Direct Policy Search for Multiobjective Optimization of the Sizing and Operation of Citizen Energy Communities

Abstract

Citizen Energy Communities (CECs) are increasingly promoted in the European Union and beyond as a form of locally operated microgrids. While traditional microgrid research is often focused on an optimized operation, individual preferences regarding conflicting objectives are becoming more important in such communities. In this study, we present an evolutionary algorithm that has previously been used for a multiobjective operation of microgrids and include the perspective of heat consumption and initial sizing decisions using direct policy search. This way, the developed tool can be used by CEC planners to integrate conflicting objectives of residents in the installation phase. We introduce the algorithm formulation and demonstrate its functionality on a case study for different ambient conditions. The results show the opportunities to size and operate CECs through the presented algorithm.