Real-time acquisition of machining task progress based on the power feature of workpiece machining

Abstract

The real-time acquisition of machining task progress is one of the most important tasks of production management and is an essential aspect of manufacturing information. Targeted to the machining mode of mixed-category workpieces in job shops, a method for the acquisition of real-time machining task progress is proposed. The method is based on both the power feature of workpiece machining and the incremental learning Lagrangian support vector machine. First, the framework for this method is presented in a straightforward manner. Second, by analysing the characteristics of power change during the machining process, the power feature vector, which reflects the characteristics of workpiece machining, is designed for Lagrangian support vector machine. Then, based on the principle of incremental learning Lagrangian support vector machine, which can address the classification of mixed-category workpieces and the problem of an insufficient number of training samples for training the initial classifiers during the actual machining process, a detailed application of this method is constructed for workpiece classification and the acquisition of machining task progress. Finally, the effectiveness of this method is empirically tested by application to a case study.