G-Code Based Toolpath Simulation for Predicting CNC Energy Consumption

For my undergraduate thesis, I developed a G-code based program in MATLAB that simulates the material removal process and predicts the energy and power consumption required to CNC machine a part. The simulation was validated by comparing the simulated energy to that of a part in both aluminum and titanium, showing an energy prediction error of 2.11% and 3.48% respectively.

Machined validation part in titanium

Empirical and predicted power vs time for machining the validation part in aluminum separated by different operations in the toolpath

Empirical and predicted power vs time for machining the validation part in titanium separated by different operations in the toolpath

Empirical and predicted energy vs time for aluminum (left) and titanium (right)

This program functions by first reconstructing the toolpath from the G-code.

The simulation then runs a virtual endmill along the toolpath in a voxel mesh of the workpiece, replicating material removal.

The model predicts the power and specific energy consumption for a timestep by relating it to the simulated material removal rate and spindle speed using existing analytical models and machine-dependant power data. An example relation is shown below.

The full thesis with comparisons and sensitivity analyses can be found here.