Short-Term Market Operations of Interconnected Local Energy Communities Under Multivariate Uncertainty Dependencies in Distribution Systems

Increasing uncertainty and variability of distributed energy resources can have a significant impact on economic operations and market transactions in local energy communities (LECs). This requires an informed decision framework to account for the impact of multivariate uncertainty. Accordingly, this paper presents a stochastic decision framework for multiple LECs that considers the inter-dependencies in modeling multivariate uncertainty in the short-term market. It is formulated as a bilevel stochastic optimization, where at the upper-level, interconnected LECs optimize their operating costs using local energy resources, inter-peer-to-peer trading and grid transactions, while the lower-level problem represents the market clearing processes to determine a grid-informed price in the day-ahead and real-time market. A regular-vine copula is applied to establish the dependency structure in the cross-coupled uncertain variables. In addition, a conditional value at risk is considered to account for risk due to uncertainties. A decomposition-based method, the modified progressive hedging algorithm, is presented as the solution. The proposed problem and algorithm are implemented on a modified IEEE 33-bus system. Numerical results demonstrate that the proposed framework effectively describes the impact of inter-dependencies on the operational and risk costs compared to independent method. Furthermore, the algorithm effectively solves the problem in a reasonable time, showing its scalability and tractability.