A combinatorial multi-objective optimisation model to support the design of urban renewable energy communities

This study proposes a novel optimisation framework for the design of urban renewable energy communities, addressing the challenge of aggregating end users in planning processes hindered by conflicting objectives and the absence of a scenario-based methodology. We developed a computational tool that integrates multi-objective optimisation with multi-criteria decision-making techniques to identify optimal district configurations for urban renewable energy community projects. Specifically, the tool allows the grouping of potential members based on environmental, economic, and technical criteria across different district scales and multiple scenarios. First, we elaborated a combinatorial optimisation problem based on the set partitioning problem, adapting the framework to numerous non-linear objectives. The algorithm was implemented in Python, including an integrated version of the Non-dominated Sorting Genetic Algorithm II with an evolution strategy and the k-means clustering algorithm. We incorporated the Technique for Order Preference by Similarity to Ideal Solution to rank the district configurations based on stakeholder preferences. We applied the algorithm to a real-world case study and carried out a comprehensive sensitivity analysis of the model. The proposed methodology enables the identification of optimal end-user groupings and the quantification of key performance indicators. This research contributes to the advancement of top-down energy community development in urban contexts. This framework assumes a virtual energy exchange compliant with current European regulations, which may differ in other jurisdictions. Further research is needed to assess the model's transferability to regulatory contexts that allow direct peer-to-peer sharing or where different energy carriers are prevalent.