Visible to the public Multivariate Uncertainty Characterization for Resilience Planning in Electric Power Systems

TitleMultivariate Uncertainty Characterization for Resilience Planning in Electric Power Systems
Publication TypeConference Paper
Year of Publication2020
AuthorsNazemi, Mostafa, Dehghanian, Payman, Alhazmi, Mohannad, Wang, Fei
Conference Name2020 IEEE/IAS 56th Industrial and Commercial Power Systems Technical Conference (I CPS)
Date PublishedJune 2020
ISBN Number978-1-7281-7195-1
KeywordsCPS Resilience, cyber physical systems, decision making, Ellipsoids, Optimization, Power systems, Probabilistic forecasting, Probabilistic logic, pubcrawl, resilience, Resiliency, stochastic optimization, Stochastic processes, Uncertainty, uncertainty sets
AbstractFollowing substantial advancements in stochastic classes of decision-making optimization problems, scenario-based stochastic optimization, robust\textbackslashtextbackslash distributionally robust optimization, and chance-constrained optimization have recently gained an increasing attention. Despite the remarkable developments in probabilistic forecast of uncertainties (e.g., in renewable energies), most approaches are still being employed in a univariate framework which fails to unlock a full understanding on the underlying interdependence among uncertain variables of interest. In order to yield cost-optimal solutions with predefined probabilistic guarantees, conditional and dynamic interdependence in uncertainty forecasts should be accommodated in power systems decision-making. This becomes even more important during the emergencies where high-impact low-probability (HILP) disasters result in remarkable fluctuations in the uncertain variables. In order to model the interdependence correlation structure between different sources of uncertainty in power systems during both normal and emergency operating conditions, this paper aims to bridge the gap between the probabilistic forecasting methods and advanced optimization paradigms; in particular, perdition regions are generated in the form of ellipsoids with probabilistic guarantees. We employ a modified Khachiyan's algorithm to compute the minimum volume enclosing ellipsoids (MVEE). Application results based on two datasets on wind and photovoltaic power are used to verify the efficiency of the proposed framework.
Citation Keynazemi_multivariate_2020