2024

ekspertyza

angielski

EN This deliverable offers a comprehensive overview of the outcomes of RENergetic work package 3 (WP3). It is a relevant document for project members, as well as for external entities who have the intention of forming an energy island (EI) and require a supporting ICT solution. It summarizes the main concepts and functionalities included in an ICT system for energy islands. Evaluation of these functionalities is provided in the WP7 deliverable: D7.5 - Final evaluation of common demonstration results & impact. This document is the second and final deliverable of WP3. In the previous deliverable D3.1, completed in April 2022, the interim version of the ICT RENergetic system is described. In this deliverable, the description of the final versions of these functionalities is given. A comprehensive description of data models and the system architecture supporting these functionalities is provided. The main concept of the RENergetic data model is an asset – an abstract representation of any device in an energy island. Assets can be related to other assets, users, and measurements. Measurements represent any time series data that is collected by sensors in an energy island. All functionalities of the RENergetic system rely on the data model. Functionalities are implemented as independent software microservices. This allows the RENergetic system to be flexible for further extensions and allows the support of various functionalities in different energy island installations. During the lifetime of the project, the RENergetic system is deployed to the supercomputing platform provided by a project member – Poznan Supercomputing and Networking Center, PSNC. One of these services provides the forecasting of time series data. The results of this service are used by various other services, such as multi-vector optimization, demand response services and dashboards of the graphical user interface (GUI). The optimization is performed in a hierarchical manner with two levels: a global multi-objective multi-vector optimization and domain-specific optimization. The multi-objective optimization service finds optimal schedules for flexible devices in different energy sectors, as well as for devices connecting multiple energy sectors, e.g., heat pumps. Domain specific optimizer utilizes the schedules of the multiobjective optimizer as guidance for the local domain specific optimization process, but also considers additional domain-specific constraints and local context to determine the local schedules. From an algorithmic point of view, there are three types of demand response services implemented in the RENergetic system: rule-based, scenario-based and reinforcement learning-based demand response. The different implementations of demand response allow energy islands to better integrate them into the operation process. In the RENergetic project, these algorithms are applied to organize demand response trials in pilot sites. The GUI of the RENergetic system contains dashboards, other data visualizations and key performance indicators defined in the project.

https://www.renergetic.eu/wp-content/uploads/2023/03/D.3.1.pdf

2024

inne