The development of in-process surface roughness prediction systems in turning operation using accelerometer

Abstract

Three in-process surface roughness prediction (ISRP) systems using linear multiple regression, fuzzy logic, and fuzzy nets algorisms, respectively, were developed to allow the prediction of real time surface roughness of a work piece on a turning operation. The surface roughness is predicted from feed rate, spindle speed, depth of cut. and machining vibration that is detected and collected by an accelerometer. A multiple regression based ISRP system was developed and examined. Feed rate, spindle speed, depth of cut and vibration average amplitude were employed as independent variables to predict surface roughness of the work piece in a turning operation in a real time fashion. Two groups of data were collected for two cutters with nose radii of 0.016 and 0.031 inches, respective. A total of 162 training data sets and 54 testing data sets for each cutter were applied to train and test the system. This ISRP system reached a prediction accuracy of averagely 92.78% in predicting the in-process surface roughness. The fuzzy logic modeling methodology was practiced to develop another ISRP system. Expert's experiences and experimental observations were the major sources to the generation of the fuzzy rule bank. By applying the fuzzy logic algorism, this system reached an average prediction accuracy of 89.06%. A Fuzzy-Nets-based in-process surface roughness prediction (FISRP) system was developed as the third to predict surface roughness in a turning operation in a real time on­ line fashion. The input variables of the FISRP system were machining parameters, such as feed rate, spindle speed, depth of cut. and average vibration amplitude per revolution. Fuzzy nets theory has been implemented to use experimental data in developing this FISRP system for real time prediction. The fuzzy nets theory consists of a five-step learning mechanism of developing knowledge base for predicting surface roughness on-line in real time. This FISRP system was developed and tested via experiments, and has been tested to have an average accuracy of 95.70%. The Fuzzy-nets-based In-process Surface Roughness Prediction System was considered the best among the three tested systems. This conclusion relies on not only the best average prediction accuracy achieved, but also the self-learning ability of the fuzzy nets algorism. In conclusion, the developed in-process surface roughness prediction system is capable to predict the in-process surface roughness of the work piece in a turning operation from feed rate, spindle speed, depth of cut and average vibration amplitude. With the developed Fuzzy-nets-based In-process Surface Roughness Prediction System, the prediction can reach an average accuracy of 95.70%.