基于系统聚类法的空调钣金振动信号的分类方法

The Classification Method of Vibration Signal From Air Conditioning Sheet Metal Based on Hierarchical Clustering Method

为解决静音家电振动检测存在的问题对稳定运行于一个特定工作状态的空调外机外壳的钣金结构进行多点的振动信号采集,提取采集所得的时域波形的均方值及峭度特征,并选取两个包含不同频率成分且具有典型分类特征的振动信号进行傅里叶变换,将所得的两个频域数据分别与其他点的振动信号的频域数据进行相关系数计算.再将所得的相关系数与均方值和峭度作为分析的指标,利用系统聚类法对振动信号进行分类.实验结果表明:该聚类方法对振动信号的分类有着明显的效果,能较好地解决传统的家电振动检测方法所存在的问题.

To solve the problem of vibration testing of Mute appliances,this paper describes the vibration signals acquisition on the stable working air-conditioning,The characteristics include RMS（Root Mean Square）and Kurtosis feature are extracted,selected signals which contain different two frequency components and typical characteristics of the classification of the signals conduct a Fast Fourier transform to calculate their correlation coefficient with others in frequency domain.Finally selecting the RMS and Kurtosis feature and the correlation coefficient as the index of Hierarchical Clustering Method to classify the signals.The experimental results show that the clustering method has a significant effect on the classification of vibration signals,and it can solve the problems of the traditional vibration testing in home appliances.