基于独立成分分析算法的纺纱锭子噪声测试  

Testing spinning spindle noise based on independent component analysis algorithm

作　　者：刘晓芝[1,2] 杨建国[1,3] 李蓓智[1] 吕志军[1,2] 马胤琛[1,3] 

机构地区：[1]东华大学机械工程学院,上海201620 [2]东华大学纺织面料技术教育部重点实验室,上海201620 [3]东华大学教育部纺织装备工程技术研究中心

出　　处：《纺织学报》2011年第9期119-124,共6页Journal of Textile Research

基　　金：上海市重点学科建设资助项目(B602)

摘　　要：纺纱锭子性能的优劣对生产效率和纱线质量具有决定性的影响。以纺纱锭子运转中的噪声信号和驱动电动机单独运转的噪声信号为考察对象,对纺纱锭子运转过程进行分析。利用频谱分析和快速独立成分分析(FastICA)算法将纺纱锭子运转噪声信号的分离信号与驱动电动机单独运转的噪声信号主频率进行对比,38.88%分离主频率与驱动电机单独运转的噪声主频率一致,44.44%分离主频率与驱动电动机单独运转的噪声主频率相近,最小误差为0.05%。实验与分析结果表明,FastICA算法用于噪声源信号主频率分离的有效性和正确性。The performance of spinning spindles has decisive influence on production efficiency and yarn quality.The goal of this paper was to investigate the noise signals generated by the spinning spindle and driving motor during operation.The spectrum analysis method and fast independent component analysis algorithm（Fast ICA） were used to compare the noise signal frequency separated from the running spindle noise with the main frequency of the motor running noise,and the results indicated that 38.88% of the noise signal frequency separated from the running spindle noise was identical with the main frequency of the motor running noise,and 44.44% of the former showed close to the later,with the minimum error 0.05%.Experimental and analytical results show the effectiveness and correctness of FastICA algorithm in use for separation of the main frequency of noise source signals.

关 键 词：纺纱锭子 FASTICA算法 噪声 信号分离 

分 类 号：TN912.3[电子电信—通信与信息系统]