Loading...

Online voltage security assessment based on wide-area measurements

Beiraghi, M ; Sharif University of Technology | 2013

877 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/TPWRD.2013.2247426
  3. Publisher: 2013
  4. Abstract:
  5. Online voltage security assessment is necessary to choose appropriate remedial and corrective actions in order to prevent a large-scale blackout. This paper presents a new online voltage security assessment method based on wide-area measurements and decision-tree algorithm. For the predicted load and generation variation scenarios (i.e., a day ahead), the database is obtained using continuation power flow. Moreover, each operating point is labeled as 'secure' or 'insecure' from voltage stability points of view based on WECC voltage security criteria. Decision trees are trained on the subset of the existing data by applying two famous splitting rules and various predictors. Bagging and adaptive boosting (AdaBoost) methods are employed to generate a combined model in order to increase the accuracy and eliminate dependency on selection of the predictors. Subsequently, this combined model is applied to predict the voltage security of the power system using continuous wide-area measurements. Since prediction by decision trees is very fast, the procedure will be completely online. The proposed method is tested on an IEEE 118-bus test system. The results indicate the applicability on the supervisory control and data-acquisition/energy-management system of a realistic power system
  6. Keywords:
  7. Decision tree (DT) ; Wide-area measurements ; Continuation power flow ; Corrective actions ; Decision-tree algorithm ; Large-scale blackout ; Operating points ; Supervisory control ; Voltage security assessment ; Wide area measurement ; Adaptive boosting ; Geodesy ; Information management ; Surveying ; Decision trees
  8. Source: IEEE Transactions on Power Delivery ; Volume 28, Issue 2 , March , 2013 , Pages 989-997 ; 08858977 (ISSN)
  9. URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6478857