CNC 工具機廣泛應用於精密製造產業，惟長時間加工所造成之能源消耗將提高製造成本，並對低碳製造形成挑戰。既有 CNC 加工能耗預測研究多以 NC code 解析為基礎，可由加工指令推估機台狀態、加工時間與總能耗。然而，NC code 本身缺乏切削過程之幾何資訊，使模型對切削負載變化之描述仍有限。因此，本研究整合既有能耗預測架構、材料移除模擬與切削功率模型，以補足切削幾何資訊並改善加工能耗估測流程。
本研究以 NC block 作為能耗計算單元，從 NC block 擷取坐標、主軸轉速、進給率與運動型式等資訊，並將加工過程區分為待機、空轉與切削等狀態。待機功率、主軸空載功率、進給系統功率與周邊設備功率由空載實驗建立；切削附加功率則由 L27 切削實驗建立，並以切削深度 、切削寬度、主軸轉速與每齒進給量作為主要輸入。為取得 NC code 未提供之切削深度與切削寬度，本研究依坐標與 G 指令將刀具運動離散化，並透過材料移除模擬判定新移除材料，以估算切削幾何參數並導入能耗計算。
驗證實驗包含獨立特徵工件與複合特徵工件。獨立特徵案例之總能耗預測誤差介於 0.06% 至 3.69%，複合特徵案例則介於 1.29% 至 3.61%。在切削附加能耗估算方面，相較於 MRR 基準模型，本研究方法於獨立特徵與複合特徵案例之平均百分誤差分別由 11.56% 降至 2.08%、由 13.48% 降至 2.15%。結果顯示，本研究方法可維持合理之總能耗預測能力，並較能描述材料移除造成之切削附加負載差異，可作為加工前能耗估算與加工策略比較之參考。

CNC machine tools are widely used in precision manufacturing, but their energy consumption during long-duration machining increases manufacturing costs and poses challenges to low-carbon manufacturing. Existing NC code-based energy prediction methods estimate machine states, machining time, and total energy consumption from machining commands. However, NC code does not directly provide cutting geometry information, limiting its ability to describe cutting load variations. Therefore, this study integrates an existing energy prediction framework, material removal simulation, and a cutting power model to supplement cutting geometry information for machining energy estimation.
In this study, NC blocks are used as the basic units for energy calculation. Coordinates, spindle speed, feed rate, and motion type are extracted from NC code, and the machining process is classified into standby, air-cutting, and cutting states. The standby, no-load spindle, feed-axis, and peripheral equipment power models are established from no-load experiments, while the cutting-induced additional power model is obtained from L27 cutting experiments. Since cutting depth ap and cutting width ae are not directly provided by NC code, tool motion is discretized according to coordinates and G-code commands, and voxel-based material removal simulation is used to estimate the cutting geometry parameters for energy calculation.
Validation was conducted using independent-feature workpieces, and composite-feature workpieces. The total energy prediction errors ranged from 0.06% to 3.69% for independent-feature cases and from 1.29% to 3.61% for composite-feature cases. For cutting-induced additional energy estimation, the proposed method reduced the average percentage error from 11.56% to 2.08% for independent-feature cases and from 13.48% to 2.15% for composite-feature cases compared with the MRR baseline model. These results indicate that the proposed method provides reasonable total energy prediction and better represents the additional cutting load caused by material removal.

[1] HEIDENHAIN. (2011) Aspects of Energy Efficiency in Machine Tools.
[2] T. Xia et al., "Efficient Energy Use in Manufacturing Systems—Modeling, Assessment, and Management Strategy," Energies, vol. 16, no. 3, 2023, doi: 10.3390/en16031095.
[3] W. Cai et al., "A review on methods of energy performance improvement towards sustainable manufacturing from perspectives of energy monitoring, evaluation, optimization and benchmarking," Renewable and Sustainable Energy Reviews, vol. 159, 2022, doi: 10.1016/j.rser.2022.112227.
[4] B. Denkena, E. Abele, C. Brecher, M.-A. Dittrich, S. Kara, and M. Mori, "Energy efficient machine tools," CIRP Annals, vol. 69, no. 2, pp. 646-667, 2020, doi: 10.1016/j.cirp.2020.05.008.
[5] G. Y. Zhao, Z. Y. Liu, Y. He, H. J. Cao, and Y. B. Guo, "Energy consumption in machining: Classification, prediction, and reduction strategy," Energy, vol. 133, pp. 142-157, 2017, doi: 10.1016/j.energy.2017.05.110.
[6] V. A. Balogun and P. T. Mativenga, "Modelling of direct energy requirements in mechanical machining processes," Journal of Cleaner Production, vol. 41, pp. 179-186, 2013, doi: 10.1016/j.jclepro.2012.10.015.
[7] F. Liu, J. Xie, and S. Liu, "A method for predicting the energy consumption of the main driving system of a machine tool in a machining process," Journal of Cleaner Production, vol. 105, pp. 171-177, 2015, doi: 10.1016/j.jclepro.2014.09.058.
[8] X. Chen, C. Li, Y. Tang, L. Li, Y. Du, and L. Li, "Integrated optimization of cutting tool and cutting parameters in face milling for minimizing energy footprint and production time," Energy, vol. 175, pp. 1021-1037, 2019, doi: 10.1016/j.energy.2019.02.157.
[9] Z. Jiang, D. Gao, Y. Lu, L. Kong, and Z. Shang, "Electrical energy consumption of CNC machine tools based on empirical modeling," The International Journal of Advanced Manufacturing Technology, vol. 100, no. 9-12, pp. 2255-2267, 2018, doi: 10.1007/s00170-018-2808-x.
[10] C. Feng, X. Chen, J. Zhang, and Y. Huang, "A generalized analysis of energy saving strategies through experiment for CNC milling machine tools," The International Journal of Advanced Manufacturing Technology, vol. 117, no. 3-4, pp. 751-763, 2021, doi: 10.1007/s00170-021-07787-9.
[11] Y. Xiao, Z. Jiang, Q. Gu, W. Yan, and R. Wang, "A novel approach to CNC machining center processing parameters optimization considering energy-saving and low-cost," Journal of Manufacturing Systems, vol. 59, pp. 535-548, 2021, doi: 10.1016/j.jmsy.2021.03.023.
[12] H.-S. Yoon, E. Singh, and S. Min, "Empirical power consumption model for rotational axes in machine tools," Journal of Cleaner Production, vol. 196, pp. 370-381, 2018, doi: 10.1016/j.jclepro.2018.06.028.
[13] Q. Xiao, C. Li, Y. Tang, Y. Du, and Y. Kou, "Deep Learning Based Modeling for Cutting Energy Consumed in CNC Turning Process," presented at the 2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2018.
[14] S. Pawanr, G. K. Garg, and S. Routroy, "Prediction of energy consumption of machine tools using multi-gene genetic programming," Materials Today: Proceedings, vol. 58, pp. 135-139, 2022, doi: 10.1016/j.matpr.2022.01.156.
[15] Q. Xiao, C. Li, Y. Tang, and X. Chen, "Energy Efficiency Modeling for Configuration-Dependent Machining via Machine Learning: A Comparative Study," IEEE Transactions on Automation Science and Engineering, vol. 18, no. 2, pp. 717-730, 2021, doi: 10.1109/tase.2019.2961714.
[16] V. S. Vishnu, K. G. Varghese, and B. Gurumoorthy, "Energy Prediction in Process Planning of Five-axis Machining by Data-driven Modelling," Procedia CIRP, vol. 93, pp. 862-867, 2020, doi: 10.1016/j.procir.2020.04.087.
[17] W. Li et al., "A novel milling parameter optimization method based on improved deep reinforcement learning considering machining cost," Journal of Manufacturing Processes, vol. 84, pp. 1362-1375, 2022, doi: 10.1016/j.jmapro.2022.11.015.
[18] G. Kant and K. S. Sangwan, "Predictive Modelling for Energy Consumption in Machining Using Artificial Neural Network," Procedia CIRP, vol. 37, pp. 205-210, 2015, doi: 10.1016/j.procir.2015.08.081.
[19] Y. He, P. Wu, Y. Li, Y. Wang, F. Tao, and Y. Wang, "A generic energy prediction model of machine tools using deep learning algorithms," Applied Energy, vol. 275, 2020, doi: 10.1016/j.apenergy.2020.115402.
[20] N. Shen, Y. Cao, J. Li, K. Zhu, and C. Zhao, "A practical energy consumption prediction method for CNC machine tools: cases of its implementation," The International Journal of Advanced Manufacturing Technology, vol. 99, pp. 2915–2927, 2018, doi: 10.1007/s00170-018-2550-4.
[21] S. Pavanaskar and S. McMains, "Machine Specific Energy Consumption Analysis for CNC-Milling Toolpaths," presented at the ASME 2015 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, 2015. [Online]. Available: https://doi.org/10.1115/DETC2015-48014.
[22] T. Kablay and M. Namoshe, "Development of a novel CNC energy consumption model and energy analyzer software: towards energy-efficient tool pathway strategies in CNC machining," Discover Mechanical Engineering, vol. 4, no. 1, p. 18, 2025.
[23] J. Cao, X. Xia, L. Wang, Z. Zhang, and X. Liu, "A novel CNC milling energy consumption prediction method based on program parsing and parallel neural network," Sustainability, vol. 13, no. 24, p. 13918, 2021.
[24] A.-M. Schmitt, E. Miller, B. Engelmann, R. Batres, and J. Schmitt, "G-code evaluation in cnc milling to predict energy consumption through machine learning," Advances in Industrial and Manufacturing Engineering, vol. 8, p. 100140, 2024.
[25] A.-M. Schmitt, E. Miller, A. Schiffler, and J. Schmitt, "Energy Prediction for CNC Machines Using G-Code Evaluation, Machine Learning and a Real-World Training Part," presented at the 2025 11th International Conference on Mechatronics and Robotics Engineering (ICMRE), 2025.
[26] M. Brillinger, M. Wuwer, M. Abdul Hadi, and F. Haas, "Energy prediction for CNC machining with machine learning," CIRP Journal of Manufacturing Science and Technology, vol. 35, pp. 715-723, 2021, doi: 10.1016/j.cirpj.2021.07.014.
[27] D. Kong, S. Choi, Y. Yasui, S. Pavanaskar, D. Dornfeld, and P. Wright, "Software-based tool path evaluation for environmental sustainability," Journal of Manufacturing Systems, vol. 30, no. 4, pp. 241-247, 2011, doi: 10.1016/j.jmsy.2011.08.005.
[28] I. F. Edem and P. T. Mativenga, "Impact of feed axis on electrical energy demand in mechanical machining processes," Journal of Cleaner Production, vol. 137, pp. 230-240, 2016, doi: 10.1016/j.jclepro.2016.07.095.