부산대학교 도서관

As the main component of the CNC cutting process, the tool directly affects the dimensional accuracy of the machined workpiece and the reliable operation of the machine. According to studies, the prediction of degradation trends is usually based on complex models with nonlinear, non-Gaussian and long-range dependence (LRD). The novelty of this paper is that a prediction model is proposed to achieve the prediction of degradation trends by using fractional Levy-stable motion (fLsm), and using Lyapunov coefficients to determine the maximum prediction range of the degradation sequence and to ensure the reasonableness of the prediction results. Finally, the performance of the proposed method is demonstrated by comparison with support vector machine (SVM) using multi-sensor vertical mill monitoring data from the 2010 Prognostics and Health Management (PHM) Challenge.