An hourly hybrid multi-variate change-point inverse model using short-term monitored data for annual prediction of building energy performance, part III: Results and analysis (1404-RP)

摘要

The hourly hybrid multi-variate change-point approach aimed at predicting building energy consumption by combining a short-term data set of monitored energy consumption, weather variables and internal loads with at least 1 year of recent utility bills. The results showed an improvement in using the hourly hybrid multi-variate change-point modeling approach in predicting energy consumption in a building when compared with the utility bills approach discussed by Abushakra and Paulus (2016). Since the method uses utility history to represent the long-term data, 2 weeks of monitoring of hourly data in many cases were found to be sufficient for estimating long-term energy consumption. This article shows the hourly time scale results of ASHRAE RP-1404, along with an analysis that provides recommendations and guidance to energy modelers in their use of short-term monitoring for long-term prediction of building energy performance. Color area plots, developed from over 1 million simulations, show that the user would have flexibility finding satisfactory short-term periods for monitoring, be it 2-week periods or longer.