本研究使用概似不確定性估計法(Generalized Likelihood Uncertainty Estimation, GLUE)配合中屯風力發電站即麥寮風力發電站資料進行日發電量不確定性以及月發電量不確定性之評估,並使用KS測試(Kolmogorov–Smirnov test)驗證其結果。在日發電量之不確定性方面,以此模式評估約有38%有過α=0.05之檢定,約56%有過α=0.01之檢定。在月發電量之不確定性部分,有通過α=0.05之檢定比例超過是90%,不同地點以相對誤差作為似然函數(likelihood function)通過α=0.05之檢定比例為58.33%,以NS效率係數(Nash-Sutcliffe efficiency coefficient) 作為似然函數則為83.33%。氣候變遷對東吉島之風力發電造成的影響會隨著氣候變遷的影響程度變嚴重,低發電量部分出現比現況更低發電量之機會提高,使得高發電量月份的不確定性提高。
This study uses Generalized Likelihood Uncertainty Estimation (GLUE) to evaluate the wind power uncertainty on daily and monthly periods, and uses the Kolmogorov–Smirnov test (KS test) to verify the results. The test shows that about 38% data sets pass the KS test on daily uncertainty at α = 0.05, and about 55% at α = 0.01. In the case of monthly uncertainty, it is more than 90% data sets pass the KS test at α = 0.05. It is 58.33% passing the KS test at α = 0.05 by using the relative error as the likelihood function at different site. In contrast, it would be 83.33% by using the Nash-Sutcliffe efficiency coefficient as the likelihood function. Finally, this study uses GLUE to evaluate the uncertainty of the impact of climate change on wind power in Tungchitao. The results show that the lower the wind power tends to be, the higher the probability of incidence of low wind power impacted by climate change will be.