Research Experiences For Undergraduates (REU)

REU 2011 Projects

Geometry and flow effects on the performance of vertical residential geothermal heating systems

Sponsor: New England Geothermal Professional Association
Advisor: Prof. Burt Tilley


Justin Boyer		Spencer Frei		Kathyrn Lockwood		Gregory Stewart

Residential geothermal heating technology has been developed over the past few decades as an alternative to fossil-fuel based heating. However, the high initialization cost of these systems has proven to be a significant barrier to their acceptance. In order to reduce these initial costs, a better understanding of how the different pipe geometries and their orientations within a grout-filled borehole needs to be better understood. We are interested in formulating novel mathematical descriptions, utilizing the disparate length scales in the radial and axial directions, which can inform how different cross-section geometries, and geometrical variations in the axial direction affect the harnessing of energy from the soil. This problem is coupled to the thermal response in the soil, and hence both fields need to be found simultaneously.

Pricing and Hedging Variable Annuities Using Monte Carlo Methods

Advisor: Dr. Matthew Willyard


John Collins	Daniel Gitelman	Kevin McDermott

Dimin Xu	Xiao Xu

Variable annuities are becoming a very popular retirement option because of the various guarantees and protections they offer. To avoid having an overly complicated set of fees, insurance companies are seeking more efficient ways to hedge against risk. We use Monte Carlo methods to determine the variable annuity's price and its sensitivities to market conditions. Ultimately, we will determine more effective hedging strategies for insurance companies to reduce their chances of catastrophic loss.

Evaluation of Glucose Sensor Simulated Use as a Predictor of Device Performance

Sponsor: Instrumentation Laboratory
Advisor: Prof. Zheyang Wu


Michelle L. DeVeaux		Sasha S. Indarte		Samantha M. Noreen

The goal of this project is to characterize the relationship between devicesimulated use and device performance. Toward that end, we analyzed trends in performance of device simulated use testing and device performance separately, as well as the associativity between these two. Statistical methodologies involve multivariate correlation estimation, McNemar's test, Chi Square test, linear and logistic regression modeling and selection, and survival analysis. In selecting models that characterize the relationship between the simulated use testing and the device performance data, various model comparison criteria (AIC, BIC, MSE, norm ratio, etc.) were explored. Model searching strategies include stepwise search and exhaustive search. Cross-validation was employed to evaluate and choose models with the highest predictive accuracy.

Maintained by webmaster@wpi.edu
Last modified: Jan 02, 2015, 22:46 UTC