电脑风扇噪声主动降噪关键技术研究

发布时间：2018-08-06 09:09

【摘要】：随着互联网技术的飞速发展,电脑已经成为了人们日常生活和工作中不可或缺的一部分,随之而来的电脑风扇噪声也越来越受到人们的重视。传统的被动降噪方法在短时间内可以起到一定的降噪效果,但是实际中环境温度的变化、风扇上灰尘的积累、风扇的老化等均会使降噪效果逐渐变差。而主动噪声控制(Active Noise Control)可以通过自适应算法实时地响应风扇噪声特性的变化,且对于低频窄带噪声有着很好的降噪效果,故ANC技术在电脑风扇噪声的控制上将会有较大的应用前景。本文主要研究如何设计ANC系统,确立其一般的设计流程以及在设计过程中所要解决的关键问题有哪些。通过研究,ANC系统的一般设计流程为：首先分析源噪声特性,然后根据分析结果选择合适的通道特性及控制特性。其中通道特性主要包括管道的尺寸、次级通道数量、电声设备的空间分布,控制特性主要包括控制结构、控制算法、算法参数的选择、误差通道的辨识。最后根据通道特性及控制特性选择合适的硬件设备。文中主要研究了管道的设计、自适应的控制算法及自适应滤波器、传感器和扬声器的空间布置以及声通道的自适应辨识。其中管道的设计依据声波导管理论以及扬声器和误差传感器位置的优化；声通道的辨识采用可控的离线自适应辨识方法。在控制算法上本文研究了三种算法,考虑到电脑风扇噪声的特性以及管道尺寸较小,采用FbLMS算法作为ANC系统的控制算法,然后通过Matlab仿真确定算法参数对降噪效果的影响,并得到如下结论：收敛系数越大,系统收敛速度越快,但过大的收敛系数回事系统发散；滤波器长度越大,收敛速度越快,但会增加数据的吞吐量；算法对于窄带噪声有很好的降噪效果,但随着带宽的增加,降噪效果明显下降；误差通道传递函数的估计可以有一定的误差,但不宜过大,否则系统发散。最后为了验证主动消声理论与控制算法,搭建了硬件试验平台,并对模拟的正弦噪声进行降噪,取得了一定的降噪效果。
[Abstract]:With the rapid development of Internet technology, computers have become an indispensable part of people's daily life and work, and the noise of computer fans has been paid more and more attention. The traditional passive noise reduction method can play a certain effect in a short period of time, but the change of environment temperature, the accumulation of dust on the fan and the aging of the fan will make the noise reduction effect worse gradually. The active noise control (Active Noise Control) can respond to the fan noise characteristics in real time by adaptive algorithm, and has a good noise reduction effect for low frequency narrow band noise. Therefore, ANC technology will have a great application prospect in the noise control of computer fan. This paper mainly studies how to design ANC system, how to establish its general design flow and what are the key problems to be solved in the design process. The general design flow of ANC system is as follows: firstly, the source noise characteristics are analyzed, and then the appropriate channel characteristics and control characteristics are selected according to the analysis results. The channel characteristics mainly include the size of pipeline, the number of secondary channels, the spatial distribution of electro-acoustic equipment, and the control characteristics mainly include control structure, control algorithm, algorithm parameter selection, error channel identification. Finally, according to the channel characteristics and control characteristics, select the appropriate hardware equipment. In this paper, the design of pipeline, adaptive control algorithm and adaptive filter, space arrangement of sensor and loudspeaker and adaptive identification of acoustic channel are studied. The design of the pipeline is based on the acoustic tube theory and the optimization of the speaker and error sensor position, and the sound channel identification adopts the controllable off-line adaptive identification method. Considering the noise characteristics of computer fan and the small size of pipeline, FbLMS algorithm is used as the control algorithm of ANC system, and then the effect of algorithm parameters on noise reduction is determined by Matlab simulation. The conclusions are as follows: the larger the convergence coefficient, the faster the system convergence speed, but the larger the convergence coefficient, the faster the convergence speed is, but the faster the data throughput is, and the larger the filter length is, the faster the convergence speed is. The algorithm has a good noise reduction effect for narrow band noise, but with the increase of bandwidth, the noise reduction effect is obviously reduced; the estimation of the error channel transfer function can have some errors, but not too large, otherwise the system diverges. Finally, in order to verify the active noise theory and control algorithm, a hardware test platform is built, and the simulated sinusoidal noise is de-noised, and the noise reduction effect is achieved.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP306.2;TB535.1

【参考文献】