机械制造与自动化2025，Vol.54Issue(3)：96-99,104,5.DOI:10.19344/j.cnki.issn1671-5276.2025.03.019
应用t-SNE流形学习方法的盾构刀盘磨损信号降维
Dimension Reduction of Shield Cutterhead Wear Signal by t-SNE Manifold Learning Method
摘要
Abstract
In order to enhance the ability of monitoring the running stability of shield cutterhead,a method for dimensionality reduction and performance evaluation of torque vibration signal of shield cutterhead based on t-SNE manifold learning was designed.According to the actual parameters of shield equipment,the identification of cutter state parameters was realized by data driven.The results show that compared with ISOMap,LLE,KPCA,etc.,using t-SNE manifold dimensionality reduction can obtain a longer t-distribution,so that the low-dimensional far end points will have a larger low-dimensional interval,and can accurately classify normal and degenerate samples,and accurately identify high-dimensional data containing low-dimensional manifold.The variation of normal operating conditions and fault maintenance sampling interval in Markov space is small,and the use of time-dependent Markov distance measure to evaluate tool performance can obtain higher precision,effectively eliminate the correlation between different dimensions,and have better performance than Euclidean distance.关键词
盾构刀盘/性能评估/t-分布随机邻域嵌入/信号降维Key words
cutter head of shield tunneling/performance evaluation/t-distributed random neighborhood embedding/signal dimension reduction分类
交通运输引用本文复制引用
黄玉华,李峰..应用t-SNE流形学习方法的盾构刀盘磨损信号降维[J].机械制造与自动化,2025,54(3):96-99,104,5.基金项目
广西壮族自治区自然科学基金项目(GXNSFBA297163) （GXNSFBA297163）
