Artificial neural networks modelling for surface roughness in wire electrical discharge machining of Incoloy 800H

The use of artificial neural networks (ANNs) in optimizing process parameters and model development has gained attention due to their ability to predict timely and accurate outputs. The current study uses ANNs to predict surface roughness during wire electrical discharge machining (WEDM) of Incoloy 800H. The prediction model utilizes a multilayer perception based on back-propagation neural network (BPNN). In neural network modelling, network weights and bias values were updated, and the process output was estimated using the Levenberg-Marquardt backpropagation training function (trainlm) and a hyperbolic tangent sigmoid (tansig) transfer function, respectively. The ANN model comprises one output neuron (surface roughness) and five input neurons (pulse on, pulse off, wire tension, servo voltage, and wire feed rate). WEDM experimental data is divided into training, testing, and validation in the ratio of 70%:15%:15%. The results show that the optimal ANN model with 5-11-1 topology provides an accurate estimate of surface roughness, with a correlation coefficient of 0.997. The proposed ANN model can predict WEDM process parameters quickly and effectively.

I. Introduction

Incoloy 800H alloys, known for their heat resistance, and corrosion resistance are specifically engineered to provide superior resistance to oxidation compared to the conventional 800 series. These materials are well-suited for applications in which temperatures can surpass over 600°C [1, 2]. Moreover, these alloys also provide exceptional resistance to carburization and nitriding processes, which can induce fracture or surface hardening in certain metals [3, 4]. Significant heat generation, high cutting forces, diminished surface quality, extensive wear, and residual stresses have been identified as the major challenges during conventional machining of Incoloy 800H alloys [5, 6]. WEDM, a non-traditional machining process, is more capable of machining and fabricating geometrically complicated materials [7–9]. Because of the numerous benefits of WEDM machining, many researchers performed WEDM experimental investigations on different materials to determine machinability performance in terms of surface roughness, material removal rate, kerf width, etc. [10–13]. It is essential to choose the right process settings because of the intricate and unpredictable nature of the WEDM process and the numerous process parameters involved. Furthermore, the performance of the machining process is greatly impacted by the choice of process parameters. It is possible to control and optimize the input parameters to achieve the desired process output by developing a reliable input-output relationship model [14].