Green machining: A framework for optimization of cutting parameters to minimize energy consumption and exhaust emissions during electrical discharge machining of Al 6061 and SKD 11

Abstract

Dielectric medium (kerosene) between the tool electrode and workpiece forms a reaction product, owing to thermal effect of discharge in the process of electrical discharge machining (EDM), which is emitted from the surface in the form of harmful gas and may get into the operator’s body by breath system inhalation and contact with skin. Therefore, exhaust emissions should be considered as important performance indicators for the green machining in the process of EDM. Moreover, energy efficiency and machining productivity should be improved and the cut process must take ecological constraints into consideration. Thereafter, the definition of green manufacturing helps to evaluate different and sometimes conflicting aspects of the manufacturing process, such as machining productivity, surface quality, energy efficiency and exhaust emissions. But it is difficult to dynamically adjust the weights of different aspects according to the actual processing conditions. This paper proposes a framework for optimization of cutting parameters to minimize energy consumption and exhaust emissions during electrical discharge machining of Al 6061 and SKD 11. Three cutting process parameters (peak current, pulse period, and duty cycle) are tested against material remove rate (MRR), surface roughness (Ra), energy efficiency per volume (EEV), and exhaust emissions characteristics (EEC) for EDM of these two materials. First, the process energy and exhaust emissions are analyzed. The results show that EEV is negatively correlated with surface quality (Ra) for both Al 6061 and SKD 11 material. Second, two data analyses (Taguchi and adaptive-network-based fuzzy inference system (ANFIS) analyses) are utilized to obtain high MRR and best Ra, EEV and EEC for the desired machining performances and environmental sustainability. It is found that the relative errors between the predicted values and experimental results of MRR, Ra, EEV and EEC are all not more than 20%, and found that the maximum of mean relative errors of them is 13.04%. Finally, a framework implemented by an optimization system, based on ANFIS, enhanced cuckoo search (ECS) algorithm and desirability function, is built to obtain the desired cutting parameters. It is found that high peak current (8.94A) and pulse period (500μs) are recommended to obtain multi-objective optimization for machining Al 6061, and that the optimal solutions of the EEV and EEC are significantly decreased, by 87.13% and 37.09%, respectively. Similarly, high peak current (8.77A) and pulse period (261.5μs) are recommended to obtain multi-objective optimization for machining SKD 11, and the optimal solutions of the EEV and EEC are significantly decreased, by 93.73% and 46.85%, respectively. Thereafter, it can be concluded from the above research that the proposed technique offers significant advantages and potential for applications in the green manufacturing field.