Tool condition monitoring of aluminum oxide grinding wheel using AE and fuzzy model

Abstract

The grinding process is situated at the end of the machining chain, where geometric and dimensional characteristics and high-quality surface are required. The constant use of cutting tool (grinding wheel) causes loss of its sharpness and clogging of the pores among the abrasive grains. In this context, the dressing operation is necessary to correct these and other problems related to its use in the process. Dressing is a reconditioning operation of the grinding wheel surface aiming at restoring the original condition and its efficiency. The objective of this study is to evaluate the surface regularity and dressing condition of the grinding wheel in the surface grinding process by means of digital signal processing of acoustic emission and fuzzy models. Tests were conducted by using synthetic diamond dressers in a surface grinding machine equipped with an aluminum oxide grinding wheel. The acoustic emission sensor was attached to the dresser holder. A frequency domain analysis was performed to choose the bands that best characterized the process. A frequency band of 25–40 kHz was used to calculate the ratio of power (ROP) statistic, and the mean and standard deviation values of the ROP were inputted to the fuzzy system. The results indicate that the fuzzy model was highly effective in diagnosing the surface conditions of the grinding wheel.