STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Production Research & Management

Regression and ANN Models in Predicting Tool Wear

  1. Mafiz Alji
  2. Zain Bamne
  3. Atique Kondvilkar
  4. Umer Aklekar
  5. Paramjit Thakur

Abstract

A modern machining system must be able to detect tool wear while milling in order to maintain the product's surface quality. The vibration signatures produced by a single point cutting tool during machining have been found to be good predictors of the tool's health. The current study used Artificial Neural Networks to forecast tool life by analysing vibration signatures when turning EN9 and EN24 steel alloys (ANN). Tool wear prediction is critical in the business for increased productivity and product quality. Cutting tool flank wear is frequently used as a tool life criterion because it determines the diametric accuracy, stability, and dependability of machining. The main goal of this project is to extend the life of the cutting tool, resulting in a better surface finish, higher quality, and a reduction in manufacturing costs and time. In this project, we will forecast tool wear using two different models: regression mathematics and artificial neural network (ANN) models. The response (output) variable measured during milling in this study is tool wear (flank wear/crater wear), while the input variables are cutting speed, feed, and depth of cut.

Keywords

Support