About INFUSE
Aim and Mission
INFUSE aims at developing a platform allowing to extract information from various sources of data – including high reporting rate measurements delivered by the new generation of smart meters (1 frame/s) and phasor measurement units with a reporting rate up to 100 frames/s. This is answering to the major challenge electric power systems face today, i.e. controlling the active distribution grids which are characterized with less and less available inertia, i.e. with high dynamics not yet studied. Discrimination from dynamic/fault behaviour from steady state operation of a grid using static converters- mediated energy transfer is requiring advanced models for each infrastructure to enable planning and realtime control and therefore INFUSE directly answers and addresses the green energy transition in all sectors of the energy system while ensuring security of supply.
Activities are grouped into five technical WPs and envisage development of a scalable and modular cloud Fiware – based platform allowing integration of heterogeneous multi-vector data sources by standardized API. The main goal is to derive digital twins for selected use cases (campus grid distribution network, prosumer microgrid with PV, EV charging and heat pumps) enabling advanced high resolution models parametrized with low time resolution information like weather forecast, e-mobility, social events etc. to achieve system resilience, flexibility, interoperability with minimal investment. Then we will develop an intelligent data analytics engine that can process and correlate information from various sources at the process level enabling anomaly detection, predictive maintenance and energy control in reduced inertia grids. Once digital twins are validated (for selected use-cases, most relevant for each partner/country), applications like distribution state estimation and control algorithm for microgrid operation will be developed and tested in specific networks for realtime information mediated by the INFUSE platform. For example, partners in Czech Republic will use the hybrid (PMUbased) state estimators for network reconfiguration in various scenarios (ensuring maximal energy transfer, minimal impact on energy users, lower stress on installations etc.)
Smart Grid Intelligence
Digital Energy Platform
Moreover, development of advanced metering infrastructures (AMI) for low-voltage (LV) networks enables the fine-grained collection of rich and diverse datasets for in situ power quality, assets integrity (PV panels) and quality of service assessment. Resulting datasets support detection and labelling of micro-scale events that are affecting the correct operation of LV networks and have been so far overseen through window-based averaging using typical approaches and measurement equipment. However, methods and techniques to first detect and label such events as anomalies in a data processing and learning pipeline are not trivial, especially for information delivered with high reporting rate. Subsequently, the labelled datasets are used either in a forecasting framework or as an early-warning system for potential imbalances in the local energy network. One key novelty is the combination of extracted features using time series data mining methods, such as the matrix profile, with state-of-the-art machine learning algorithms, including automated machine learning to
optimize classification metrics in real time, across various model/algorithm structure. We propose embedding such preprocessing algorithms for deriving power profile prosumer models.
In this way the proposal is answering to several key challenges mentioned in the Call: Data platforms/APIs,
interoperability, and standardisation, Resilience and data security, Artificial intelligence/machine learning, Big data and management of big data from different sources, smart metering, sensors, and automation. Consortium will benefit from experience and tested solutions in previous ERA-NET projects like FISMEP and I-GRETA (UNSTPB, GRE), advanced testing infrastructure and now-how related to asset monitoring (ICPE), PMU deployment and optimal placement in view of hybrid state estimators (UNSTPB, BUT), microgrid planning and operation (UNSTPB, BUT) and industrial software development (ELVAC). Real-life testbed will be provided by GRE, an NGO with prosumers installations where we are able to test various network configurations, emulate power profiles and enable flexibility either using energy storage or modifying behavioural use of energy.