中文摘要：

为解决工程应用中切削参数一致的孔系加工质量一致性评估的难题,提出了一种基于振动信号特征波动可视化的聚类分析方法。首先采用振动传感器监控孔系钻削过程,提取各孔振动信号小波包能量谱和高阶统计量特征;然后利用雷达图得到各孔振动信号特征矩阵分布图,提取信号特征雷达图多边形重心特征;最后采用模糊C-均值（FCM）算法对雷达图平面重心点集进行聚类分析。理论分析结果与人工检测结果对比表明：该方法可直观呈现孔系钻削质量分布情况,简便、可靠地实现孔系钻削质量的一致性评估。

英文摘要：

To analyse the consistency evaluation of hole series part drilling quality,a clustering analysis method was presented based on fluctuation of vibration signal features.A vibration sensor was used to monitor the drilling process,and the vibration signal wavelet packet energy spectrum and higher-order statistics features of each hole in drilling process were extracted to construct the monitoring signal characteristic matrix.Then radar chart was used to reconstruct the distribution graphs of wavelet packet energy spectrum features and high-order statistics features,and the center of gravity features were extracted from these radar charts.Finally,fuzzy C-means（FCM）algorithm was used to complete the clustering analyses of these center of gravity features points.By comparison,it shows that hole drilling quality distribution can be visually presented and hole drilling quality consistency evaluation can be realized simply and reliably by clustering analysis based on radar chart.