Scholarlink

This study investigated the prediction of end milling parameters on surface roughness during the symmetrical cryogenic end milling process of AISI D2 tool steel. The varied milling parameters are the flow rate of cryogenic cooling, cutting speed, feed rate, and axial depth of cut. The experimental design selected is an L18 orthogonal array based on the Taguchi method. Experiments were randomized entirely and repeated twice. The prediction methods applied were fuzzy inference system (FIS) Mamdani type 1 and Sugeno type 1, utilizing the Gaussian membership function. A genetic algorithm was selected to tune FIS. The smallest value of RMSE is applied to determine the best FIS for prediction. The result showed that FIS Sugeno type 1 with rules and output tuned had the smallest RMSE. The error between the predicted and measured values was only 0.99%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Surface Roughness Prediction in End Milling with Cooling Liquid Nitrogen on AISI D2 Tool Steel Material Using Fuzzy Inference System

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Muhammad Fiky Izzulhaq,
  • Arif Wahjudi,
  • B. O. P. Soepangkat

摘要

This study investigated the prediction of end milling parameters on surface roughness during the symmetrical cryogenic end milling process of AISI D2 tool steel. The varied milling parameters are the flow rate of cryogenic cooling, cutting speed, feed rate, and axial depth of cut. The experimental design selected is an L18 orthogonal array based on the Taguchi method. Experiments were randomized entirely and repeated twice. The prediction methods applied were fuzzy inference system (FIS) Mamdani type 1 and Sugeno type 1, utilizing the Gaussian membership function. A genetic algorithm was selected to tune FIS. The smallest value of RMSE is applied to determine the best FIS for prediction. The result showed that FIS Sugeno type 1 with rules and output tuned had the smallest RMSE. The error between the predicted and measured values was only 0.99%.