Answers are grounded in this paper and cite their sources.
Machine learning techniques are being widely adopted across the globe for their reliability and flexibility when compared with other traditional methods. However, the selection of suitable machine learning techniques has a major role in a process for the prediction of optimal process parameters. In this study, we have adopted two different machine learning techniques Adaptive Neuro-Fuzzy Inference System (ANFIS) and Unified Convolutional Neural Network (UCNN) for the identification of optimal process parameters for the SS 316 L base alloy cladded with Er-NiCr-3 filler by Tungsten Inert Gas (TIG) cladding process. The ANFIS methodology will develop a model with a range of process parameters that can be used to determine the theoretical values, whereas the UCNN uses images for the identification of any defect in the samples the images are broken as different pixels based on the algorithms employed. Here, we have correlated the machine learning outputs with the actual experimental values (microhardness and tensile values are considered for the correlation). Whereas, for the UCNN technique we have procured the grain structures of the cladded samples. It is inferred from the comparison that the machine learning technique had shown sound and reliable outputs with an error percentage (≈ 0.1–2.0 %) in line with the actual data. Therefore from the study, it is revealed that the adoption of machine learning techniques can be utilized wisely for a process in the prediction of optimal process parameters in a flexible manner when compared with the other traditional optimization techniques.
No takes yet. Share an insight, caveat, or question.
R et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: