Grinding parameters optimisation of white marble using response surface methodology integrated with NSGA-II algorithm approach

This study combined response surface methodology (RSM), desirability function approach (DFA) and non-dominated sorting genetic algorithm (NSGA-II) to estimate the optimal grinding parameters that obtain the minimum surface roughness and contouring error value of white marble. A series of experiments was used to determine the effect of grinding parameters of robotic manipulators such as spindle speed, feed speed, grinding width and grinding depth on surface roughness (Ra) and contouring error (Cl). The Ra values of samples were measured by a stylus-type equipment and Cl values were obtained by employing a three-coordinate measuring machine. According to the results of the experimental design, the RSM was used to establish the prediction models of Ra and Cl. The parameters-optimised process that leads to minimum Ra and Cl was divided into three stages. Firstly, the established prediction model was optimised by DFA method. Secondly, the results obtained from DFA were chosen as the initial point of NSGA-II. Finally, the optimum grinding parameter values were obtained by using NSGA-II method. The experimental results showed that the Ra value of white marble reaches 3.166 μm, and the Cl value reaches 0.053 mm by the integrated optimisation approach.