In-process stochastic tool wear identification and its application to the improved cutting force modeling of micro milling

Micro milling aims to manufacture miniature structures with high quality and complex features, the stochastic time-varying tool wear is a crucial factor which has great influence on machining efficiency of micro process. To improve precision sustainability cutting tools, in-process conditions should be identified updated ahead time. In this work, an improved integrated estimation method proposed based long short-term memory (LSTM) network particle filter (PF) algorithm predict values. The PF-LSTM identification methodology developed progression basis historical measurement data. With wear, force model modified, in run-out trochoidal trajectory edge are also considered. modified were validated by experiments workpiece material Al6061. It can seen from comparison results that availability have been improved, prediction accuracy could increased 3.4% compared without considering wear.