Estimation of grinding conditions by machine learning

Abstract

In ultra-precision grinding, it is necessary to consider the effect of grinding conditions on form error. However, even if the set value is an ideal value, in reality there will be an error from the ideal grinding conditions due to various reasons. Therefore, the purpose of this research is to construct a system that measures the error of grinding conditions and carry out compensation grinding. In this report, the grinding condition is estimated from the state of the ground surface using a machine learning technique.