Balancing REC self-consumption and DSO grid constraints through integrated power flow optimization

Renewable Energy Communities (RECs) are increasingly viewed as a strategic framework for achieving a climate-neutral economy, providing the grid with essential flexibility through coordinated self-consumption and distributed storage. However, the widespread integration of RECs introduces significant operational challenges for medium-voltage (MV) and low-voltage (LV) distribution networks, as Distribution System Operators (DSOs) must balance grid stability with the costs and benefits of decentralized production while REC members seek to maximize local energy sharing. Critically, when RECs optimize for self-consumption without grid awareness, their operational decisions may violate voltage limits and thermal capacity constraints, creating conflicts between community objectives and network reliability. This thesis investigates these operational conflicts through a three-scenario comparative optimization framework built on a common Mixed-Integer Linear Programming (MILP) formulation of REC dispatch, integrated with a linearized distribution power flow model (LinDistFlow/Branch Flow Model). In Scenario I, REC dispatch is optimized purely against REC-side economic objectives, with no representation of the distribution network and no enforcement of voltage or thermal feasibility. Scenarios II and III both couple the REC-side economic objective to a network-aware DSO-side objective within a single weighted multi-objective formulation, solved jointly, with the REC-side term weighted more heavily by design; the two scenarios differ only in whether the DSO-side objective is restricted to active power losses (Scenario II) or additionally penalizes voltage magnitudes exceeding a specified upper bound (Scenario III). The case study network is a radial IEEE 33-bus feeder hosting a REC with residential, commercial, and dedicated photovoltaic/battery storage members, representing a realistic distribution feeder under significant distributed energy resource penetration. Results quantify the trade-offs: grid-blind REC optimization could achieve the maximum self-consumption rate but may lead to operational limits violation, while pure grid optimization ensures reliability at the cost of community energy independence while limiting the self-consumption potential. The coordinated approach demonstrates that information sharing between RECs and DSOs enables feasible solutions that preserve substantial self-consumption while respecting grid constraints. This work provides DSOs with a practical framework for assessing REC integration impacts and designing coordination mechanisms that align community interests with network operational requirements.