Development and Application of the Output Model for Wind Turbines

• Main Authors: Chin-yen Chen, 陳清嚴
• Other Authors: Horng-Ching Hsiao
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/13656267445780458891

碩士 === 國立臺灣科技大學 === 電機工程系 === 96 === The purpose of the study is to establish highly reliable wind power generation output model in order to achieve the object of incorporating wind power generation into the power system. The wind power generation has been developmented significantly for several decades, but how to combine the wind power generation into the electric power system is the core topic of technology exploration in recent years. And how to incorporate the wind power generation output into the electric power control operation is the key to the practicability of wind power generation. To establish a reliable and stable wind power generation output model, the study investigate the environmental variables of wind turbine output model, definition, categorization and comparison of regression analysis respectively. It is found that the polynomial regression analysis could be the representation of the best characteristics of wind turbine output. A six order polynomial mathematical model is adopted for the nonlinear model of multivariable output curve for wind turbine. The study compared and analyzed the deviation of 8 wind turbines at Jian-Shan Wind Farm in Penghu County based on the location, months, seasons, and different turbine manufactures as the case study. To verify the correctness and reliability of the wind turbine output model, the deviation rate between the simulated output and the actual commercial operational output are compared. Finally, the Jian-Shan Wind Farm output model and coefficients are used to predict the possible output and compared with actual commercial operational output as well as deviation calculation. The result showed the monthly and daily deviations between the simulation of wind turbine output in January 2007 and the measured output in January 2006 are less than 6% daily. The monthly averaged predict deviation rate is 2.35%. the result showed the wind turbine output models derived were extremely valuable for assessment of power system operation.