Centralized stochastic energy management framework of an aggregator in active distribution network

Abstract

This paper develops a detailed and sequential procedure for short term operation of an aggregator to minimize the cost of the consumer and risk of the aggregator, which includes day-ahead (DA) and real-time (RT) operation. DA operation has two stages: consumer load scheduling and risk-based energy procurement. Consumer load scheduling implemented over a radial distribution network by a cost minimization objective function which considers electricity price and solar PV uncertainty, peak-to-average ratio, and phase unbalance. This model is formulated as a mixed integer non-linear program. In the second stage, a risk based energy procurement is formulated, here, the aggregator has choice of energy procurement from wholesale market either in DA market or RT market to meet the scheduled power. In RT operation, the aggregator take a decision on share of RT purchases and battery schedules to minimize the cost of scheduled load power deviations. In order to realize this, a rolling window optimization is implemented which is modeled as mixed integer problem. The framework is demonstrated with a detailed case study of 15 node radial active distribution network consisting of 420 number of residential customers.