Predicting the acoustical properties of 3d printed resonators using a matrix of impulse responses and mode interpolation

Abstract

Accurately predicting acoustical properties of 3D printed models is of interest to instrument designers who explore novel geometries. We introduce a technique to carry out these estimates using a database of impulse responses and mode interpolation. 3D models are organized as a function of their physical characteristics and placed into a multidimensional space/matrix. The models at the boundaries of this space define the limits of our prediction algorithm and they are produced using 3D printing. Impulse responses of these models are measured, and modal information is extracted from each object. Mode parameters are interpolated within the matrix to predict the frequency response of unprinted models that fall within the geometrical space of the test matrix. A physical model using modal synthesis also allows us to listen to the resulting resonator.

Let’s Print an Ecology in 3D (and 4D)

“…Customization parameters such as layer thickness and infill patterns improved energy efficiency by up to 11.51% and carbon emission reduction by up to 49.91%. Predicting the acoustical properties of materials using a matrix of impulse responses and mode interpolation [78] or the quality of electronics structures manufactured by a 3D inkjet printing process [79,80] was performed with different mathematical models to limit experimental trials save energy and material use during the process. More recently, artificial intelligence, machine learning, and deep learning have become integral components of 3D printing used in various aspects of additive manufacturing, including design optimization, predicting 3D printing parameters and process control, material development, part orientation, support generation, defect detection, quality control, etc.…”