Neural Networks for Active Noise Control-Using Frequency-Shaped Secondary-Path

• Main Authors: Chiung-Hung Cheng, 陳俊弘
• Other Authors: Jong-Yih Lin
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/90213718663177373458

碩士 === 國立中興大學 === 機械工程學系 === 93 === This thesis plans to first design a feedback controller for a secondary-path in an active noise control (ANC) system. The feedback controller is designed based on a nominal model of the secondary-path by use of H-infinity control theory. The controlled secondary-path is thus acquired a preliminary capability to reduce noise level and a dynamic characteristic close to that of an ideal secondary-path for later design. A neural network is then designed with a Back-Propagation algorithm based on the controlled secondary path to further reduce noise level. An intelligent ANC controller is finally formed by connected the feedback controller and the neural network in series. Results of computer simulation and experiment show that our proposed ANC controller has a good performance of noise reduction.