Sustainable High-Precision Hard Turning Under Various Near-Dry Machining Conditions Using Vegetable Oil-Based Nanofluids: Neural Network Modeling and Metaheuristic Optimization


Abderazek H., Hamdi A., YAPAN Y. F., UYSAL A., Merghache S. M.

Process Integration and Optimization for Sustainability, 2026 (ESCI, Scopus) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1007/s41660-026-00738-5
  • Dergi Adı: Process Integration and Optimization for Sustainability
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus
  • Anahtar Kelimeler: CO2 carbon footprints, Energy consumption, Green turning, Hard steel, Machining cost, Nanometric fluid
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

From the green vision to its practical implementation, modern machining is fully aligned with the dynamics of sustainable manufacturing, aiming to balance industrial performance with environmental preservation. This encompasses efforts to minimize total energy consumption (), limiting overall carbon dioxide emissions (), decreasing machining costs (), and improving the surface quality of the machined parts (). Here, the reduction of machining costs represents green or sustainable finance. The primary objective of this study is to analyze the combined impact of cooling conditions (), the parameters of the minimum quantity lubrication (MQL) system, specifically the flow rate () and nozzle distance (), as well as cutting parameters (speed, feed, and depth of cut), during the hard turning process of C45 steel (52 HRC). To this end, three lubrication configurations are compared: (i) MQL using pure biodegradable vegetable oil, (ii) MQL enriched with a nanofluid containing hexagonal boron nitride (hBN) nanoparticles, and (iii) MQL with a nanofluid formulated with multi-walled carbon nanotubes (MWCNT). The experimental data analysis methods used include analysis of variance, artificial neural networks, k-fold cross-validation, and the algorithm called Weighted Optimization Framework based Speed Constrained Multi-Objective Particle Swarm Optimization (WOFSMPSO) to determine the optimal solution in each optimization scenario. The results show a proportional correlation between energy consumption and its share in CO2 emissions. Among the parameters studied, lubricant flow rate proves to be the most influential on CO2 emissions, contributing 39.18%. Next in importance are cooling conditions, contributing 34.19%, and cutting speed, with a more limited impact of 7.96%. Additionally, the inclusion of nanoparticles, such as MWCNT nanograins or hBN nanoparticles, in the lubrication improves both surface quality and process efficiency.