Rehabilitating missing energy use and weather data when determining retrofit energy savings in commercial buildings

Due to the character of the original source materials and the nature of batch digitization, quality control issues may be present in this document. Please report any quality issues you encounter to digital@library.tamu.edu, referencing the URI of the item.Includes bibliographical references (leaves 124-129).Issued also on microfiche from Lange Micrographics.This thesis evaluates four methods for rehabilitating short periods of missing data. Single variable regression, polynomial models, Lagrange interpolation, and linear interpolation models are developed, demonstrated, and used to fill 1-6 hour gaps in weather data, heating data and cooling data for commercial buildings. The methodology for comparing the performance of the four different methods for filling data gaps uses 11 one-year data sets to develop different models and fill over 250,000 "pseudo-gaps'' which are created by assuming data is missing and then comparing the "filled'' values with the measured values. The major findings may be summarized as follows: 1. It was also found that different data types have different gap frequency distributions. Data gaps of 1-6 hours cover all missing NWS temperature and dew point data. One to six hour gaps also cover 50-70% of the total missing LO-STAR temperature and humidity data, and 50-70% of total missing LO-STAR energy use such as cooling, heating, motor control use and electricity data. 2. The polynomial model and the linear interpolation model are comparable and more accurate than other models. The linear interpolation model is slightly better than the polynomial model for filling both missing weather data gaps and missing cooling data gaps. The least accurate is the Lagrange model, particularly as the length of the data gap increases. The single variable regression (SVR) method can not deal with missing weather data due to the pattern of weather data. Based on these findings, a polynomial model with hour-of-day as an independent variable and a linear interpolation model are recommended to fill 1-6 hour data gaps in cooling, heating and weather data