We integrate the power of data-driven design and optimization, bringing a new level of intelligence to the creative process. By harnessing the vast capabilities of artificial intelligence, we can explore design spaces that were previously unimaginable.
Inverse design framework for broadband acoustic metamaterials
These researches aim to overcome the limitations of conventional inverse design methods for acoustic metamaterials by developing a machine learning-based framework capable of generating non-parametric structures with target acoustic responses. By leveraging variational autoencoders and optimization in the latent space, the proposed method enables efficient structure generation and significantly enhances sound attenuation performance—achieving broader transmission loss bandwidth and multi-peak acoustic response beyond the capabilities of traditional design approaches.
Data-driven mechanical metamaterial-based design
Mechanical metamaterials are emerging as an innovative solution for controlling deformation in structures exposed to intensive mechanical and thermal stimuli, such as wearable devices, solar substrates, and semiconductor circuit boards. Our research group is developing a deep learning-based design framework for mechanical metamaterials tailored to these diverse applications.
[Related works]
Ji Gyo Park, Baekgyu Kim, Jin Yeong Song, Keon Ko, Ho Kyoung Lee, Dongwhi Choi**, Seunghun Baek**, Sang Min Park**, "Machine learning-driven optimization of locally resonant metamaterials for simultaneous vibration control and effective triboelectric sensing", Nano Energy, 142, Part A, 111224 (2025) (IF = 17.1, Top 5.35% in JCR 2024)
Min Woo Cho*, Seok Hyeon Hwang*, Jun-Young Jang*, Jin Yeong Song, Sun-kwang Hwang, Kyoung Je Cha, Dong Yong Park**, Kyungjun Song**, Sang Min Park**, "Inverse design of non-parameterized ventilated acoustic resonator via variational autoencoder with acoustic response-encoded latent space", Materials Today Communications, 47, 113055 (2025) (IF = 4.5, Top 36.74% in JCR 2024)
Keon Ko*, Min Woo Cho*, Kyungjun Song, Dong Yong Park**, Sang Min Park**, "Inverse design of non-parametric acoustic metamaterials via dual variational autoencoder and iterative transfer learning to break parametric boundaries in dataset", Engineering Applications of Artificial Intelligence, 151, 110735 (2025) (IF = 8.0, Top 2.86% in JCR 2024)
Min Jik Kim, Seon Yeong Yang, Woo Seok Yang, Sehyeok Oh, Sang Min Park** , Da Seul Shin**, "Exploration of physics-related latent vectors in hot working of Inconel 718 superalloy using autoencoder", Journal of Materials Research and Technology, 35, 6749–6762 (2025) (IF = 6.6, Top 10.42% in JCR 2024)
Jin Yeong Song, Seok Hyeon Hwang, Min Woo Cho, Keon Ko, BaekGyu Kim, Kyungjun Song, Sang Min Park, "Inverse design of ventilated acoustic resonators using a sound transmission loss-encoded variational autoencoder", Journal of Mechanical Science and Technology, 39, 2, 508–519 (2025) (IF = 1.7, Top 64.84% in JCR 2024)
Min Woo Cho, Seok Hyeon Hwang, Jun Young Jang, Sun-kwang Hwang, Kyoung Je Cha, Dong Yong Park**, Kyungjun Song**, Sang Min Park**, "Beyond the limits of parametric design: latent space exploration strategy enabling ultra-broadband acoustic metamaterials", Engineering Applications of Artificial Intelligence, 133, Part F, 108595 (2024) (IF = 7.5, Top 2.79% in JCR 2023)
Min Woo Cho, Keon Ko, Majid Mohammadhosseinzadeh, Ji Hoon Kim, Dong Yong Park, Da Seul Shin**, Sang Min Park**, "Inverse design of Bézier curve-based mechanical metamaterials with programmable negative thermal expansion and negative Poisson’s ratio via data augmented deep autoencoder", Materials Horizons, 11, 2615–2627 (2024) (IF = 12.2, Top 8.90% in JCR 2023)