Tolcha Mesay Alemu, Lemu Hirpa Gelgele, Adugna Yosef Wakjira
Jimma Institute of Technology, Jimma University, Jimma, Ethiopia.
Department of Mechanical and Structural Engineering, University of Stavanger, Stavanger, 4036, Norway.
Sci Rep. 2025 Mar 25;15(1):10215. doi: 10.1038/s41598-025-95446-4.
This study investigates the machinability of a novel LM25 aluminum alloy reinforced with vanadium carbide composite material (LM25Al/VC) using computer numerical control (CNC) lathe operation. By optimizing CNC lathe process parameters such as depth of cut, feed rate, and cutting speed, the aim is to maximize material removal rate, minimize surface roughness, reduce power consumption, and optimize costs. The study employs analytical modeling, deep neural networks (DNN), and grey relational grade (GRA) coupled with response surface methodology (RSM) for performance evaluation. The effectiveness of these methods was compared using four objective verification mechanisms. In this case, the DNN technique delivered superior results among the methods considered. Additionally, new analytical models and DNN programming were developed in this work to predict machining costs, power consumption, material removal rate, and surface finish quality. These findings contribute to creating energy-efficient, cost-effective machining techniques and promote sustainable practices in the manufacturing industry.
本研究利用计算机数控（CNC）车床加工操作，对一种新型的碳化钒复合材料增强LM25铝合金（LM25Al/VC）的可加工性进行了研究。通过优化数控车床的工艺参数，如切削深度、进给速度和切削速度，目标是最大化材料去除率、最小化表面粗糙度、降低功耗并优化成本。该研究采用解析建模、深度神经网络（DNN）以及结合响应面方法（RSM）的灰色关联度（GRA）进行性能评估。使用四种客观验证机制对这些方法的有效性进行了比较。在这种情况下，DNN技术在所考虑的方法中取得了更好的结果。此外，本研究还开发了新的解析模型和DNN编程，以预测加工成本、功耗、材料去除率和表面光洁度质量。这些研究结果有助于创建节能、经济高效的加工技术，并促进制造业的可持续发展实践。
