Coherence-based sensor set expansion for optimal sensor placement in active road noise control. (15th April 2022)

• Record Type: Journal Article
• Authors: Park, Yun Seol; Cho, Mun Hwan; Oh, Chi Sung; Kang, Yeon June
• Is Part Of: Mechanical systems and signal processing. Volume 169(2022)
• Publication Date: 2022-04-15
• Subjects: Active road noise control -- Optimal sensor placement -- Effective independence method -- Coherence function -- Fisher information matrix; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2021.108788
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A method is proposed to determine the optimal sensor set for active road noise control. A reference sensor set was expanded using the coherence-based information matrix. Noise reduction level was calculated using the optimal filter—Wiener filter. Proposed method demonstrated better performance than the conventional estimation method. Abstract: Road noise is the dominant noise source that directly affects the vehicle noise, vibration, and harshness (NVH) performance. The active noise control (ANC) technology is a promising solution for reducing interior noise from tire–road interactions. Active road noise control (ARNC) systems include a set of acceleration sensors, which acquire vibrations that cause road noise, and the locations of the sensors is important to accurately determine the vibration transfer path. Therefore, it is necessary to find an optimal sensor set among the candidate sensors to maximize the performance of the ARNC system. The trial-and-error method for simulating all possible sensor combinations requires a high computational cost. This paper proposes a method that determines the set of reference sensor locations with the coherence analysis and the Fisher information matrix. The methodology starts with one sensor highly correlated with the output sound pressure level signals. The initial sensor set is iteratively expanded to the desired number of sensors by maximizing the determinant of the coherence information matrix. This approach can significantly reduce the amount of computation and subsequently shorten the time required. In addition, the obtained local optimal results indicated a reduction in broadband road noise of approximately 7 dBA and showed an error within 0.2 △dBA of the targeted noise reduction results in all driving conditions.