Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
398926 | International Journal of Electrical Power & Energy Systems | 2013 | 7 Pages |
This paper deals with decision tree based power system security assessment for multiple contingencies. In this paper, we propose a contingency grouping technique to produce good, but reduced number of decision trees for multiple contingencies. The grouping is based on the degree of overlap in class boundary regions or post-contingency performance measures of every contingency, as captured by the graphical index progressive entropy. The proposed method has been demonstrated on French EHV system for a voltage collapse problem in west France, Brittany region.
► Proposes decision tree-based security assessment for multiple contingencies. ► Devised a novel graphical metric, progressive entropy, for deriving common trees. ► Illustrated on the Brittany region of French power system. ► Obtained lesser operating rules that performed well for many contingencies.