Improved Variable Step Size Least Mean Square Algorithm for Pipeline Noise

• Published in: Scientific programming Vol. 2022; pp. 1 - 16
• Main Authors: Zhang, Xiaohui, Yang, Songnan, Liu, Yuanyuan, Zhao, Wei
• Format: Journal Article
• Language: English
• Published: New York Hindawi 09.02.2022; John Wiley & Sons, Inc
• Subjects: Active noise control; Adaptive filters; Air conditioning; Algorithms; Convergence; Error reduction; Error signals; HVAC; Indoor air quality; Mean square values; Microphones; Noise control; Noise levels; Noise reduction; Outdoor air quality; Pollutants; Porous materials; Smoothness; Sound waves; VOCs; Volatile organic compounds
• ISSN: 1058-9244; 1875-919X
• DOI: 10.1155/2022/3294674

In this study, we employ the active noise control (ANC) method to eliminate the low-frequency part of the noise generated by the rotation of the axial fan in heating, ventilation, and air-conditioning (HVAC) pipelines. Because the traditional variable step size least mean square (VSS-LMS) algorithm has poor tracking performance, we propose a variable step size filtered-X least mean square (FXLMS) algorithm based on the arctangent function to improve the adaptive filtering method of the convergence speed and noise cancellation effect. The step size of the proposed algorithm can be adjusted according to the error. When the error signal is significant, a larger step can be obtained, and when the error is small, the step size smoothness of the algorithm can be optimized. Compared with the traditional VSS-LMS algorithm, the convergence speed of the proposed algorithm is increased by 29%, the noise reduction effect is enhanced by 19%, and the mean square error (MSE) is reduced by 23% (0.0084). In addition, we developed a hardware experimental platform based on noise characteristics. In the noise reduction test using a GB/T 5836.2-06 standard PVC pipeline, the system reduced the noise by 12–17 dB.