Tool Fault Diagnosis System of Machining Center Based on Multi-sensor Data Fusion
Wenzhou Public Welfare Science and Technology Project
In this project, the on-line tool state recognition and prediction system of machining center is studied, and the method of tool damage feature extraction and damage state prediction based on steady subspace method under small samples is explored. In order to overcome the technical difficulty that the existing methods are not suitable for the data features of machining centers, a new efficient method for extracting inherent features of tool damage state is studied. Combined with machine learning theory and intelligent calculation method, a systematic on-line recognition and prediction model of tool state of CNC milling machines is constructed, and a prototype system for on-line monitoring of tool state is developed, which provides information support for monitoring and identifying tool state of CNC milling machines.
