Self-Organizing Map Neural Network Algorithm for the Determination of Fracture Location in Solid-State Process joined Dissimilar Alloys

TL;DR

This paper applies a Self-Organizing Map neural network to accurately predict fracture locations in dissimilar alloys welded by friction stir, demonstrating high prediction accuracy using process parameters.

Contribution

It introduces the first use of a neurobiological based unsupervised learning algorithm for fracture location prediction in dissimilar alloys.

Findings

Achieved 96.92% prediction accuracy.

Successfully classified fracture locations at TMAZ of copper or aluminium.

Demonstrated effectiveness of SOM neural network in materials failure analysis.

Abstract

The subject area known as computational neuroscience involves the investigation of brain function using mathematical techniques and theories. In order to comprehend how the brain processes information, it can also include various methods from signal processing, computer science, and physics. In the present work, for the first time a neurobiological based unsupervised machine learning algorithm i.e., Self-Organizing Map Neural Network is implemented for determining the fracture location in dissimilar friction stir welded AA5754-C11000 alloys. Too Shoulder Diameter (mm), Tool Rotational Speed (RPM), and Tool Traverse Speed (mm/min) are input parameters while the Fracture location i.e. whether the specimen fracture at Thermo-Mechanically Affected Zone (TMAZ) of copper or it fractures at TMAZ of Aluminium. The results showed that the implemented algorithm is able to predict the fracture location with 96.92% accuracy.