PlumX Metrics
Embed PlumX Metrics

Research on the acoustic performance of MPP and inverse design

Journal of Applied Physics, ISSN: 1089-7550, Vol: 137, Issue: 9
2025
  • 0
    Citations
  • 0
    Usage
  • 0
    Captures
  • 0
    Mentions
  • 0
    Social Media
Metric Options:   Counts1 Year3 Year

Article Description

Under specifying the noise frequency and sound absorption coefficient in engineering practice, the traditional method cannot quickly get the microperforated panel (MPP) structure size that meets the condition. Therefore, this paper establishes the MPP acoustic impedance model and calculates the sound absorption coefficient of MPP, revealing the change rule of the damping state. Based on the finite element method, the corresponding numerical model is built to explain the MPP sound absorption principle. Experiments were carried out using impedance tubes to verify the accuracy of the acoustic impedance model. It is found that the micropore diameter is directly proportional to the maximum sound absorption frequency, and the plate thickness and acoustic cavity depth are inversely proportional to the maximum sound absorption frequency. Aiming at the low efficiency of MPP inverse design, this paper proposed an inverse design method of MPP based on 1DCNN by constructing a network architecture adapted to the characteristics of MPP data and establishing a mapping model between dimensional parameters and absorption performance. The model training results show that the loss of the test set is as low as 0.703, and the R values of the absorption performance indexes are all higher than 0.997. Considering the actual physical constraints and taking the engineering requirements as input, the output designed MPP has an absorption bandwidth of 349 Hz, and the maximal acoustic absorption frequency is α = 0.997; the result meets the engineering demand and provides an essential theoretical basis and engineering application reference for the design of the MPP structure.

Bibliographic Details

Provide Feedback

Have ideas for a new metric? Would you like to see something else here?Let us know