A Prediction Model for Additive Manufacturing of Inconel 718 Superalloy

Abstract

Inconel 718 is a nickel-based superalloy and an excellent candidate for the aerospace, oil, and gas industries due to its high strength and corrosion resistance properties. The machining of IN718 is very challenging; therefore, the application of additive manufacturing (AM) technology is an effective approach to overcoming these difficulties and for the fabrication of complex geometries that cannot be manufactured by the traditional techniques. Selective laser melting (SLM), which is a laser powder bed fusion method, can be applied for the fabrication of IN718 samples with high accuracy. However, the process parameters have a high impact on the properties of the manufactured samples. In this study, a prediction model is developed for obtaining the optimal process parameters, including laser power, hatch spacing, and scanning speed, in the SLM process of the IN718 alloy. For this purpose, artificial neural network (ANN) modeling with various algorithms is employed to estimate the process outputs, namely, sample height and surface hardness. The modeling results fit perfectly with the experimental output, and this consequently proves the benefit of ANN modeling for predicting the optimal process parameters.

Cost-Aware Design and Fabrication of New Support Structures in Laser Powder Bed Fusion: Microstructure and Metallurgical Properties

“…A high microhardness value of 460.5 HV was achieved in the as-fabricated IN718 sample built on top of a pin support structure. The results were comparable with the LPBF IN718 in the literature for as-fabricated LPBF sample (322 HV) [21], heat-treated LPBF (335 HV), heat-treated plus hot isostatic pressing (478 HV) [37], and as-fabricated wrought and cast (353 HV) [37].…”

“…On the other hand, at a relatively high value of the laser power (P), Parimi et al [20] observed increased numbers of entrapped bubbles within the melt pools, thus a decreased level of density. Ravichander et al [21] conducted a comprehensive study on the effect of laser processing parameters on the geometry accuracy of the as-fabricated sample. Out of all other energy density parameters (laser power (P), hatch spacing (h), and layer thickness (t)), they found laser scan speed (v) as a dominant factor towards grain length, which affects the final dimension of the part.…”

Scanning Strategy Investigation for Direct Powder Bed Selective Laser Processing of Silicon Carbide Ceramic

“…Additive Manufacturing (AM) can offer highly complex parts with minor post-processing operations [9][10][11]. AM is a developing technology that can produce parts from 3D CAD models, as defined by ISO/ASTM 52900 [12].…”

Surface Roughness Analysis and Prediction with an Artificial Neural Network Model for Dry Milling of Co–Cr Biomedical Alloys

“…An artificial neural network (ANN) is a computational system that uses a network of functions (nodes) to process data input into a desired output. It was inspired by the biological nervous system; therefore, the network is composed of interconnected elements (nodes) called artificial neurons [ 21 , 59 , 60 ]. The ANN computational system consists of three layers: the input layer, which contains input neurons that represent input data; the hidden layer, which contains a specific number of hidden neurons; and the output layer, which consists of the output data (the number of neurons is equal to the number of desired outputs).…”

“…ANN is a rapid and reliable method broadly used for solving complex problems by using various algorithm architectures [ 33 , 58 , 59 ]. Consequently, this modeling approach can be used as a prediction instrument for determining nonlinear relationships between input and output results for different manufacturing processes [ 59 ].…”