AN OPTIMIZATION APPROACH TO UNCERTAINTY PROPAGATION IN BOUNDARY LOAD FLOW
Andrija T. Sarić, A.M. Stanković · 2005
Abstract − Boundary load flow exemplifies a class of power (load) flow models that explicitly incorporates uncertainties in system data. Its importance for operation and control of modern power systems is increasing due to changes in ways the system is operated following deregu-lation and due to the emergence of new types of sources (e.g., distributed generation). Modeling based on interval computations is a rigorous mathematical tool for worst-case analysis of uncertain systems. This paper proposes an optimization-based approach to minimizing uncertainty spread in a boundary load flow model. The approach combines interval computations with a linear program-ming-based heuristic that is effective in avoiding the ex-cessive conservativism often associated with direct inter-val computations. We consider the case of uncertainty in both measurements (e.g., SCADA) and network parame-ters. The method is potentially applicable to large-scale power systems, and we study its performance on a two benchmark examples: New England/New York intercon-nection with 68 nodes and standard IEEE test system with 300 nodes.