Examination of electrochemical machining parameters for AA6082/ZrSiO4/SiC composite using Taguchi-ANN approach

Aluminum alloy is a widely utilized material in the modern automotive industry due to its lightweight properties and corrosion resistance. Unconventional machining processes, particularly electrochemical machining (ECM) offer effective means to work with such materials. This study focuses on assessing the influence of four specific parameter combinations on the machining of AA6082/ZrSiO4/SiC alloy. This work also analyzes the impact of critical ECM process parameters, including tool feed rate, applied voltage, electrolytic concentration, and electrode type on the output response variables. These variables encompass characteristics such as material removal rate (MRR) and surface roughness (SR), and their relationships are explored through the application of the Taguchi design of experiments methodology. The analyzed experimental data were employed to train an Artificial Neural Network (ANN) model aimed at achieving more accurate predictions to increase the MRR and reduce SR. The ANN setup is a multilayer perceptron utilizing a feed forward architecture, denoted as (4–20–2). This notation indicates that there are 4 nodes in the input layer, twenty neurons in the hidden layers, and 2 nodes in the output layer. The ANN predictions yield an R2 value of 0.98003 and MSE within the range of 0.02413, specifically for the experiment dataset. The results of the regression study strongly indicate that the ANN model can effectively and reliably predict both MRR and SR with a high degree of precision. The scanning electron microscope (SEM) micrograph of the surface also indicates an improved surface finish with brass tool as compared to graphite.