Tool condition monitoring Based on Mel-frequency cepstral coefficients and support vector regression

Manufacturing process performances become a key issue for reliability improvement. In order to decreasing loss of production due to machine stopping, to achieve this goal to maintain the equipments in operational condition, make products quickly, economically and with high quality. This requirement can be satisfied by implementing appropriate maintenance strategies, Condition based maintenance (CBM) used condition monitoring technologies for health assessment and remaining useful life (RUL) estimation of equipment working under different operating conditions. To achieve this goal; A new method for tool wear condition monitoring based on Mel-frequency cepstral coefficients(MFCC) and support vector regression (SVR); The raw signals are first processed to extract the MFCC features, which are then exploited to learn models that represent the evolution of the cutting tool's degradation. SVR is a learning technique where the goodness of fit is measured; the idea is based on the computation of a nonlinear regression function in a high dimensional feature space where the input data are mapped via a nonlinear function. The proposed method is applied on real world cutting tools degradation. The experimental results show that the health indicator can reflect effectively the tool wear degradation.