Active Noise Control of a Moving Evaluation Point in One Dimensional Acoustic Field (3rd Report, Proposal of Time-delay Separation Method in the Secondary Path Identification)

Active Noise Control of a Moving Evaluation Point in One Dimensional Acoustic Field (3rd Report, Proposal of Time-delay Separation Method in the Secondary Path Identification)
复制标题

一维声场中移动评价点的主动噪声控制(第三次报告,二次路径识别中时延分离方法的提出)

DOI:
10.1299/kikaic.72.402
复制
发表时间:
2006
期刊:
Transactions of the Japan Society of Mechanical Engineers. C
影响因子:
--
通讯作者:
S. Koike
S. Koike
中科院分区:
--
文献类型:
--
作者:
H. Iida;Mitsuru Nakazawa;S. Koike

文献摘要

被引文献

相似文献

在我们以前的论文中已经研究了移动点的有源噪声控制。该模型的特点是噪声控制源与评价点(目标降噪点)之间的传递特性随评价点的移动而变化。为了解决这一问题,引入了真实的实时字符识别算法。然而,辨识所用的收敛算法仅在系统平稳时才有效,这也是辨识精度随评估点速度增加而降低的原因之一。此外,在计算辨识过程中,评价点会发生移动,会导致辨识结果与真实的系统存在偏差。我们提出了新的技术,其中的传输特性分为时间延迟因子和主要成分。一旦对字符进行了划分,由运动引起的传输特性的变化主要体现在时延因子上,由于时延因子是位置的函数,因此很容易得到时延因子。应用这一思想,可以预期识别方面的显著改进。本文对该方法进行了数值和实验研究。
Active noise control of a moving point has been examined in our previous papers. The feature of the model is that the transfer characteristics between the noise control source and the evaluation point (aimed noise reduction point) change with the movement of the evaluation point. Real time character identification algorithm have been introduced to handle the problem. However, the convergent algorithm used in the identification works effectively only when the system is stationary, and this is one of the reason that the identification accurcy decreases when the speed of the evaluation point increases. Besides, the evaluation point moves during the identification process of the calculation and it can cause discrepancy between identification results and the real system. We propose new technique where transfer characteristics are divided into time delay factor and main components. Once the characters are divided, the change of transfer characteristics caused by the movement appears mostly in time delay factor, and since it is the function of the location, it is easy to get the delay factor. Applying this idea, one can expect remarkable improvement in the identification. This method is examined numerically and experimentally in this paper.