Back to top

Pilots' applications


Lead
12 large-scale pilots (LSPs) across 10 countries aim to cultivate the shifting towards predictive and prescriptive analytics and to enable multiple data source analytics for a wide range of applications. The large geographical coverage of the pilot sites aims to support the large-scale EU-wide replicability and market take-up of services and solutions in different socio-economical contexts to maximize the impact of BD4NRG services across Europe.
image/svg+xml
CENTRICA-Aggregator Belgium
EKL, BORZEN & CS-TSO & Power Market Operator Slovenia
ENELX-DSO Italy
TILOS-Smart Island Microgrid Greece
ENERCOUTIM-RES Operator Portugal
CARTIF & VEOLIA-ESCO Spain
AJSCV & VEOLIA-ESCO & Municipality Spain
LEIF-Fund Latvia
REN & R&D NESTER-TSO Portugal
ELES-TSO Slovenia
ASM-DSO Italy
OEDAS-DSO Turkey
0
Large Scale Pilots

BD4NRG framework is demonstrated and validated in real-life pilots in 12 demo-sites across 10 countries.

Operation of Electricity Networks

Body

BD-4-NET Large-Scale Pilots focus on predictive analytics towards forecasting and increasing the efficiency and reliability of the electricity network.

NET

Paragraphs
Accordion items
Body
  • WHERE: Portugal
  • DESCRIPTION: The pilot aims to assess the condition of two grid assets, Circuit Breakers (CB) and overhead lines, in order to provide answers to predictive maintenance and implementing maintenance plans. Both are core elements of the electricity network and represent different challenges related to data processing.
  • GOAL: Asset management departments of system operators are traditionally data intensive. This pilot will provide new data services that can be used to foster predictive maintenance strategies for system operators. A data analytics application will be developed to evaluate the operation of CBs and determine the probability of failure considering different operational conditions. Furthermore, inspection data will be used to generate a semi-automatic maintenance plan for the overhead lines. Main expected benefits will be consisting in optimizing maintenance costs and hence improve reliability of the grid operations.

Body
  • WHERE: Slovenia
  • DESCRIPTION: Operation, maintenance and planning activities of asset management life cycle are included to optimize the delivery of market transactions. The pilot aims to Data analytics to support the seamless execution of any market transaction with the assurance of adequate transmission capacity. While the capacity is inherently limited, it can be temporarily increased by the actions of power grid operators.
  • GOAL: Operation of the power system closer to the operational limits, with a large share of distributed renewable energy sources, requires new operating paradigm. So, a new service from the TSO to a market actor will be demonstrated, in which the TSO will consider temporarily changing the lifecycle delivery of grid assets to accommodate market transactions. For this purpose, an additional signal pathway will be included in the maintenance decision-making algorithm. Finally, the pilot will also propose the remuneration structure for the service.

Body
  • WHERE: Italy
  • DESCRIPTION: The pilot aims to deploy cross-functional predictive analytics to support integrated DSOs asset management and network operation. Specifically, new methodologies and tools will be used for the anticipated prediction of transformer asset failures that will support the efficient management of the infrastructure. A variety of local data-driven optimized energy management strategies and analytics will be deployed. (e.g., flexible scheduling of EV recharging). Also, economic incentives will be modelled by utilizing BD4NRG framework. Near real time smart meters will be leveraged and integrated with off-grid data. Finally, predictive analytics capabilities for optimal grid operation will be deployed.
  • GOAL: ASM Terni faces already severe network stability problems to the power distribution grid. The goal is to improve the asset management and the operational efficiency of the local smart distribution grid, while at the same time leveraging on and coordinating with commercial electricity end users, such as e-mobility operators. Hence yearly expenditures for operation and maintenance will also be optimized by advanced remote control of the asset.

Body
  • WHERE: Kütahya, Turkey
  • DESCRIPTION: The resulting huge amount of cross-functional data will be used to deploy Assets (Transformers) Predictive maintenance to support optimized grid planning and operation. AI will be used for forecasting future maintenance and failure of the network assets and hence failure along the grid. AI based predictive maintenance results will allow operators to avoid breakdowns, connect new supplies at the optimal points of the network (for MV) and plan investments in the places where they are most needed. It will be possible to diagnose the different problems to improve the maintenance and operation of the system, power quality and AMR systems.
  • GOAL: Due to the high penetration of PV, there are some power quality problems, such high voltage, voltage fluctuation, congestion, capacitive, etc. So, preventing of faults will increase customer satisfaction and reduce SAIDI, SAIFI indexes resulting to reducing costs associated with maintenance. Another goal is to achieve more accurate investment planning through analytics.

Management of Distributed Energy Resources

Body

BD-4-DER Large-Scale Pilots (LSPs) focus on optimizing the management of assets connected to the grid.

DER

Paragraphs
Accordion items
Body
  • WHERE: Belgium
  • DESCRIPTION: Residential flexibility that is valorised on multiple energy services is becoming increasingly interesting. Residential assets such as electric vehicles, electric boilers, micro-CHP’s or heat pumps are an untapped potential of flexible energy to integrate renewable energy by providing balancing services and mitigate grid congestion.  In order to control the assets efficiently, flexibility providers are building models of the assets to optimize their control actions. This can be a labour-intensive and time-consuming process that limits scalability. CENTRICA and IMEC are therefore approaching this challenge by leveraging on data-driven reinforcement learning algorithms.
  • GOAL: Validation of the developed cross-context transfer learning algorithms for residential flexibility asset profiling with a view to reduce the time necessary for model training and hence optimize management of asset. Moreover, proof-of concept setup of using residential DR assets to retrieve local grid information and use this information in the aggregation activations.

Body
  • WHERE: Slovenia
  • DESCRIPTION: This pilot aims to enhance the coordinated management of grid and third party-owned grid connected assets and optimize their flexibility potential to support near real time market transaction predictions and subsequent anticipated settlement. The DSO’s reactive and preventive operational applications will be complemented based on data analytics. Pilot’s aim is to create a fluid data backbone, establish services to support evidence-based business decisions. Smart meter data will be exploited to increase the efficiency and reliability of the network operation.
  • GOAL: DSO that are at the core of utility industry transformation and local energy systems are becoming key assets that need to be properly managed and coupled with the infrastructure. The pilot aims at exploiting smart meter data to increase efficiency and reliability of network operation and will result in the design of rules and complementary toolbox for implementation of the demand-response services and settlement among the involved and impacted parties.

Body
  • WHERE: Catania, Italy
  • DESCRIPTION: The pilot aims to use the data produced by the testing system to improve BESS reliability to predict critical events and reduce plant unavailability. The capability to identify unexpected energy storage system behaviors or underperformance areas will allow the maintenance optimization through the shifting from CBM (Condition Based Monitoring) approach to proactive predictive system analysis able to improve plant performance, while reducing operating and maintenance costs.
  • GOAL: The ENELX use case is focused on BESS (Battery Energy Storage Systems) plants integrated in the power system to provide ancillary services and will be deployed by ENELX company of the ENEL group. Big Data and AI/ML platforms and algorithms will be developed by complement in ENELX big data platform and the experimental data storage platform with BD4NRG edge enablers will optimize battery energy and power utilization to cover primary/secondary regulation services. Proper data analysis and selection approach will improve plant reliability and early-stage problem detection is proposed as well to improve plant performances.

Body
  • WHERE: Greece
  • DESCRIPTION: This pilot aims to upscale and further develop the local EMS to offer improved analytics-based forecasting for RES local generation and flexible/controllable loads. In that respect, this pilot will demonstrate BD4NRG at level of cross-domain analytics via integration of weather service data, mutual benefits from cross domain data sharing, integration of non-energy domain data (e-mobility, water network data), including EVs and water pumping systems as controllable electrical loads. The resulting more accurate prediction will support real time control, fast response – optimization, security of the energy availability and preventing power interruptions / shutdowns for residential and commercial customers, hence demonstrating as well BD4NRG framework at cross-functional level.
  • GOAL: In TILOS island a microgrid has been developed, comprising of a hybrid power station of RES-storage and a network of smart meters and DSM panels spread across the residential and tertiary sector of the island, also including 8 water pumping stations. The pilot aims at better quality of the provided energy, better and faster response to unplanned events and black-out avoidance, better matching of customers demand, better power system response, improvements of environmental footprint and reduced operational costs.

Body
  • WHERE: Portugal
  • DESCRIPTION: The aim of the pilot is to demonstrate BD4NRG at level of scalable AI-based analytics pushed at the edge of the network, through the edge-level optimized deployment of big data management micro-services close to the point-of-need to support AI-based analytics tailored to asset-level anomaly detection and prediction and provide an integrated support for near real time optimized decision-making on coordinated flexibility provisioning. Secondly, the proposed pilot aims to demonstrate scalable integration of cross-functional and cross-stakeholders heterogeneous data sources to support regional-level analytics-based cross-resources. These include additional renewable energy generation sources, water distribution, waste management, university campuses, and additional municipal assets.
  • GOAL: In ENERC Solar Demonstration Platform there are connected infrastructures with a variety of data sources from meteo to radiation to multivariate indoor sensors to energy smart meters and beyond. ENREC aims to gain improvement both in cost reduction for coordinated decentralized RES generation asset management and in the Virtual Power Plant optimized management. The pilot will also create the setting for a region-wide set of data lakes which will include the demonstration of efficiency gains in the water-energy nexus, identification of new value streams through data analysis and optimized implementation of decentralized energy systems based on big data analytics.

Investments & Efficiency in Buildings

Body

BD-4-ENEF Large Scale Pilots (LSPs) focus on increasing the efficiency and comfort of buildings, and de-risking investments in energy efficiency.

ENEF

Paragraphs
Accordion items
Body
  • WHERE: Spain
  • DESCRIPTION: This pilot aims to integrate vast amounts of data in a harmonised way by means of scalable Big-Data mechanisms, including cadastre, BIMs, and Energy Management Systems at Building and District level (data provided by VEOLIA partner as building/district operator), and will demonstrate BD4NRG at two different levels. First cross-domain predictive and prescriptive analytics will provide near real time dynamic support to predict the optimal building energy retrofitting planning dates. SecondlyBD4NRG cross-domain analytics will be tailored to predict EPC performance assessment by dynamically confronting EPCs and aggregated building-level energy consumption, with a view to provide actionable information to ESCOs for consequently adapting in automated way the energy consumption building-level profiles, and for supporting the initial stage of defining the EPC acceptable threshold via EPC reliability improvement in order to develop dynamic Energy Performance Certification (EPCs) to support public authorities, private investors, building owners, ESCOs.
  • GOAL: This will contribute to EPC being more reliable, user-friendly, cost-effective, better quality, and EU legislationcompliant, while contributing to remove “reliability” barriers which are actually hindering the high potential of EPCs to realize significant energy efficiency in the building sector. Consequently, the ex EPC usage would be increased across Europe and end-users. 

Body
  • WHERE: Spain
  • DESCRIPTION: This pilot will test big data analytics on merge buildings’ comfort, buildings renovation, smart grids, renewable energy production, demand analysis and predicting energy consumption. In particular, the Thermal Comfort Validator (TCV) tool will be leveraged, a cross-platform web-app that estimates the Predicted Mean Vote comfort inside a building to predict occupant’s comfort levels per proposed intervention. This will facilitate considering both energy consumption savings and thermal comfort in EPC and building renovation machine learning (ML) to bridge the gap between controllable building parameters and thermal comfort, by conducting an extensive study on the efficacy of different ML techniques for modelling comfort levels, while trading off with energy efficiency.
  • GOAL: Pilot’s main goal is to help public administrations foster new private-oriented business models to implement energy retrofitting to buildings. The main gained benefit will be consisting of finer grained and more accurate thermal comfort prediction, which is expected to be linked linearly with HVAC set points and actions on controllable loads, while taking into due consideration potential changes in energy consumption and how this could be reflected into building energy efficiency. Finally, the pilot will create new business models to increase the efficiency and comfort of buildings and big data analytics to provide information to the private sector.

Body
  • WHERE: Latvia
  • DESCRIPTION: LEIF is the only institution in Latvia that has reliable data on investments in energy efficiency and actual performance of investments in terms of energy savings. The pilot concept will demonstrate BD4NRG framework through cross-domain integration of a variety of heterogeneous data on financial performance, underlying energy efficiency impact of the investments. The aim is to collect and process data from smart meters, and apply ML algorithms, to better predict energy consumption and calculate and monitor the energy savings achieved. As a result, EPC and investments will be more reliable, cost-effective, and of better quality.
  • GOAL: The capability offered by emerging near big data analytics to integrate cross-domain financial and energy consumption is the key for building the necessary market confidence in energy efficiency projects. The use of historical data pooled from major market segments can encourage more energy efficiency investments and de-risk investments. The pilot will reduce the uncertainty linked to energy efficiency investments, strengthen debt and equity financing of energy efficiency projects. Specifically, the platform will provide investors/financiers and project developers the opportunity to evaluate quickly and easily KPIs for projects, explore past and future, make a profound risk measurement and evaluation.