Engineering Commons^™

Extending The Convolution In Graph Neural Networks To Solve Materials Science And Node Classification Problems, Steph-Yves Mike Louis

Theses and Dissertations

The usage of graph to represent one's data in machine learning has grown in popularity in both academia and the industry due to its inherent benefits. With its flexible nature and immediate translation to real life observed objects, graph representation had a considerable contribution in advancing the state-of-the-art performance of machine learning in materials.

In this dissertation proposal, we discuss how machines can learn from graph encoded data and provide excellent results through graph neural networks (GNN). Notably, we focus our adaptation of graph neural networks on three tasks: predicting crystal materials properties, nullifying the negative impact of inferior graph …

An Artificial Intelligence Approach To Fatigue Crack Length Estimation From Acoustic Emission Signals, Shane T. Ennis

Theses and Dissertations

As in service aircraft begin to age and fatigue, a method for evaluating the operational life they are currently operating under and have remaining comes into question. Structural health monitoring is (SHM) is a popular method of structural analysis with growing interest in the aerospace industry. SHM is capable of damage assessment and structural life estimations.

The ultimate goal of the research presented in this thesis is to develop a methodology of classifying the length of a fatigue crack though the use of machine learning. The thesis has three major chapters as described below.

The first chapter deals with the …

Classification Of Acoustic Emission Data Into Load Steps Using An Artificial Neural Network, Allen Ross

Theses and Dissertations

The average age of bridges in South Carolina is approaching 40 years, very close to the 50-year service life. Almost 11% of the bridges in South Carolina are rated as structurally deficient, greater than the national average of 7.5% (South 2021). Health monitoring is the concept that in-situ sensors can continuously monitor civil structures and send real time signals of damage. This process allows for automation and could save time and money on the inspection and load rating of bridges. Health monitoring requires sensors that provide uninterrupted data without sacrificing the functionality of the bridge. A prime candidate for this …

Simulation-Based And Data-Driven Approaches To Industrial Digital Twinning Towards Autonomous Smart Manufacturing Systems, Kaishu Xia

Theses and Dissertations

A manufacturing paradigm shift from conventional control pyramids to decentralized, service-oriented, and cyber-physical systems (CPSs) is taking place in today’s Industry 4.0 revolution. Generally accepted roles and implementation recipes of cyber systems are expected to be standardized in the future of manufacturing industry. Developing affordable and customizable cyber-physical production system (CPPS) and digital twin implementations infuses new vitality for current Industry 4.0 and Smart Manufacturing initiatives. Specially, Smart Manufacturing systems are currently looking for methods to connect factories to control processes in a more dynamic and open environment by filling the gaps between virtual and physical systems.

The work presented …

Artificial Intelligence Approaches For Structural Health Monitoring Of Aerospace Structures, Kimberly A. Cardillo

Theses and Dissertations

Structural health monitoring (SHM) and non-destructive evaluation (NDE) have been a significant research topic to help with damage detection in aerospace structures. SHM and NDE techniques are based on extracting damage sensitive features to determine the criticality of damage and lifetime of a structure. Acoustic emission (AE) signal detection is an important technique in SHM and NDE especially for fatigue crack growth. AE signals for thin aerospace structures consist of ultrasonic guided Lamb waves that propagate through the structure. This thesis focuses on AE signal repeatability, load at which AE signals occur, feature extraction, artificial intelligence and electro-mechanical impedance of …

Explainable Ai Using Knowledge Graphs, Manas Gaur, Ankit Desai, Keyur Faldu, Amit Sheth

Publications

During the last decade, traditional data-driven deep learning (DL) has shown remarkable success in essential natural language processing tasks, such as relation extraction. Yet, challenges remain in developing artificial intelligence (AI) methods in real-world cases that require explainability through human interpretable and traceable outcomes. The scarcity of labeled data for downstream supervised tasks and entangled embeddings produced as an outcome of self-supervised pre-training objectives also hinders interpretability and explainability. Additionally, data labeling in multiple unstructured domains, particularly healthcare and education, is computationally expensive as it requires a pool of human expertise. Consider Education Technology, where AI systems fall along a …