Transition-based Soft Phase Partition Algorithm of Multiphase Batch Processes Based on K-means Optimal Clustering

In the multi-period batch process, there are many time periods, and the fuzzy transition areas corresponding to the process characteristics of adjacent areas from one period to another cannot be strictly divided. If it is impossible to accurately classify the transition period, misjudgment is likely...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Chinese Control Conference S. 5806 - 5811
Hauptverfasser: Shao, Mengya, Lyu, Feng, Sun, Hao, Du, Wenxia, Guo, Zhenxing
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: Technical Committee on Control Theory, Chinese Association of Automation 01.07.2018
Schlagworte:
Accuracy
Clustering algorithms
Fault diagnosis
K-means
Machine tools
Monitoring
MPCA
optimal clustering
Partitioning algorithms
phase transition mode
Principal component analysis
Production
Reliability theory
Uncertainty
ISSN:1934-1768
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In the multi-period batch process, there are many time periods, and the fuzzy transition areas corresponding to the process characteristics of adjacent areas from one period to another cannot be strictly divided. If it is impossible to accurately classify the transition period, misjudgment is likely to occur when using the multi-way principal component analysis (MPCA) method. In order to make the fault diagnosis of the complex system in industrial production more accurate, this paper proposes a method based on the best clustering of k-means to divide the soft period. On the basis of the traditional phase division, this method adds the theory of K-means best cluster phase transition interval screening. And this method integrates the soft-time segmentation algorithm to improve the reliability of the sub-period division so as to make the transition period more accurate. The effectiveness of the method is proved by the simulation experiment of machine tool processing data and by compared with other methods.
ISSN:1934-1768
DOI:10.23919/ChiCC.2018.8482829