A fuzzy-nets-based in-process surface roughness prediction system in turning operations

Abstract

A Fuzzy-Nets-based in-process surface roughness prediction (FISRP) system was developed to predict surface roughness in turning operations in a real time fashion. The input variables of the FISRP system were machining parameters, such as feed rate, spindle speed, depth of cut, and machining vibration per revolution. An accelerometer was employed to gather real-time vibration signals. Two groups of data were collected for two cutter bits with nose radii of 0.016 and 0.031 inches, respectively. Fuzzy nets theory was implemented to use the experimental data in developing the system for real-time prediction. The fuzzy nets theory is a five-step learning procedure for developing a knowledge base to predict surface roughness in real time. This FISRP system was tested to have an average prediction accuracy of 95.70%.