A model for three-dimensional simulation of acoustic emissions from rotating machine vibration

Area of Science:

• Acoustics and Noise Control
• Mechanical Engineering
• Vibration Analysis

Background:

• Industrial noise poses challenges for worker health and productivity.
• Conventional active noise control (ANC) has limitations in achieving broad spatial coverage, especially for mobile operators.
• Combined passive and active noise control strategies are effective but require advanced modeling for optimal performance.

Purpose of the Study:

• To propose a novel three-dimensional vibration-acoustic model for predicting and controlling industrial noise.
• To enhance the spatial coverage of active noise control (ANC) systems in dynamic environments.
• To provide a predictive tool for designing effective ANC solutions for machinery noise.

Main Methods:

• Development of a 3D vibration-acoustic model using accelerometer data from a centrifugal pump as input.
• Utilizing Auto-Regressive with eXogenous inputs (ARX) models to represent single-input, single-output (SISO) systems linking machine vibration to noise levels.
• Employing spatial interpolation techniques to estimate model parameters across a defined area.

Main Results:

• The proposed model demonstrated good agreement between predicted noise levels and experimental measurements.
• The model successfully captured the relationship between machine vibration and acoustic output within the test environment.
• Validation indicates the model's potential for designing improved ANC systems.

Conclusions:

• The developed 3D vibration-acoustic model is a promising tool for enhancing active noise control (ANC) strategies.
• The model can aid in designing more effective noise attenuation solutions, particularly for scenarios requiring broad spatial coverage.
• This approach offers a pathway to improved noise mitigation in industrial settings with moving equipment or personnel.