Contribution
Computational Mapping of End-Correction for Perforated Plates using a Deep Neural Network
* Presenting author
Abstract:
In this work the end-correction used for lumped (semi analytical) perforated plate models is computed in detail and its dependency on frequency and geometry parameters is mapped using a deep neural network (DNN). To achieve this, the end correction is computed exactly in the linear acoustic regime using a full linearized Navier-Stokes (FLNS) model (aka. performing a thermoviscous acoustic simulation). The associated complex-valued end-correction is deduced and a DNN is trained for further efficient modeling of perforated plates.