The prediction of computer numerical control (CNC) machining time critically impacts productivity. Computer-aided manufacturing (CAM) software typically has machining time prediction capabilities, allowing users to know the machining time for parts during toolpath planning. However, CAM software does not consider the kinematics of the CNC machines and the control principle of the CNC controllers. Hence, the predicted machining time is often much shorter than the actual one. To address this problem, we developed two neural network-based machining time prediction models for milling operations using MATLAB and TensorFlow. The results show that using the models proposed in this research could achieve prediction errors within 2%, while the CAM software had about 12% error.
Paper
Full text
Accurate Prediction of CNC Machining Time for Milling Operations Using Neural Networks
Semantic Scholar · Engineering · 2024
Abstract
The prediction of computer numerical control (CNC) machining time critically impacts productivity. Computer-aided manufacturing (CAM) software typically has machining time prediction capabilities, allowing users to know the machining time for parts during toolpath planning. However, CAM software does not consider the kinematics of the CNC machines and the control principle of the CNC controllers. Hence, the predicted machining time is often much shorter than the actual one. To address this problem, we developed two neural network-based machining time prediction models for milling operations using MATLAB and TensorFlow. The results show that using the models proposed in this research could achieve prediction errors within 2%, while the CAM software had about 12% error.