Optimizing a Parametric Energy Model for Use in Citywide Residential Overheating Analysis

Abstract

City-wide urban building energy modeling may help estimate which residential buildings in a city are at risk of occupant overheating during a heat wave. Six parametric energy modeling methods are presented for use in an automated analysis to generate energy models using existing building information or GIS data. An evaluation of the six methods generates models that mimic detailed energy models for three existing homes in Cleveland, Ohio, USA. A comparison of parametric to detailed models using R2 and CV-RMSE indicates that five out of six models adequately represent the detailed models when analyzing operative temperature and wetbulb globe temperature.