A fast and accurate mean radiant temperature model for courtyards: Evidence from the Keyuan Garden in central Guangdong, China

Renzhi Wu, Xiaoshan Fang, Shuang Liu, Ariane Middel

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

Most models that simulate mean radiant temperature (Tmrt) for complex geometries are computationally expensive. We developed a fast and accurate Tmrt model for courtyards (CMRT) based on a parallelized quasi-physical algorithm. Here, we present the CMRT implementation, validate the model for Keyuan Garden in Central Guangdong, and compare model results and performance to ENVI-met and Ladybug Tools. The new model simulated Tmrt for various locations in the test bed courtyard with a root mean square error (RMSE) of 1.82 °C, outperforming Ladybug (5.00 °C) and ENVI-met (8.27 °C) and meeting the accuracy requirement for heat and cold stress. The CMRT model simulates Tmrt two orders of magnitude faster than ENVI-met and Ladybug Tools. Model performance increased with resolution of the domain and number of view facets. A sensitivity analysis shows that the solar beam fraction is the largest contributor to Tmrt after air temperature. Cloud cover has a non-linear effect on Tmrt, and every 1 °C increase in surrounding surface temperatures raises Tmrt by 0.28 °C. With faster simulations and reliable results, the CMRT model will facilitate future climate-resilient courtyard design.

Original languageEnglish (US)
Article number109916
JournalBuilding and Environment
Volume229
DOIs
StatePublished - Feb 1 2023

Keywords

  • Courtyard
  • GPU
  • Heat stress
  • Mean radiant temperature
  • Quasi-physical model
  • Sensitivity analysis

ASJC Scopus subject areas

  • Environmental Engineering
  • Civil and Structural Engineering
  • Geography, Planning and Development
  • Building and Construction

Fingerprint

Dive into the research topics of 'A fast and accurate mean radiant temperature model for courtyards: Evidence from the Keyuan Garden in central Guangdong, China'. Together they form a unique fingerprint.

Cite this