise wave scattering 26 . The combination of MDP and RL was also employed in structural optimisation 27 , 28 , material design 29 , 30 , optics 31 , and chemistry 32 .…”

Section: Introductionmentioning

confidence: 99%

Reinforcement learning optimisation for graded metamaterial design using a physical-based constraint on the state representation and action space

Rosafalco,

De Ponti,

Iorio

et al. 2023

Sci Rep

View full text Add to dashboard Cite

The energy harvesting capability of a graded metamaterial is maximised via reinforcement learning (RL) under realistic excitations at the microscale. The metamaterial consists of a waveguide with a set of beam-like resonators of variable length, with piezoelectric patches, attached to it. The piezo-mechanical system is modelled through equivalent lumped parameters determined via a general impedance analysis. Realistic conditions are mimicked by considering either magnetic loading or random excitations, the latter scenario requiring the enhancement of the harvesting capability for a class of forcing terms with similar but different frequency content. The RL-based optimisation is empowered by using the physical understanding of wave propagation in a such local resonance system to constrain the state representation and the action space. The procedure outcomes are compared against grading rules optimised through genetic algorithms. While genetic algorithms are more effective in the deterministic setting featuring the application of magnetic loading, the proposed RL-based proves superior in the inherently stochastic setting of the random excitation scenario.

show abstract

“…More recently, topological metamaterial, developed from the topological insulator in condensed matter, has shown great advantages on robust wave guiding and manipulation, owing to topological protection features [32][33][34]. For example, Wang et al [32] found the Fano resonance in a topological metamaterial is robust against random perturbations.…”

Section: Introductionmentioning

confidence: 99%

Vibration Energy Harvesting from the Subwavelength Interface State of a Topological Metamaterial Beam

Wang

Zhu

et al. 2022

Micromachines

View full text Add to dashboard Cite

Topological metamaterial has been a research hotpot in both physics and engineering due to its unique ability of wave manipulation. The topological interface state, which can efficiently and robustly centralize the elastic wave energy, is promising to attain high-performance energy harvesting. Since most of environmental vibration energy is in low frequency range, the interface state is required to be designed at subwavelength range. To this end, this paper developed a topological metamaterial beam with local resonators and studied its energy-harvesting performance. First, the unit cell of this topological metamaterial beam consists of a host beam with two pairs of parasitic beams with tip mass. Then, the band structure and topological features are determined. It is revealed that by tuning the distance between these two pairs of parasitic beams, band inversion where topological features inverse can be obtained. Then, two sub-chains, their design based on two topologically distinct unit cells, are assembled together with a piezoelectric transducer placed at the conjunction, yielding the locally resonant, topological, metamaterial, beam-based piezoelectric energy harvester. After that, its transmittance property and output power were obtained by using the frequency domain analysis of COMSOL Multiphysics. It is clear that the subwavelength interface state is obtained at the band-folding bandgap. Meanwhile, in the interface state, elastic wave energy is successfully centralized at the conjunction. From the response distribution, it is found that the maximum response takes place on the parasitic beam rather than the host beam. Therefore, the piezoelectric transducer is recommended to be placed on the parasitic beam rather than host beam. Finally, the robustness of the topological interface state and its potential advantages on energy harvesting were studied by introducing a local defect. It is clear that in the interface state, the maximum response is always located at the conjunction regardless of the defect degree and location. In other words, the piezoelectric transducer placed at the conjunction can maintain a stable and high-efficiency output power in the interface state, which makes the whole system very reliable in practical implementation.

show abstract

Machine-learning-driven on-demand design of phononic beams

Cited by 42 publications

References 51 publications

Vibration transmission characteristic prediction and structure inverse design of acoustic metamaterial beams based on deep learning

Vibration transmission characteristic prediction and structure inverse design of acoustic metamaterial beams based on deep learning

Reinforcement learning optimisation for graded metamaterial design using a physical-based constraint on the state representation and action space

Vibration Energy Harvesting from the Subwavelength Interface State of a Topological Metamaterial Beam

Contact Info

Product

Resources

About