The thesis develops a planning framework for ADNs to achieve their dispatchability by means of ESS allocation while ensuring a reliable and secure operation of ADNs. Second, the framework is extended to include grid reinforcements and ESSs planning. Finally, the thesis also develops a distribution network expansion planning (DNEP) strategy to consider hosting newly integrated stochastic generation resources and customers.
The first part of the thesis develops the framework for the optimal sizing and siting ESSs to achieve ADNs dispatchability. The planning strategy is developed by embedding operational constraints of the grid by means of both linearized and exact optimal power flow models. In this regard, the proposed formulation relies on the so-called Augmented Relaxed Optimal Power Flow (AR-OPF) method: it expresses a convex full AC optimal power flow, which is proven to provide a global optimal and exact solution in the case of radial power grids. The AR-OPF is appropriately modified to be coupled with the proposed dispatching control resulting in a two-level optimization problem. In the first block, the optimal level of dispatchability and the ESS allocation is determined based on the trade-off between the imbalance penalty cost and the ESS investment cost. The following block determines the site and size of the ESS assets by evaluating the system states over operating scenarios representing seasonal variability and prediction uncertainty. To solve such a large-scale planning problem, the Benders decomposition technique is utilized to break down the planning problem into investment and parallel operation problems, each representing the daily operation of a typical day type.
The second part of the thesis tackles how the optimal investment in existing lines has to be integrated to develop a joint planning problem considering ESSs and line reinforcement to achieve the optimal dispatch level while ensuring sufficient hosting capacity for increasing stochastic renewable generation. The line reinforcement investment is suitably modeled along with corresponding adjustments on the network admittance matrix and the grid constraints to be incorporated in the modified AR-OPF model.
Finally, the above methods are extended to formulate a distribution network expansion planning (DNEP) strategy that considers hosting newly integrated stochastic generation resources and customers. The investment in new lines is suitably modeled and embedded into the AR-OPF model-based operation. This is achieved by converting the model to account for the change of network topology and adjacency matrix associated with assets investment. The large computational burden of the DNEP problem is mitigated by employing a specific sequential algorithm that consists of two sub-stages: the first sub-stage sequentially integrates new nodes while determining lines for reinforcement and nodes for ESS allocation, while the second sub-stage determines the capacity of lines and ESSs. The performance of the proposed methodology is verified through simulations on various sizes of distribution networks by showing that it can determine the optimal level of dispatchability while securing the required hosting capacity of the ADN under increasing stochastic prosumption.