基于多模型的自适应有源噪声控制算法研究  

An adaptive active noise control algorithm using multiple models

作　　者：赵剑[1] 徐健[1] 李晓东[1] 田静[1] 

机构地区：[1]中国科学院声学研究所,北京100080

出　　处：《振动工程学报》2007年第6期549-555,共7页Journal of Vibration Engineering

摘　　要：针对有源噪声控制中次级通道存在很大不确定性的问题,建立次级通道模型集以覆盖次级通道的不确定性,每个模型都有一个相对应的采用Modified FXLMS算法进行权值更新的自适应滤波器。在系统运行过程中,算法首先依据切换准则实时判断当前次级通道所处状态,选出最匹配的次级通道模型,同时把采用最匹配模型进行权值更新的自适应滤波器的参数复制给参与控制的滤波器。仿真和实验结果表明,采用多模型自适应的有源噪声控制算法提高了系统的鲁棒性。Multiple models are set up to cover the uncertainty of the real secondary path to deal with a grand uncertain secondary path problem in active noise control （ANC）. For each model there is a corresponding adaptive filter, whose coefficients are updated using the Modified FXLMS algorithm. During the operating process, firstly the algorithm determines the specific situation of the real secondary path and chooses the best one among those models according to a switching rule, and then the corresponding adaptive filter＇s coefficients are copied into the actual control filter to implement the ANC system. Computer simulations and experimental results show that the proposed ANC algorithm improves the robustness of the ANC system.

关 键 词：有源噪声控制 自适应控制 多模型 次级通道 切换准则 

分 类 号：TB535[理学—物理] TP273.2[理学—声学]