Engineering Commons^™

Multiscale Topology Optimization With A Strong Dependence On Complementary Energy, Dustin Dean Bielecki

All Dissertations

A discrete approach introduces a novel deep learning approach for generating fine resolution structures that preserve all the information from the topology optimization (TO). The proposed approach utilizes neural networks (NNs) that map the desired engineering properties to seed for determining optimized structure. This framework relies on utilizing parameters such as density and nodal deflections to predict optimized topologies. A three-stage NN framework is employed for the discrete approach to reduce computational runtime while maintaining physics constraints.

A continuous representation that uses complementary energy (CE) methods to solve a representative element's homogenized properties consists of an embedded structure that is …

Performance Based Design And Machine Learning In Structural Fire Engineering: A Case For Masonry, Deanna Craig

All Theses

The volatile and extreme nature of fire makes structural fire engineering unique in that the load actions dictating design are intense but not geographically or seasonally bound. Simply, fire can break out anywhere, at any time, and for any number of reasons. Despite the apparent need, fire design of structures still relies on expensive fire tests, complex finite element simulations, and outdated procedures with little room for innovation. This thesis will make a case for adopting the principles of performance-based design and machine learning in structural fire engineering to simplify the process and promote the consideration of fire in all …

Algorithm Optimization And Hardware Acceleration For Machine Learning Applications On Low-Energy Systems, Jianchi Sun

All Dissertations

Machine learning (ML) has been extensively employed for strategy optimization, decision making, data classification, etc. While ML shows great triumph in its application field, the increasing complexity of the learning models introduces neoteric challenges to the ML system designs. On the one hand, the applications of ML on resource-restricted terminals, like mobile computing and IoT devices, are prevented by the high computational complexity and memory requirement. On the other hand, the massive parameter quantity for the modern ML models appends extra demands on the system's I/O speed and memory size. This dissertation investigates feasible solutions for those challenges with software-hardware …

A Novel Computationally Efficient Ai-Driven Generative Inverse Design Framework For Accelerating Topology Optimization And Designing Lattice-Infused Structures, Darshil Patel

All Dissertations

Multiscale topology optimization (TO) provides an inverse design computational framework for designing globally and locally optimized hierarchical structures. Triply periodic minimal surfaces (TPMS), a subclass of parametrically-driven lattice structures, exhibit unique properties such as large surface area, significant volume densities, and good strength-to-weight ratio, which makes them favorable for novel engineering applications. The recent advances in additive manufacturing and its ability to fabricate high-resolution structures have spurred interest in multiscale TO and TPMS for computationally designing finer and high-resolution designs. While multiscale TO and TPMS bring transformative opportunities in various applications, their potential for everyday use remains idle due to …

A Quantitative Comparison Of Algorithmic And Machine Learning Network Flow Throughput Prediction, Cayden Wagner

All Theses

Applications ranging from video meetings, live streaming, video games, autonomous vehicle operations, and algorithmic trading heavily rely on low latency communication to operate optimally. A solution to fully support this growing demand for low latency is called dual-queue active queue management (AQM). Dual-queue AQM's functionality is reduced without network traffic throughput prediction.

Perhaps due to the current popularity of machine learning, there is a trend to adopt machine learning models over traditional algorithmic throughput prediction approaches without empirical support. This study tested the effectiveness of machine learning as compared to time series forecasting algorithms in predicting per-flow network traffic throughput …