Enhancement of the machining surface roughness model with cuckoo algorithm

Abstract

Quality of machining products is generally associated with the surface roughness (Ra) which is one of the important aspects that could affect machining performances. This study seeks to find the minimum value of Ra for a modern machining process, Abrasive Water Jet (AWJ). This study proposed Cuckoo algorithm to improve the surface roughness of the AWJ process because, since 2013, there has only been one study on the algorithm. Ra model in AWJ has five machining parameters comprising traverse speed (V), water jet pressure (P), standoff distance (h), abrasive grit size (D) and abrasive flow rate (m). These parameters in the Ra model have contributed to the best solution for machining performance by achieving the minimum values. In this study, to obtain better results, the parameters in the standard Ra model were enhanced in the modeling process where the number of parameters has been reduced. The enhanced parameters in the model are represented by “Q”. Computational techniques were used to evaluate the model whereas SPSS were used to validate the results. The result shows that minimum Ra value of the proposed model is 2.1709 µm. It was much lower than the results of three computational techniques which are Artificial Neural Network (ANN), Regression analysis and Support Vector Machine (SVM) by about 29.2 %, 23.8 % and 19.3% respectively.