Statistical analysis of wind-induced pressure fields and PIV measurements on two buildings

Bubryur Kim, K. T. Tse, Akihito Yoshida, Yukio Tamura, Zengshun Chen, Pham Van Phuc, Hyo Seon Park

Research output: Contribution to journalArticle

4 Citations (Scopus)


This study investigated the aerodynamic characteristics of two tall buildings through an independent component analysis (ICA) and principal component analysis (PCA) of wind-induced pressure fields. The pressure patterns identified by ICA and PCA were compared using the wind pressures on two building models measured in a series of wind tunnel tests under five cases consisting of different gaps between two buildings. To further investigate the flow patterns, particle image velocimetry was employed to measure the instantaneous wind flow patterns. The ICA results in the 1st, 2nd, and 4th modes indicated different pressure distributions on the inside building surfaces for different gaps, while the 3rd mode indicated a suction phenomenon on the outside and leeward surfaces; the PCA results indicated that the gap only influenced the pressure on the inside surfaces in the first four modes. Only the first PCA mode had a higher correlation with the original data than the ICA modes, and the ICA modes generally had higher correlations than the other modes. In wind forces, the 1st and 3rd PCA modes provided similar information, whereas all the ICA modes provided different information. Overall, ICA provided more diverse information than PCA, which yielded rather limited and homogenous information.

Original languageEnglish
Pages (from-to)161-174
Number of pages14
JournalJournal of Wind Engineering and Industrial Aerodynamics
Publication statusPublished - 2019 May

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Renewable Energy, Sustainability and the Environment
  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Statistical analysis of wind-induced pressure fields and PIV measurements on two buildings'. Together they form a unique fingerprint.

  • Cite this