A new robust narrowband active noise control system with online secondary-path modeling and frequency mismatch compensation

A new robust narrowband active noise control system with online secondary-path modeling and frequency mismatch compensation
复制标题

DOI:
10.1109/icamechs.2018.8507129
复制
发表时间:
2018-08
期刊:
2018 International Conference on Advanced Mechatronic Systems (ICAMechS)
影响因子:
--
通讯作者:
Yaping Ma;Yegui Xiao;Liying Ma;K. Khorasani;Jian-ming Ding;Jianhui Lin
Yaping Ma;Yegui Xiao;Liying Ma;K. Khorasani;Jian-ming Ding;Jianhui Lin
中科院分区:
其他
文献类型:
--
作者:
Yaping Ma;Yegui Xiao;Liying Ma;K. Khorasani;Jian-ming Ding;Jianhui Lin

文献摘要

相似文献

窄带有源噪声控制(ANC,NANC)系统在抑制由旋转机械产生的恼人噪声信号方面提供了优异的性能,所述恼人噪声信号可以被建模为加性噪声中的正弦信号。在实际应用中,为了实现足够的降噪,NANC系统需要解决两个重要问题,包括次级路径建模(SPM)和频率失配(FM),即非声学定时信号传感器的误差。在本文中,提出了一种新的鲁棒NANC系统,它包含三个子系统,即:1)在线次级路径建模子系统,设计用于跟踪时变次级路径,2)FM补偿子系统,为每个ANC通道提供修改的频率,以及3)成本降低子系统,其中无论目标频率的数量如何,仅放置两个x滤波块。这三个子系统无缝集成,使得整个NANC系统在降噪和计算效率方面都具有很大的优势。仿真结果以及应用到真实的噪声进行证明我们提出的系统的有效性和能力,以及鲁棒性。
Narrowband active noise control (ANC, NANC) systems provide excellent performance in suppressing the annoying noise signals generated by rotating machines, which may be modeled as sinusoidal signals in additive noise. In realworld applications, two important issues including secondary path modeling (SPM) and frequency mismatch (FM), namely, the error of the nonacoustic timing signal sensor, need to be addressed by the NANC system in order to achieve sufficient noise reduction. In this paper, a new robust NANC system is proposed which contains three subsystems, namely: 1) an online secondary-path modeling subsystem designed to track the time-varying secondary path, 2) an FM compensation subsystem providing modified frequencies to every ANC channel, and 3) a cost reduction subsystem where only two x-filtering blocks are placed regardless of the number of the targeted frequencies. The three subsystems are seamlessly integrated such that the entire NANC system enjoys great advantages in both noise reduction and computational efficiency. Simulation results as well as application to real noise are conducted to demonstrate the effectiveness and capabilities as well as robustness of our proposed system.