著者

小池 関也 小林 優揮 飯田 裕 下嶋 浩

出版者

一般社団法人日本機械学会

雑誌

日本機械学会論文集. C編 (ISSN:03875024)

巻号頁・発行日

vol.67, no.654, pp.321-328, 2001-02-25

参考文献数

10

被引用文献数

2

---

This paper proposes an active noise controller applicable to a moving evaluation point in one dimensional acoustic field. A combined adaptive feedforward/feedback controller based on filtered-x LMS algorithm is proposed. The feedforward controller is utilized for the noise reduction, and the feedback controller is utilized for the reduction of howling, which occurs with use of acoustic reference signal at feedforward controller. For the sake of stable noise control under moving evaluation point, necessity for a real time identificator of transfer function of the error path, which uses a microphone fixed in the neighborhood of the control speaker, is shown. Stable noise reduction without howling is realized by using above combined controller and the real time identificator with use of acoustic reference signal aiming practical use. A straight duct is chosen as a one dimensional acoustic field. Computer simulations and experimental analyses are carried out to clarify the validity of the proposed controller.

著者

小池 関也 小林 優揮 飯田 裕 下嶋 浩

出版者

一般社団法人日本機械学会

雑誌

日本機械学會論文集. C編 = Transactions of the Japan Society of Mechanical Engineers. C (ISSN:03875024)

巻号頁・発行日

vol.67, no.660, pp.2521-2527, 2001-08-25

参考文献数

13

被引用文献数

2

---

A basic noise control system applicable to a moving evaluation point was proposed in the previous report, and its effectiveness was examined numerically and experimentally. Although the results showed that the proposed system is effective, it can only handle the case of slow evaluation point velocity. This restriction seemed to be due to the slow convergent speed of the adaptive filter. In this paper the cause of restriction is examined in detail, and an advanced control system is proposed. The features of the proposed system are as follows : 1) Improved feedback controller where main filter is less affected by the movement of the evaluation point. 2) Introduction of IIR filter together with the learning identification method to the error path identifier. The new system is examined both in simulation and experiment and shows great improvement compared with the basic model.