Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 3 of 3
Full-Text Articles in Engineering
Gearbox Baffle Optimization, Megan Arduin
Gearbox Baffle Optimization, Megan Arduin
Masters Theses
Current literature reveals there is limited consensus on the placement of baffles within a gearbox to reduce churning losses. Thus, there is a need for a process to identify baffle clearances that result in maximum and minimum churning losses. There are two types of baffles: axial and radial. While both axial and radial baffles cause reductions in churning losses to various degrees, the focus is on the effect of radial baffles. The effect of a board (rectangular plate) baffle location on the churning losses of a single gear gearbox are evaluated using computational fluid dynamics (CFD) implemented in Ansys. Several …
A Reinforcement Learning Approach To Spacecraft Trajectory Optimization, Daniel S. Kolosa
A Reinforcement Learning Approach To Spacecraft Trajectory Optimization, Daniel S. Kolosa
Dissertations
This dissertation explores a novel method of solving low-thrust spacecraft targeting problems using reinforcement learning. A reinforcement learning algorithm based on Deep Deterministic Policy Gradients was developed to solve low-thrust trajectory optimization problems. The algorithm consists of two neural networks, an actor network and a critic network. The actor approximates a thrust magnitude given the current spacecraft state expressed as a set of orbital elements. The critic network evaluates the action taken by the actor based on the state and action taken. Three different types of trajectory problems were solved, a generalized orbit change maneuver, a semimajor axis change maneuver, …
Evaluation Of Data Collection Operations For Real-Time Influenza Surveillance During An Emergency, Yuwen Gu
Evaluation Of Data Collection Operations For Real-Time Influenza Surveillance During An Emergency, Yuwen Gu
Dissertations
It is unclear how data collection operations for surveillance alter the disease portrayal that influenza reported trends attempt to provide during an emergency. This study developed a model that simulates the collection and testing of influenza specimens after an outbreak is declared in Michigan. It performed simulation based optimization to understand which operational factors affect the biases between the growth rates of original and observed influenza incidence trends, and to quantify the predictive power of the influenza incidence trends at different points of data collection. The results show that emergency driven high risk perception increases the reporting, which leads to …