Applying Markov chain model to forecast short-term wind speed

• Main Authors: Chung-Han Yang, 楊仲涵
• Other Authors: Hsu-Hao Yang
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/37908165537578491490

碩士 === 國立勤益科技大學 === 工業工程與管理系 === 101 === 　　In recent years, global warming and energy crisis issues have become significant. Pollution-free andlower cost of wind power is becoming the global mainstream. Wind speed is the most important factor to impact the generation of wind power. Building the wind speed model for simulation and predictions becomes an important issue. 　　This studyinvestigatesdifferentMarkov chain modelsbased onclassified states and wind speed conversion for wind turbines in Taiwan.We further develop the first, second and third-order Markov chain to compare the simulation results of the statisticalparameters,error index and autocorrelation function to identify the most suitable Markov model. The results show that the speed simulated by Markov chain, which can effectively preserve the original speed of mean and standard deviation, and the second-order Markov chain is better than first-order and third-order. 　　The results of this study cansimulate the wind speed model accurately so that turbine procurement, construction and distribution of electricity can be greatly benefited.