Empirical power-consumption model for material removal in three-axis milling

• Energy consumption of a vertical milling machine was decomposed.
• Material removal power was empirically modeled with respect to the process parameters and tool wear.
• The difference between power under severe-wear and slight wear conditions was modeled.
• Rotational speed, feed, and depth of cut simultaneously affect on the material-removal power.
• Appropriate process parameters could be selected considering the tool wear states.

Environmental impact concerns have prompted various ecodirectives and legislation requiring manufacturing industries to take a greater role in reducing energy consumption and carbon emission. Research has indicated that monitoring and assessment of energy consumption is a basic approach that can be used to achieve energy savings. In this study, we decomposed the energy elements of a milling machine used for cutting to model additional material-removal power caused by the cutting load with respect to various process parameters. Because the material-removal power is directly related to the cutting force, current and power monitoring has been a good solution for indirect monitoring system. However, the effect of the process parameters on the material-removal power is more complex. In this study, the material-removal power consisted of ∼7.6% of the total power consumption of the machine tool, and increased with the flank wear of the tool. The difference between power under severe-wear and slight-wear conditions was empirically modeled using response surface methodology. The cutting power varied in terms of all process parameters, and its increase caused by the tool wear depended on the process conditions. Though both the material-removal power and its difference have a positive correlation with the material-removal rate, the absolute difference was not proportional to the material-removal power itself. With the constructed model, the overall energy consumption of the machine tool could be estimated more precisely, and the tool-wear states could be more accurately predicted in accordance with various process parameters.