| Summary: | 碩士 === 國立臺灣大學 === 生物環境系統工程學研究所 === 103 === Recently green energy, including wind energy, is attached great importance. Moreover, with the improvement of wind energy technology the cost of wind energy development is decreasing. If we can properly assess the wind energy potential in a region, we can provide an incentive to investors. In the process of assessing wind power generation, there exists uncertainties, such as distribution of wind speed, selection of parameters, and estimation of wind generating capacity, etc. Considered the uncertainty of each step, the wind energy output can be estimated accurately, then developers can have more information to make right judgments.
This study uses the Generalized Likelihood Uncertainty Estimation (GLUE) to assess the uncertainty of wind energy capacity. By revising the difference between simulation and measured results and giving a proper weight, the GLUE method can give more accurate simulation results than the widely used Monte Carlo method. This study use the data from 2002 to 2011, collected from the wind power plant in Penghu Jhongtun, and divide the data into 15 simulated scenarios, i.e, 12 months, strong wind period (from January to March and October to December), low wind period (April to September) and the whole year, to explore simulation uncertainty.
The research results indicate that the GLUE method has better simulation results than Monte Carlo method does. In most of the scenarios, the GLUE method can describe the realities of the situation. This result also shows that the wind power capacity is steady in summer and winter, while in spring and autumn, because the wind speed changes largely, the uncertainty distribution is relatively larger.
|
|---|
