Title :
The Application of Sequential Minimal Optimization Algorithm In Short-term Load Forecasting
Author :
Qian, Zhou ; Jie, Zhai Yong ; Pu, Han
Author_Institution :
North China Electr. Power Univ., Baoding
Abstract :
A new approach based on sequential minimal optimization (SMO) Algorithm for the electric power system load forecasting was presented. The proposed algorithm introduced the conception of limited memory, selected proper kernel parameters and improved SMO algorithm. SMO algorithm decomposes the QP problem in SVM into a series of sub block data to ensure the convergence. The Improved SMO algorithm can keep the length of the block data has remained unchanged and it provides the excellent forecasting accuracy proved by the result of the experiment.
Keywords :
load forecasting; power engineering computing; power station load; quadratic programming; support vector machines; QP problem; SVM; electric power system load forecasting; sequential minimal optimization algorithm; short-term load forecasting; Control systems; Convergence; Kernel; Lagrangian functions; Load forecasting; Quadratic programming; Support vector machines; Virtual colonoscopy; Limited memory; SMO; Short-term load forecasting;
Conference_Titel :
Control Conference, 2007. CCC 2007. Chinese
Conference_Location :
Hunan
Print_ISBN :
978-7-81124-055-9
Electronic_ISBN :
978-7-900719-22-5
DOI :
10.1109/CHICC.2006.4346950
