Energy value chain vision: To develop advanced distributed edge AI technologies for applications and services at the intersection of mobility, energy and building industrial sectors to support the European stakeholders driving the engine for innovation and entrepreneurship in sustainable energy. The value chain integrates the edge AI technological developments into four demonstrators.
VCD 2.1: Intelligent Water Tank Energy Optimisation demonstrator (lead: FE, NXTEN, NXTECH) – aims to provide an intelligent, scalable, automated, user-friendly energy monitoring, analysing, and controlling solution within a building that enables self-learning using ML and DL algorithms adapted to different configuration of a local energy network. The goal is to implement an edge computing topology with embedded AI into the micro- and deep-edge nodes to increase the processing and sensing capabilities of the edge devices.
VCD 2.2: Embedded Edge Approaches for Intelligent Distributed Energy Systems demonstrator (lead: SINTEF, FE, NXTEN, NXTECH) – aims to investigate embedded edge approaches for developing a system that provides automated edge processing with AI capabilities, as well as mesh communication among devices, to enable the autonomous operation of appliances, solar installations, storage, and electric vehicles to self-regulate and optimise the consumption and exchange of energy.
VCD 2.3: Mesh Intelligent Lighting demonstrator (lead: PROLUX, SINTEF, NXTECH) – aims to develop AI-enabled hardware/software nodes that cooperate with the driver electronics of the light fixture, and a reconfigurable AI-based mesh architecture. Such AI-nodes can provide human centric lighting, auto commission systems, usage pattern recognition, recognition of individuals and advanced sensor usage in general without high bandwidth networks and central processing.
VCD 2.4: Efficient Attack-detection Based on Neural-networks for Security Sensitive Systems demonstrator (lead: NXP-DE, UZL) – aims at providing an intelligent monitoring module integrated in the edge hardware to detect and react against possible physical attacks. It further aims at providing the edge hardware with the following security features low power, low latency, and yet robust attack detection.; AI-based attack detector with False Alarm Minimization (FAM) technique and 100% attack detection accuracy with 0% instance level false alarm rate.