Motivated by the government's clean energy targets, the penetration of Distributed Energy Resources (DER) is increasing. These DERs interconnections bring the added generation and storage capacity at the distribution level. Also, with the increasing implementation of smart inverters and Advanced Distribution Management Systems (ADMS), the flexibility of the DERs can be leveraged to solve the distribution grid issues like over abnormal voltages, intermittencies, and thermal overloads. The proposed work focuses on the development of a robust distributed control architecture to control and optimally manage the load and PV variations using energy storage by creating the virtual clusters of the distribution grid.
The proposed approach adopts a distributed control approach by partitioning the grid into manageable clusters through an improved spectral clustering-based approach. The cluster of the distribution grid represents a good balance of local load and DER generation. An approach for reactive power to voltage sensitivity is also proposed for voltage regulation purposes at the cluster level based on the grid measurements. The clusters configurations adapt to accommodate the varying grid topology or changing load and DER generation. For the distribution grids receiving the set-points at the substation level for management of transmission power flow, an Alternating Direction Methods of Multipliers (ADMM) based sharing optimization is proposed to share the area set-points among the clusters based on the state of charge of each cluster. The controllable assets of each cluster (BESS) are managed through model predictive control. The improvement in the grid resiliency is demonstrated through the implementation of the complete framework to support the loads on the healthy part of the grid without interruptions during the contingency/outage scenarios on the distribution grid.