Modeling and parameter optimization for cutting energy reduction in MQL milling process

• Published in: International Journal of Precision Engineering and Manufacturing-Green Technology, 3(1) Vol. 3; no. 1; pp. 5 - 12
• Main Authors: Jang, Duk-yong, Jung, Jeehyun, Seok, Jongwon
• Format: Journal Article
• Language: English
• Published: Seoul Korean Society for Precision Engineering 01.01.2016; Springer Nature B.V; 한국정밀공학회
• Subjects: Algorithms; Artificial neural networks; Back propagation; Back propagation networks; Cutting energy; Cutting parameters; Cutting speed; Energy; Energy conservation; Energy consumption; Energy Efficiency; Engineering; Feed rate; Flow rates; Flow velocity; Global optimization; Industrial and Production Engineering; Lubrication; Mathematical models; Natural resources; Neural networks; Particle swarm optimization; Process variables; Raw materials; Sustainable Development; 기계공학; Environmentally conscious manufacturing; Minimum quantity lubrication; Specific cutting energy; Neural network; Particle swarm optimization
• ISSN: 2288-6206; 2198-0810
• DOI: 10.1007/s40684-016-0001-y

Environmentally conscious manufacturing (ECM), a key concept for modern manufacturing, emphasizes the efficient and optimal use of raw materials and natural resources and minimization of the negative effects on nature and society. This study focused on achieving ECM for milling processes. Toward this end, a model predicting the specific cutting energy was developed and optimized to determine the cutting conditions that minimize the specific cutting energy. A minimum quantity lubrication scheme was employed to minimize the amount of cutting oil used, thereby minimizing the associated process cost. Four process variables or cutting conditions (cutting speed, depth of cut, feed rate, and flow rate) were selected for the specific cutting energy model, and their appropriate ranges were determined through a preliminary experiment. The specific cutting energy model was developed based on an artificial neural network, where the Levenberg-Marquardt back propagation algorithm was implemented and the number of hidden layers was determined through comparison with controlled experimental data. The cutting conditions to minimize the specific cutting energy were determined using a global optimization process-the particle swarm optimization algorithm. In this algorithm, all computations were confined within the experimental range via constraint conditions, and the resulting optimized process variables were experimentally verified.