Title: Data Anonymization and Aggregation Approaches for Local Energy Communities

In this paper, we propose an approach for data anonymization and aggre- gation for local energy communities (LECs). We utilize the k-anonymity principle with an efficient Mondrian k-anonymization algorithm, providing generalization and suppression to raw datasets to be shared across data spaces. We analyse a data aggregation approach, as a means for organiza- tion and representation of extracted data from LECs. Furthermore, we train Self-Organizing maps from datasets, to provide a model in a compact form that can be used by involved stakeholders for visualization and inference, as well as for projection of new datasets on LECs’ historical data. We present experimental results and discussion for the methods, based on synthetic datasets, for a LEC with 60 metering devices. We also discuss the potential of the proposed approches for deployment in energy data spaces, which are able to provide access to massive datasets via federation.