时域-频域-非线性特征融合的钢丝圈转速变化监测与分析OA

The variation characteristics of traveler speed in different usage stages in ring spinning were investigated.A multi-dimensional time series analysis method was proposed,integrating time-domain,frequency-domain and nonlinear dynamical features of traveler speed.An independently developed inductive sensor was utilized to acquire real-time traveler speed data during its adaptation period,stabilization period and deterioration period.A fused feature set was constructed by combining time-domain statistical indicators,frequency-domain energy descriptors and recursive quantification analysis,and K-means algorithm was used to perform unsupervised clustering on the fused features.The results showed that the consistency rate of the clustering results with the usage stage of the traveler was 91.7%,and the interpretation rate of the fused features for the yarn hairiness was 66.7%.It is considered that extracting fused features from real-time traveler speed signals can provide a reliable basis for assessing the usage state of the traveler and yarn quality.