Adaptive and Intelligent Edge Computing Based Building Energy Management System


This project main result can be described as a smart platform able to efficiently manage the energy consumption in buildings, reducing the consumption by the means of suggestions that aims to a behavioral change in end users, and optimizing the power grid load. To this end, the project will advance in the state of the art of several disruptive research fields (edge computing, virtual organizations, social computing, XAI, DLTs…) generating multiple outputs, such as:

  • A complete smart platform based on virtual organizations and designed as a 3-tier architecture able to ingest data from multiple sources and provide tailored responses for an efficient energy consumption pattern (Software).
  • An IoT network and edge gateways able to satisfy the data ingestion needs of the platform, to be deployed in the pilot stage, and to encrypt data at the hardware level (Hardware).
  • A data security and privacy protocol integrating cutting-edge cryptography solutions and a DLTs based approach (Software).
  • A social machine able to manage the information processed by the platform, classifying and monitoring the information, identifying different scenarios and providing tailored responses (Software).
  • New XAI approach (Deep Symbolic Learning) using hybrid neuro-symbolic artificial intelligence algorithms for a better integration of machine reasoning and learning capacities. (Software).
  • New predictive and optimization models based on hybrid symbolic learning. (Software).
  • Datasets gathered from the buildings involved in the demonstration phase, that will retrieve information about consumption at buildings (anonymized/aggregated and plenty compliant with all ethic and privacy recommendations/legal framework) (Database).
  • Models (e.g. Energy consumption, suggestions vs. pattern modification, energy demand) considering also social and human behavioral aspects (Data correlation).