Numerical Assessment of Rectangular Side Inlet/Outlet Plenums Internally Equipped with Two Crossed Baffles Using an FEM, Neural Network, and GA Method

摘要

In this paper, a rectangular plenum internally equipped with two crossed baffles within a fixed space is assessed. To simplify the optimization process, a simplified objective function (OBJ) is constructed using the finite element model (FEM) in conjunction with the polynomial neural network model (ANNM). To assess an optimal plenum, the best OBJ will be numerically searched using a genetic algorithm (GA). Before the GA operation is performed, the accuracy of the FEM is verified using the analytical data. In addition, a case study of shape optimization on a space-constrained plenum at three targeted tones (1950 Hz, 2450 Hz, and 2850 Hz) has been introduced and carried out. The results reveal that the maximum value of the sound transmission loss (STL) can be accurately obtained at the desired frequencies. Consequently, the algorithm proposed in this study provides an efficient way to develop optimal rectangular plenums internally equipped with two crossed baffles.