Multi-objective optimization of MQL system parameters for the roller burnishing operation for energy saving, product quality and air pollution

Internal burnishing operation is a prominent solution to improve the hole quality. In this study, minimum quantity lubrication (MQL) system parameters, including the diameter of the nozzle ( N ), impingement angle ( I ), the pressure of the compressed air ( P ), the flow rate of the lubricant ( L ), and the distance between the nozzle and workpiece ( D ) are optimized for decreasing the total energy consumption ( TE ), average surface roughness ( AR ), and air quality indicator ( AI ) of the internal roller burnishing process. The predictive models of performance measures were developed using the adaptive neuro-based-fuzzy inference system (ANFIS) approach, while a novel model is developed to compute the total burnishing cost ( BC ). The neighborhood cultivation genetic algorithm (NCGA) and the VIKOR method were used to generate a set of prominent solutions and determine the best selection. The outcomes presented that the optimizing values of the N , I , P , L , and D are 1.0 mm, 35 deg., 0.3 MPa, 70 ml/h, and 10 mm, respectively. The TE , AI , AR , and BC are decreased by 3.1, 9.3, 20.6, and 7.9%, respectively, at the chosen point. The proposed performance measures could be utilized to precisely forecast the responses in the practical burnishing. The developed optimizing method combining the ANFIS, NCGA, and VIKOR could be effectively utilized to deal with complicated optimization issues for machining operations. The observed findings provided efficient information, which could help machine operators to select the optimal MQL system parameters and enhance the burnishing performances.