Title: A study of grinding wheel sharpness using neural network and fuzzy logic approaches

Authors: Hamid Baseri

Addresses: Department of Mechanical Engineering, Babol University of Technology, Shariati Street, POB 484 Babol, Iran

Abstract: In this study, two different models were developed to estimate the skewness factor of an alumina grinding wheel which is a criterion for wheel sharpness. Development of the models is based on the back propagation neural network (BPNN) and the fuzzy logic approaches. Some experiments have been performed to train the BPNN to get it to estimate the wheel sharpness. Also, all membership functions and rule-bases have been presented in details for a fuzzy model. In experiments procedure, the grits of an alumina grinding wheel is dressed using a rotary diamond disc dresser. The predicted values of the skewness factor by using the neural network represented that this model has an acceptable precision in estimation of wheel sharpness. Also well acceptable performance of the fuzzy model was achieved. Comparison of two models shows that they were successful in estimation of the wheel sharpness.

Keywords: BP neural networks; BPNN; fuzzy logic; rotary diamond disc dressing; wheel sharpness; alumina grinding wheels; fuzzy modelling.

DOI: 10.1504/IJAT.2010.036965

International Journal of Abrasive Technology, 2010 Vol.3 No.4, pp.316 - 337

Received: 16 Apr 2010
Accepted: 13 May 2010
Published online: 18 Nov 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article