Forecasting Multivariate Time Series Meteorological Data for Solar Thermal Cogeneration Systems

• Main Authors: Tan, Lit Ken, Ong, Sie Meng, Nor Azwadi, Che Sidik, Asako, Yutaka, Lee, Kee Quen, Gan, Yee Siang, Goh, Chien Yong, Tey, Wah Yen, Ngien, S. K., Chuan, Zun Liang
• Format: Conference or Workshop Item
• Language: English
• Published: Higher Education Forum 2017
• Subjects: TA Engineering (General). Civil engineering (General)
• Online Access: http://umpir.ump.edu.my/id/eprint/19837/1/SICASE-0002.pdf; http://umpir.ump.edu.my/id/eprint/19837/

The high usage of fossil fuel to produce energy for the increasing demand of energy has been the primary culprit behind global warming. Alternative energy supply is thus necessary in order to prevent the situation from worsening. Recently, renewable energies such as solar energy has emerged as potential alternative energy resources due to its abundance all over the globe Solar energy can be harnessed using available system such as solar thermal cogeneration systems. However, fluctuations of solar radiation is one of the main challenge faced by the implementation of solar thermal cogeneration system due to its high variability. In order to have solar thermal cogeneration systems function smoothly and continuously, knowledge on solar radiation’s intensity several minutes in advance are required. While there exist various solar radiation forecast models, most of the proposed model are time consuming. In this research, a new methodology to forecast solar radiation via several meteorological data that incorporates dimension reduction technique is proposed. Based on the proposed methodology, two prediction models, Artificial Neural Network and statistical are established.