Optimum Prediction of Electricity Demand Island-Wide in Taiwan Using Genetic-Gray Forecasting Model

• Main Authors: Kai-Jung Hsu, 許凱榮
• Other Authors: Jyh-Horng Chou
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/04461264295225180595

碩士 === 國立高雄第一科技大學 === 機械與自動化工程所 === 95 === In this paper, the electricity demand island-wide in Taiwan is predicted by using genetic-gray forecasting model. First, the researcher built the Gray Prediction Model DGM(1,1) to predict limited information. Second, the researcher adjusted the accuracy of Gray Prediction Model based on Genetic Algorithm. The calculating results were four data and background rate, 0.5669 which can meet the best prediction results. Comparing with other traditional prediction methods, the prediction which applied Mix Gray Prediction and Genetic Algorithm is better then traditional ones. The results of the prediction can provide the government authorities and enterprise with a reference of electricity prediction in the future.