Accurate Prediction of Machining Cycle Times by Data-driven Modelling of NC System's Interpolation Dynamics
Accurate prediction of cycle times of machining part programs plays a crucial role in process planning and part flow optimization on shop floors. This paper presents a data-driven approach to model the trajectory generation (interpolation) strategy embedded in the Numerical Control (NC) system of CNC machine tools to accurately predict their machining cycle times. Artificial Neural Networks (ANN) are trained to learn and accurately mimic how the CNC plans its feedrate profile as it accelerates, decelerates, and interpolates along complex part programs. Proposed approach is validated experimentally and shown to predict machining cycle times with >95% accuracy along tested complex machining toolpaths.