Design Of An On‐Line Diagnostic System Of Finished Surfaces In An Automated Mechanized Process By CNC Turning

The primary aim of this study was to design and develop an on‐line diagnostic system for the real‐time prediction of finished surfaces (Ra) in automated mechanized process by turning CNC. In the first phase the characterization parameters of cutting‐force signals (Fx, Fy, Fz) in the time and the frequency domain were assessed. Thereafter, the parameters that significantly influence finished surfaces and underpin predicted models based on regression techniques have been evaluated. The global monitoring system was developed with the Dasylab 9.0 (Labview family) that enables: the visual display of data; signal acquisition, processing and analysis; and the real‐time prediction of finished surfaces (Ra). Thus, the system predicts the Ra parameter with a relative error percentage below 5.8%, which is similar to the precision obtained in post‐process quality control using a stylus contact profilometer.