Engineering Commons^™

Automating Intensity Modulated Radiation Therapy Treatment Planning By Using Hierarchical Optimization, Paras Babu Tiwari

McKelvey School of Engineering Theses & Dissertations

The intensity modulated radiation therapy (IMRT) optimizes the beam’s intensity to deliver the prescribed dose to the target while minimizing the radiation exposure to normal structures. The IMRT optimization is a complex optimization problem because of the multiple conflicting objectives in it. Due to the complexity of the optimization, the IMRT treatment planning is still a trial and error process. Hierarchical optimization was proposed to automate the treatment planning process, but its potential has not been demonstrated in a clinical setting. Moreover, hierarchical optimization is slower than the traditional optimization. The dissertation studied a sampling algorithm to reduce the hierarchical …

Economic Analysis And Justification For Automated Welding Systems, Roxanne Raye

Industrial and Manufacturing Engineering

This report looks at designing a new welding process for a company in the Oil & Gas industry, ensuring method consistency and performance quality of their steel drill bit while increasing overall production. The current product is welded manually with a flux core wire, resulting in high cycle times and numerous cases of required rework.

The use of automation systems will be analyzed for their increased consistency, quality, safety, and production estimates. Specifically robotic cell design will be considered, since it can work on more parts with fewer operators, adding efficiency and flexibility to the production floor. Other areas of …

Solar Decathlon Instrumentation And Controls, Timothy Ambrose, Mateo Begue, Andrew Elliott

Mechanical Engineering

The Instrumentation and Controls team has designed a control and instrumentation system for the 2015 Cal Poly Solar Decathlon house that will monitor temperature, humidity, and energy usage throughout the house and control the phase change material duct. It will relay information to the user through a tablet application developed by the Computer Science team. The team has also designed a lighting control scheme for use with Lutron’s HomeWorks QS lighting control system.

Forest Sign Maker, Victor Espinosa Iii, Kevin Ly, Lisa Yip

Mechanical Engineering

Executive Summary:

The Inyo National Forest is arguably one of the most beautiful locations in California, containing natural masterpieces such as Mount Whitney and the Ancient Bristlecone Pine Forest. Despite its magnificence, the Inyo National Forest can be a treacherous region. The Friends of the Inyo take pride in being able to facilitate the viewing experience for all outdoorsmen by maintaining the mountain trails, which includes providing adequate trail signage.

Unfortunately, there is a fundamental issue with the recent state of trail signage in the Inyo National Forest: the rate at which signs are being vandalized or naturally destroyed is …

Semi-Automated Switching Regulator Modeling Method And Tool, Michael Heath Leonard

Graduate Theses and Dissertations

This thesis presents the results of research targeted at automating the behavioral modeling process for switching voltage regulators. These regulators are commonly used in many application areas including discrete use in larger systems, integrated in a System on a Chip (SoC), or as the primary use case for a design. When used in an integrated system these regulators can be a significant force in slowing down simulations. A common method for removing this slowdown is to use a behavioral model of the switching regulator.

Creating behavioral models can be very time consuming and requires expertise.

The thesis discussion begins by …

Robust Models For Operator Workload Estimation, Andrew M. Smith

Theses and Dissertations

When human-machine system operators are overwhelmed, judicious employment of automation can be beneficial. Ideally, a system which can accurately estimate current operator workload can make better choices when to employ automation. Supervised machine learning models can be trained to estimate workload in real time from operator physiological data. Unfortunately, estimating operator workload using trained models is limited: using a model trained in one context can yield poor estimation of workload in another. This research examines the utility of three algorithms (linear regression, regression trees, and Artificial Neural Networks) in terms of cross-application workload prediction. The study is conducted for a …