Analysis and Comparison for the Unit Commitment Problem in a Large-Scale Power System by Using Three Meta-Heuristic Algorithms

• Main Authors: Hung-Yi Chang, 章弘毅
• Other Authors: Yuan-Kang Wu
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/5fwrqc

碩士 === 國立中正大學 === 電機工程研究所 === 102 === This thesis applied three algorithms including charged search system (CSS), particle swarm optimization (PSO), and ants colony system (ACS) to solve the unit commitment problem in a large-scale power system. The three algorithms were written by Matlab programming and applied to 10-, 20-, 40-, 60-, 80-, 100-bus testing systems to solve the UC problem. Then, this study compares the total generation cost and calculation time obtained from the three algorithms. This thesis also discussed the tunable parameters in the three algorithms, and compared the solutions on UC cost and computation time by setting different parameters. Finally, this thesis considered the effect of wind power forecasting uncertainty on unit commitment, in which seven forecasting methods were utilized and the difference between forecasting and actual values represents wind power forecasting uncertainty.