A RSCMAC Based Forecasting for Solar Irradiance from Local Environmental Information

• Main Authors: Xiao Ru Li, 李効儒
• Other Authors: Ching-Tsan Chiang
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/21918023548931977582

碩士 === 清雲科技大學 === 電機工程所 === 100 === In recent years, PV system installations increase is not only in quantity, but also in large system or even in power plant. The accumulated installed capacity is getting large, and this could affect entire grid power management and scheduling. Based on this consideration, solar irradiance prediction becomes very important to estimate PV power generation, and the generated power will affect power deploy, schedule, or even the entire power grid stability. This project is aimed to provide an efficient solar irradiance prediction model to predict the installed PV system power generation and also for the evaluation of future large-scale grid-connected PV system or PV power plant. The purpose of this project is to apply Recurrent S_CMAC_GBF (RSCMAC) to predict extreme short-term Solar Irradiance. Currently, most studies of the solar irradiance prediction focus on Global Hourly Solar Irradiations (GHSI) and their purpose is to predict the installed PV system power generation, so it can be used to evaluate the installation benefit. Therefore, this project utilizes RSCMAC to develop a solar irradiance prediction model and to verify its feasibility.