Numerical Noise Transfer Analysis of a Flexible Supported Gearbox Based on Impedance Model and Noise Transfer Function

Authors:
Open Access
Publisher website
Google Scholar
Abstract
To investigate the gearbox radiation noise properties under various rotational speeds, a noise prediction method based on impedance model and noise transfer function (NTF) is proposed. One only needs to extract the NTF of the housing once rapid gearbox noise prediction under different working conditions is realized. Taking a flexible supported gearbox as a research object, the external excitation of the housing (the bearing excitation load and isolator excitation load) is calculated through a gear-housing-foundation-coupled impedance model, and the noise transfer function is simulated through the vibroacoustic-coupled boundary element model; then the radiation noise is obtained. Based on this model, the noise transfer analysis of the housing is carried out, different excitation components and NTF components are compared, and the contributions of different excitation components to noise are compared. Results show that the radiation noise of gearbox is mainly excited by the high-speed bearing, while the low-speed bearing and isolator have little influence on noise. At low speed, vertical force, axial force, and moment excitation of bearings all contribute to the radiation noise while at high speed, the gearbox radiation noise is mainly generated by vertical excitation force of bearings.
Funding Information
  • National Natural Science Foundation of China (51535009, B13044, 20192070, 20192007, 20192053)
  • 111 project of China (51535009, B13044, 20192070, 20192007, 20192053)
  • Scientific Research Starting Foundation of Taiyuan University of Science and technology (51535009, B13044, 20192070, 20192007, 20192053)
  • Scientific Research Starting Foundation of Taiyuan University of Science and technology (51535009, B13044, 20192070, 20192007, 20192053)
  • Scientific Research Starting Foundation of Taiyuan University of Science and technology (51535009, B13044, 20192070, 20192007, 20192053)

This publication has 18 references indexed in Scilit: