• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2023 Fiscal Year Final Research Report

Study of broadband noise prediction generated from a low-pressure fan based on machine learning

Research Project

  • PDF
Project/Area Number 21K12294
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 64020:Environmental load reduction and remediation-related
Research InstitutionNagasaki University

Principal Investigator

Sasaki Soichi  長崎大学, 工学研究科, 助教 (00304965)

Project Period (FY) 2021-04-01 – 2024-03-31
Keywords広帯域騒音 / ファン / 風洞 / 機械学習 / 圧力PSD
Outline of Final Research Achievements

Among the aerodynamic noises generated by low-pressure fans, broadband noise is considered an important issue in the field of fluid mechanics. In this study, we developed a low-noise wind tunnel with an anechoic room and used machine learning to generate pressure power spectral density (PSD), which is a key parameter for predicting broadband noise. Based on this pressure PSD, it is possible to predict the broadband noise generated from a flat plate with high accuracy. Moreover, from the comparison between the measured aerodynamic noise in the experimental apparatus of a low-pressure fan and the noise predicted by machine learning, we revealed that the broadband noise generated by the fan was caused by the pressure PSD of Karman vortex shedding.

Free Research Field

流体工学

Academic Significance and Societal Importance of the Research Achievements

厚生労働省の新型コロナウィルス感染症対策専門会議は、感染リスクのある空間として「換気の悪い密閉空間」を挙げていた。その屋内換気のために常時稼働する換気用のファンには、排気性能だけでなく、静穏な環境を保つために低騒音であることも求められる。このファンから発生する広帯域騒音の解析的な予測では、その精度や計算コストが現実的な課題となっている。機械学習による広帯域騒音の予測では、流れの数値シミュレーションのように運動方程式や質量の保存則を満足しながら計算する必要がない。このため、低圧ファンの空力騒音を予測するための計算コストを大幅に低減することが可能になる。

URL: 

Published: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi