Hybridization of cuckoo search and bat algorithm for optimizing machining performances in deep hole drilling

Abstract

Deep Hole Drilling (DHD) is a machining process employed to produce holes with a length exceeding ten times of its diameter. The machine is utilized to assemble high-precision workpieces. The significant issue in DHD is in producing the best results of machining performances at the optimal value through the selection of machining parameters. In machining, achieving the optimum value of machining parameters is related to the performance and quality of products. Hence, the modelling and optimisation approaches are suitable for identifying the optimal DHD parameters to improve DHD performance. In this study, the real DHD experimentation based on an experiment (DOE) of full factorial with added centre points is conduct to investigate the influence of DHD machining parameters: feed rate (f), spindle speed (s), depth of hole (d) and Minimum Quantity Lubricant, MQL (m) on surface roughness (Ra), roundness (Rd), and cylindricity (Cy). The modelling process employed in the regression analysis consists of four types of mathematical models-multiple linear regression (MLR), two-factor interaction (2FI), multiple polynomial regression (MPR), and stepwise regression (SR)-were developed based on experimental data and used as an objective function for optimisation process. In the optimisation, Cuckoo Search (CS) was implemented in order to optimize the DHD machining performances. However, previous research indicates that CS has some weaknesses: trapping in local optima and slow convergence rate. Thus, a new hybridization between Cuckoo Search and Bat Algorithm (CS-BA) was developed to improve the DHD performance. Analysis of the results indicates that, CS-BA produced the minimum values and outperformed the standard CS algorithm and established computational techniques: ABC, GR-SVM, Integrated GA-SA-Type1, and Integrated GA-SA-Type2. Overall, it can be concluded that CS-BA hybridization has enhanced the quality and productivity of DHD problems significantly.