Answers are grounded in this paper and cite their sources.
Computational modeling study demonstrates accurate material inversion and tone prediction in leather drumheads using physics-informed neural networks, highlighting pathways for digital instrument...
This study proposes a physics-informed neural network framework for quantitative analysis of the coupling mechanism between leather drumhead material properties and radiated acoustic wave characteristics. A transversely isotropic Ogden hyperelastic constitutive model is embedded into nonlinear shell vibration equations and incorporated as a physical constraint within the learning architecture. The proposed framework jointly utilizes transient displacement fields captured through high-speed optical measurements and acoustic wave information acquired by an array-based microphone system. Through vibration–radiation coupling modeling, structural dynamic responses are linked to farfield sound pressure distributions and spectral characteristics. Higher-order statistical moments of the Mel-frequency spectrum are employed to characterize multidimensional timbre attributes, while a differentiable material-parameter layer enables end-to-end inversion from target acoustic wave features to constitutive parameters. Experimental results demonstrate relative inversion errors of 2.3%, 2.5%, and 2.0% for Young’s modulus, shear modulus, and damping coefficient, respectively. The model further exhibits high consistency in acoustic feature reconstruction and superior capability in capturing high-frequency modal responses compared with conventional linear constitutive approaches. The results reveal a quantitative relationship among material nonlinearity, vibration behavior, and acoustic wave propagation characteristics. The proposed framework provides an engineering methodology for wave-based parameter identification, vibration– radiation analysis, and acoustic field prediction, supporting intelligent design and digital restoration of percussion instruments.
No takes yet. Share an insight, caveat, or question.
Shi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: