Chih-Hsing
Chu
Sponsored
By: CODEF
Abstract
Experimental
studies show that burr formation is a highly complex process depending
on a number of parameters such as material properties, tool geometry,
the depth of cut, cutting speed, feed rate and exit angle. Not only
affect these factors on burr type as well as burr size, but they also
influence each other. This work utilized machine learning techniques
for burr type prediction. With the capability of feature weighting,
those approaches may be able to enhance the understanding and predictability
of burr formation. They also provide a feasible approach to querying
in database systems consisting of experimental data in an accumulated
way.
View the entire report:
MS Word
PDF
Go Back
to the 1998 Index
