An Approach to Solve Transient Stability-Constrained Optimal Power Flow Problem Using Support Vector Machines. Issue 6 (3rd April 2017)

• Record Type: Journal Article
• Authors: Nguyen-Duc, Huy; Tran-Hoai, Linh; Cao-Duc, Huy
• Is Part Of: Electric power components and systems. Volume 45:Issue 6(2017)
• Page Start: 624
• Page End: 632
• Publication Date: 2017-04-03
• Subjects: power system dynamics -- power system stability -- power system optimization -- computational intelligence -- Support Vector Machines -- Optimal Power Flow; Electric machinery -- Periodicals; 621.3104205
• DOI: 10.1080/15325008.2017.1292328
• Languages: English
• ISSNs: 1532-5008

Abstract

Abstract: The Transient Stability-Constrained Optimal Power Flow (TSC-OPF) is a challenging optimization problem, and is the subject of several recent researches. This paper proposes a novel approach to solve TSC-OPF. In the proposed framework, Support Vector Machines (SVMs) are used to classify whether an operating condition satisfies predefined transient contingencies. A novel classification strategy is proposed to ensure the optimal solution satisfies all considered contingencies with certain security margin. Besides, the weight coefficients of the SVM are used as sensitivity measures in order to help optimization solver find solutions more effectively. The proposed approach is demonstrated for the New England system and the IEEE 300 bus system.