Wind Mapping of Malaysia Using Ward's Clustering Method

Azhar, Amar and Hashim, Huzaifa (2024) Wind Mapping of Malaysia Using Ward's Clustering Method. Energies, 17 (7). p. 1563. ISSN 1996-1073, DOI https://doi.org/10.3390/en17071563.

Full text not available from this repository.
Official URL: https://doi.org/10.3390/en17071563

Abstract

Malaysian wind maps used in wind analysis and engineering design are dated back to the last 50 years of wind data. Cotemporally, the mean wind speed was used as the basic wind design and due to changes in the weather condition worldwide, wind monitoring assumption and structure design are jeopardized. Therefore, this study aims to map the wind based on the trend of the highest wind speed recorded. The wind speed was analyzed based on a trend basis. The study included 42 Malaysian Metrological Department weather stations and the annual wind speed was acquired based on the monthly highest wind speed (1990-2019). The data were then processed using the 95% confidence interval method to determine the mean of the month, and the annual wind trendline was obtained. The trendline was then clustered using Ward's method with the assistance of the high-level programming language, PYTHON v3.11.6. The clustering analysis produced two clusters for the Malaysian Peninsula and two for Sabah and Sarawak. The recommendation is to use the highest wind speed recorded on the map as the design wind speed. The recommendation is expected to help experts to compensate for the uncertainties in wind speed during the design stage and avoid incidents in the field.

Item Type: Article
Funders: UNSPECIFIED
Uncontrolled Keywords: mean wind speed; coastal; inland; wind trendline; wind clustering; Ward's method
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Divisions: Faculty of Engineering > Department of Civil Engineering
Depositing User: Ms. Juhaida Abd Rahim
Date Deposited: 14 Oct 2024 07:42
Last Modified: 14 Oct 2024 07:42
URI: http://eprints.um.edu.my/id/eprint/45346

Actions (login required)

View Item View Item