Comparative analysis of steel mechanical properties for generalized cutting energy prediction in CNC machining
Journal
International Journal of Advanced Manufacturing Technology
Journal Volume
140
Journal Issue
7-8
Start Page
4179
End Page
4190
ISSN
02683768
Date Issued
2025-10
Author(s)
Abstract
The continuous rise in global manufacturing demand has significantly increased electricity consumption, with Computer Numerical Control (CNC) machine tools identified as major energy consumers—yet often operating at efficiencies below 30%. Although numerous studies have investigated energy consumption in machining, clear definitions of energy sources and a comprehensive examination of the influence of mechanical properties on power consumption remain insufficient. To address these, this work analyzed the power behavior of spindle and feed axis motors and evaluated the impact of steel properties on cutting power, aiming to develop a generalized cutting power model. In particular, the predictive accuracy of such models varies significantly across different steel types, underscoring the importance of robustness and generalizability in evaluating model performance. This study systematically compares the correlations between cutting power and four material properties—yield strength, tensile strength, elongation, and hardness. Cutting power models based on each property were constructed using five types of steel: P1, SKD61, P20, P3, and NAK80. The proposed model was first validated using S50C and subsequently tested on 420J2 stainless steel, achieving a prediction accuracy of 97.6%. In addition, the model was evaluated under a variable cutting path designed to simulate realistic machining conditions, yielding an accuracy of 97.2%. These results collectively highlight the model’s strong generalizability across different steel types and cutting scenarios.
Subjects
CNC machining
Cutting power prediction
Energy consumption modeling
Milling process
Specific cutting energy (SCE)
Steel machining
Publisher
Springer Science and Business Media Deutschland GmbH
Type
journal article