A Design Method for Second-Order Helmholtz Resonators Based on Deep Learning

Abstract

The invention discloses a second-order Helmholtz resonator design method based on deep learning, which is used for two resonance frequencies f based on the second-order Helmholtz resonator 1 and f 2 , to design the geometric parameters of the second-order Helmholtz resonator; the method includes: generating N sound insulation curves that meet the requirements according to the specific design requirements, and inputting the N sound insulation curves into the pre-established and trained The deep neural network model outputs N sets of corresponding equivalent electrical parameters; based on the conversion relationship formula between the second-order Helmholtz resonator geometric parameters and equivalent electrical parameters, according to N sets of equivalent electrical parameters, N sets of two The geometric parameters of the second-order Helmholtz resonators; the corresponding sound insulation curves are calculated according to the geometric parameters of N groups of second-order Helmholtz resonators, and the optimal sound insulation curve is selected from the N sound insulation curves. The geometric parameters of the second-order Helmholtz resonator corresponding to the sound insulation curve are used as the designed structural parameters.

Description

technical field [0001] The invention relates to the field of deep learning, in particular to a deep learning-based second-order Helmholtz resonator design method. Background technique [0002] Acoustic artificial structure is a kind of artificially designed composite structure. The properties of this composite structure mainly depend on its appearance, size, shape and arrangement, so it has many novel physical properties that natural materials do not have. In the practical application of acoustic artificial structures, we often need to design the geometric structure of the unit according to the desired physical properties. This design is a typical reverse design problem. The traditional design route can be divided into two parts: First, we can use some classic physical models and the experience and intuition accumulated from previous practice to determine the framework of the model. For example, we can achieve sound absorption and isolation based on the Helmholtz resonator ...

AI Technical Summary

Problems solved by technology

However, due to the large number of geometric parameters of THR and the coupling between parameters, it is difficult to analyze its physical characteristics

Therefore, on the one hand, the traditional THR design method requires rich experience for structure selection, and on the other hand, it also needs to make a lot of attempts on the target parameters, which not only has high requirements for the designer itself, but also often consumes a lot of time.

Images