Multi-objective Soft Computing Approaches to Evaluate the Performance of Abrasive Water Jet drilling Parameters on Die Steel

Abstract

In the current manufacturing methods, improvement in process performance will be resulting in huge benefits to the industries. For the improvement in part or product quality, the machining process must be operating with the optimal parameters, and thus, quality attributes can be produced. It applies to one of the modern machining techniques such as abrasive water jet (AWJ) as the use of this technique is growing rapidly in manufacturing industries for needful applications. Therefore, the present work is proposed to find the optimal level of AWJ parameters such as water jet pressure, stand-off distance, abrasive mass flow rate for drilling die steel with the simultaneous minimization of drilled hole features such as surface roughness, circularity, and cylindricity by using an unusual metaheuristic technique, namely harmony search algorithm (HSA). In addition, taguchi grey relational analysis (TGRA) was performed and their results were compared with the HSA technique. The multiple linear regression models were developed for each response, and the same were used in HSA to determine the optimum parameters for the minimization of responses where the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method was employed to convert the multi-objectives into single objective. In this study, the efficacy of HSA technique was demonstrated with the hole features of die steel. The results proved that the HSA technique had outperformed the TGRA based on the overall improvement in AWJ drilling performance on die steel, and the predicted process attributes such as surface roughness of 1.53625 μm, circularity of 0.04370 mm, and cylindricity of 0.02459 mm were obtained at water jet pressure of 275.4 MPa, stand-off distance of 3.96 mm, and abrasive flow rate as 0.25 kg/min. Also, the percentage deviation error (< 6%) of the predicted hole features of HSA is acceptable and analogous to the experimental results. Hence, it is confirmed that a new metaheuristic algorithm, namely HSA was found suitable for the AWJ drilling process.