Reserve Capability Assessment Considering Correlated Uncertainty in Microgrid

This paper presents a novel approach of assessing the required reserve capability in order to meet the forecast uncertainty dynamics in microgrid. The historical data of forecast variables is used to generate energy balance scenarios. The dynamics of these scenarios are represented as instances between capacity, ramp-rate and ramp-duration variables. A polytopic model is used to enclose these instances with its surface defining the worst case scenarios. This approach captures the correlated nature of the dynamics and provides a compact representation. In relevance, the capability of the generators and the power import from the grid are modeled as convex envelopes. A vertex based method is proposed that allocates the polytope among the resource envelopes. During this process, the operational cost is minimized while considering the resource location and the network constraints. The proposed method is examined for a microgrid test case based on the CIGRE medium voltage network. The results show new insights in the allocated demand and reserve capabilities. As an additional result, it is observed that the reserve requirements can be decreased by allocating reserves for each time instance separately as compared to the fixed percentage of load approach.