“20th PSCC 2018 papers submission and review platform

Full Program »

Prediction of Umbrella Constraints

Security-constrained optimal power flow (SCOPF) problems are essential tools to transmission system operators for long-term and operational planning and real-time operation. SCOPF is used in many system studies. However, the solution procedure of SCOPF problems is challenging due to the inherent size and scope of modern grids. As empirical evidence and past research show, relatively few constraints in SCOPF problems are necessary and sufficient to enclose the feasible set of solutions. These constraints are called umbrella constraints. An optimization-based formulation, called umbrella constraint discovery---UCD, has been proposed to identify such constraints. Additionally, it was shown that umbrella sets are relatively insensitive to load changes in the system. This paper proposes use of a data-driven approach to predict the umbrella set of a system based on the UCD results for similar conditions. The proposed approach is based on the cyclic nature of electric loads as well as the relative insensitivity of umbrella sets to the system parameter changes. Artificial neural networks have been used for this purpose. The results show the applicability of such methods for prediction of umbrella sets.

Author(s):

Ali Jahanbani Ardakani    
Iowa State University
United States

Francois Bouffard    
McGill University
Canada

 

Powered by OpenConf®
Copyright ©2002-2014 Zakon Group LLC