Development of smart machining

• Main Author: Seah, Yee Loong
• Other Authors: Yeo Swee Hock
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Manufacturing; Engineering::Aeronautical engineering
• Online Access: https://hdl.handle.net/10356/141412
• Institution: Nanyang Technological University

Smart Machining has been extremely popular in the manufacturing industry since its debut. This caused companies to leverage on technology capabilities, leading to the automation and intelligence known as Industry 4.0. The Mazak CNC QuickTurn 250 machine was equipped with sensors which record data such as the Tri-Axial Cutting Force, Tri-Axial Acceleration, Cutting Temperature, Coolant Pressure, Power, and Acoustic Emission. Aluminium was used as the main material for cutting experiments and the relationship between the variables was studied. This can be further expanded to cover different working materials with minimal modifications. In addition to the sensor data, the CNC machine provided data such as the feed rate and cutting speed. Surface roughness readings were also recorded using a surface roughness tester and through experiments, it was proven that this is affected by the cutting speed, feed rate, and coolant pressure. Data analytics and Machine Learning were subsequently done to generate a regression model that was able to predict the cutting force and surface roughness based on the dependent variables. Additionally, Decision Trees, Supported Vector Machine, and Neural Networks algorithms were built, which could classify between a sharp and worn cutting tool up to a 90% accuracy.