Urban Energy Modeling: Evaluating approaches for districtwide calibration considering the Urban Heat Island effect

  • Luis G. R. Santos

Student thesis: Master's Thesis

Abstract

Over the past decade, building energy modelling research has increasingly focused on urban-scale models. The shortcomings of analyzing urban buildings in isolation are well known and far from negligible (mainly, the inability to account for urban heat island, shading from neighboring obstructions and obstructed wind flow). The aim of this paper is to evaluate the impact of the urban context via urban-scale modelling and inverse parameter estimation (calibration) using metered energy consumption of each building. This aims to enhance current urban and building models based on Building Energy Softwares, incorporating adjacent shading from buildings and estimating the Urban Heat Island effect. We describe an automated calibration method for 56 buildings in a representative downtown district of Abu Dhabi, which has undergone a detailed energy audit, collecting data from 2008 to 2010. Since it is well known that, due to the urban heat island effect, the urban ambient air temperature can differ significantly from the reference rural air temperature used in most building simulations, the calibration procedure will also consider this differential. Three main approaches of district calibration are proposed, using Genetic Algorithm. The first one estimates seven building unknown parameters together with four variables related to the weather (describing annual average and seasonal variation of the UHI effect on both temperature and humidity). The second approach uses the software UWG to pre-process the weather in order to generate the urban weather, thereby reducing the number of estimated parameters. The third approach is based on using a Surrogate model instead of detailed building models. The computationally fast surrogate model is extensively trained to mimic the detailed models' calibration error. Two approaches are looked at for the calculation of the ASHRAE 14 calibration error metrics (CvRMSE and NMBE). One approach is too look at the whole district as one aggregate building, while the other, introduced for the first time herein, consists in deriving the weighted average of the error of each building. The main contribution of the paper is to propose a faster way to calibrate a large number of buildings, considering effects of the outdoor environment. For the weighted average approach, CvRMSE found is between 17.84% and 18.07%. NMBE is between 15.36% and 15.50%. For grouped buildings, CvRMSE is between 4.93% and 5.16%. NMBE is between -3.21% and 4.65%.
Date of AwardDec 2017
Original languageAmerican English

Keywords

  • Urban areas
  • Urban Heat Island phenomenon
  • Building Energy Model
  • Artificial Neural Network modeling
  • Calibration Methodology.

Cite this

'