The Lagrange model exhibits mean bias errors of 0.175 % (-0.010 F) which is better than the SVR model with temperature as the independent variable, which exhibits mean bias errors up to 0.909 % (0.062 F). For filling 1-6 missing hours of cooling data, heating data or weather data, a linear interpolation model or a polynomial model with hour-of-day (HOD) as the independent variable both provide a mean bias error of less than 0.087 % (0.005 F). Comparisons are made using six statistical parameters including mean bias error, coefficient of determination, and coefficient of variation of the root-mean-square-error. 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 50,000 "pseudo-gaps" which are created by assuming data is missing and then comparing the "filled" values with the measured values. 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 paper evaluates four methods for rehabilitating short periods of missing data. Data from five buildings is used to explain and illustrate the baseline analysis techniques and the exploratory work conducted.įilling short gaps (a few hours) in hourly energy use and weather data can be useful for (i) retrofit savings analysis and calculation, and for (ii) diagnostic purposes. Work to date has centered on investigation of Principal Component Analysis, an improved goodness-of-fit indicator for n-parameter change-point models, and calibrated simulation modeling. Substantial effort has been devoted to exploratory analysis intended to refine the analysis performed with the baseline techniques. Regression analysis with hourly scheduling profiles will be used for baseline analysis of hourly data. In addition to PRISM, two- and four- parameter linear, segmented change-point models are expected to be suitable for at least preliminary analysis of monthly and daily data for all buildings in the program. PRISM has been adopted as the baseline technique for buildings which are appropriate for treatment with one-, three- and five- parameter segmented linear, change-point models. The analysis effort during the first year of the Texas LoanSTAR Monitoring and Analysis Program has emphasized selection and development of baseline analysis techniques to cover the range of buildings expected in the program.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |