Abstract:
In modern competitive manufacturing industry, machining processes are expected
to deliver products with high accuracy and good surface integrity. This should be
achieved through shorter production cycle times with reduced operator intervention
and increased
exibility. In order to accomplish this, the trend is towards increased
use of machine intelligence in machining processes. Grinding process is usually em-
ployed to machine harder materials, or, as a nishing process. A fast, accurate
and e cient grinding process contributes greatly to the productivity in a production
setup.
In the current work, a theoretical model was developed and used to predict the vi-
brations resulting from the grinding process. A controller based on adaptive neural
fuzzy inference system (ANFIS), was developed for the cylindrical grinding machine-
tool. The main aim of this study was to optimize the grinding process by adaptively
controlling the speed of the grinding wheel based on the infeed and the speed of the
workpiece. This would help in prevention of excessive vibrations that would a ect
the machining process, resulting in poorly nished surfaces and degraded grinding
wheel. Also, experimental work was carried out to validate the model.
From this study, it was demonstrated that, ANFIS based controller controlled vibra-
tions during grinding through in-process adjustment of speed of the grinding wheel
so that there would be minimal vibrations. The proposed controller was tested ex-
perimentally and was seen to be e ective in reducing the machining vibrations by as
much as 90 percent.
