Using a One-Dimensional Linear Microphone Array to Identify Work Area Noise Sources

The paper addresses the critical problem of noise pollution in production and the necessity of accurate noise assessment methods. This paper describes the technologies used to implement linear microphone arrays and data processing methods to localize the noise source. Experiments were conducted in an anechoic chamber and in a machine shop to detect and analyze noise sources. The paper also discusses the limitations of beamforming algorithms and the importance of factors such as microphone array geometry and sampling rate in achieving accurate results. A data processing algorithm using Python was developed to determine the arrival pattern of sound waves arriving at a microphone array. The effectiveness of the algorithm is demonstrated through experiments and simulations using Simulink. The results show that the algorithm accurately detects the source of the most intense noise and can be used for spectrum analysis and equipment diagnosis. However, the algorithm is less accurate when identifying multiple sources with different frequencies and is sensitive to interference.