Noise Control for a Moving Evaluation Point Using Neural Networks

13th International Conference on Motion and Vibration Control; 12th International Conference on Recent Advances in Structural Dynamics

Maeda, Toshiki^1 ; Shiraishi, Toshihiko^1

Grad. Sch. of Environment and Information Sciences, Yokohama National University, 79-7 Tokiwadai, Hodogaya-ku, Yokohama

240-8501, Japan^1

关键词: Control performance;    Filtered-x LMS algorithms;    Learning abilities;    Low-Frequency Noise;    Moving evaluation point;    Overfitting;    Principle of superposition;    Time varying;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/744/1/012183/pdf
DOI  :  10.1088/1742-6596/744/1/012183

来源: IOP

【 摘 要 】

This paper describes the noise control for a moving evaluation point using neural networks by making the best use of its learning ability. Noise control is a technology which is effective on low-frequency noise. Based on the principle of superposition, a primary sound wave can be cancelled at an evaluation point by emitting a secondary opposite sound wave. To obtain good control performance, it is important to precisely identify the characteristics of all the sound paths. One of the most popular algorithms of noise control is filtered-x LMS algorithm. This algorithm can deliver a good result while all the sound paths do not change. However, the control system becomes uncontrollable while the evaluation point is moving. To solve the problem, the characteristics of all the paths are must be identified at all time. In this paper, we applied neural networks with the learning ability to the noise control system to follow the time-varying paths and verified its control performance by numerical simulations. Then, dropout technique for the networks is also applied. Dropout is a technique that prevent the network from overfitting and enables better control performance. By applying dropout for noise control, it prevents the system from diverging.