Multi-objective vector evaluated PSO with time variant coefficients for outlier identification in power systems

Author

Feng, Li ; Liu, Ziyan ; Ma, Chao ; Huang, Lin ; Zhao, Li ; Chen, Tao

Author_Institution

Chongqing Electr. Power Corp., Chongqing

fYear

2008

fDate

1-4 Sept. 2008

Firstpage

1

Lastpage

6

Abstract

A generic method for multi-objective optimization for bad data identification is presented based on multi-objective vector evaluated particle swarm optimization (VEPSO) algorithm. This multi-objective VEPSO is made adaptive in nature by allowing its vital parameters to change with iterations. This adaptability helps the algorithms to explore the search space more efficiently. After the bad data are detected, eigencurves of correlating load extracted by Kohonen network are used to modify the bad data. The application of the proposed clustering algorithm to the problem of unsupervised classification of electric load data is investigated. The results of simulation show the effectiveness of the algorithm.

Keywords

iterative methods; particle swarm optimisation; power system identification; Kohonen network; clustering algorithm; electric load data unsupervised classification; iterative methods; multiobjective vector evaluated particle swarm optimization algorithm; power system identification; power system operation data; search space; Change detection algorithms; Clustering algorithms; Data mining; Machine learning algorithms; Particle swarm optimization; Partitioning algorithms; Power system security; Power system simulation; Power systems; Space exploration;

fLanguage

English

Publisher

ieee

Conference_Titel

Universities Power Engineering Conference, 2008. UPEC 2008. 43rd International

Conference_Location

Padova

Print_ISBN

978-1-4244-3294-3

Electronic_ISBN

978-88-89884-09-6

Type

conf

DOI

10.1109/UPEC.2008.4651496

Filename

4651496