Research project Opti-VNL

Optimised procedures for vertical grid load forecasting using machine learning

Credit:

BMWE

The reliable operation of power grids is complex, as the loading of individual power lines is hard to predict due to the fluctuating nature of renewable energy sources. This can lead to renewable energy plants being curtailed more than necessary, or expensive power plant capacities having to be brought online on short notice to keep the grid stable (so-called redispatch, i.e., system-stabilizing adjustments of the generator’s unit commitment). Against this background, the research project Opti-VNL, which is funded by the Federal Ministry for Economic Affairs and Energy, aims at improving the forecast of the vertical grid load (Germ.: Vertikale Netzlast; VNL), i.e., the loading of the transformers that connect the different voltage levels of the electricity grid. The project contributes to a more efficient use of renewable energy and to keeping the power grid stable.

Research project Opti-VNL

 

Duration

April 2025 to March 2028

Funded by

Federal Ministry for Economic Affairs and Energy

Project participants

  • Institute of Networked Energy Systems
  • Institute of Solar Research
  • EWE Netz GmbH
  • Siemens Energy Global GmbH & Co. KG
  • University of Oldenburg
  • Westnetz GmbH (associated)

Specifically, Opti-VNL involves the development of novel machine learning methods for short-term forecasting of the VNL with integrated uncertainty modeling. The goal is to be able to react quickly not only upon feed-in fluctuations, but also upon network reconfiguration (switching) and fault conditions, thereby avoiding unnecessary curtailment of renewable generation. At the beginning of the project, reference models will be built that serve as the baseline for a comparative assessment of the developed approaches for modeling and forecasting the VNL. The next steps are to examine existing forecasting methods for their ability to incorporate uncertainty models and to implement new methods for quantifying uncertainties, particularly those arising from weather scenarios.

Real-world data provided by the industrial project participants will be considered in the grid state evaluation, taking into account the assessment and propagation of uncertainties. The study will investigate how the injection of fluctuating renewable energy in the considered grid area could affect the VNL forecast. In a case study, the added value of the enhanced VNL forecast for the congestion prediction will be demonstrated. Finally, the developed methods will be packaged into a toolbox that is supposed to assist grid operators in their grid management and redispatch processes.

Grid operation will be modeled and evaluated primarily by the Institute of Networked Energy Systems. Their tasks include modeling and simulation to predict congestion in upstream 110-kV grids as well as the optimized curtailment of renewable energy plants. Additionally, they will process meteorological data, refine methodologies and analyze simulation results for use in VNL forecasting. In collaboration with the industrial project participants, the institute will also develop a monitoring system for VNL forecasting under changing weather situations. To this end, the project team will optimize camera-based methods and satellite image-based methods and compare them with conventional forecasts based on weather models. The comparison will focus on weather situations initially identified as relevant for redispatch. Finally, the Institute of Networked Energy Systems will define suitable metrics to assess the technical and economic benefits of the new forecasts from which the impact of ultra-short-term PV generation predictions on the derived VNL forecasts can be quantified.

Research infrastructure

Cloud camera network Eye2Sky

Eye2Sky can be used to generate short-term forecasts of solar radiation with unrivalled spatial and temporal resolution. This data can be used to stabilise the electricity grid, among other things.

Departments, research groups and challenges related to the Opti-VNL project:

Department

Energy Systems Technology

The Energy Systems Technology department focuses on the interaction between system-relevant technologies within decentralised networked structures, particularly at the low-voltage and medium-voltage level.

Department

Energy Systems Analysis

The Energy Systems Analysis department generates system-analytical knowledge, which we provide across sectors up to the global level and based in part on methods and modelling tools developed in-house.

Research Group

Flexibilities and Ancillary Services

The Flexibilities and System Ancillary Services research group is investigating which tools, operational management strategies and additional infrastructure are required in order to operate the electricity grids in a stable, reliable and robust manner.

Research Group

Power Grid Technologies

The Power Grid Technologies research group maps the complex interaction between controlling, controllable and non-controllable grid participants in our laboratories in order to understand and optimise its functioning at software and hardware level.

Research Group

Energy Meteorology

In the Energy Meteorology research group, we analyse how various observation systems and numerical models can be methodically optimised. We operate our own measurement networks for ground- and satellite-based collection of energy meteorological data to provide optimal information for the energy supply.

Challenge

Strategies for sustainable energy systems

Energy transition is a dynamic process that must be continuously adapted to new influences such as technical innovations or socially negotiated framework conditions.

Challenge

Balancing energy supply and demand

The simultaneity of electricity generation and consumption is the basis for a stable energy system. While fossil and nuclear power plants have so far been deployed according to demand, the dominant renewable energies in the electricity system are heavily dependent on weather - and therefore not available at any given time.

Challenge

Grid management: Electricity, gas, and heating grids

To ensure that energy is not only available on the balance sheet, but also in the right place at the right time, electricity, gas, and heating grids must be analysed in anticipation of future challenges, adapted in their operational management and modernised and expanded in terms of infrastructure.

Contact

Flexibilities and Ancillary Services

Research Group
Institute of Networked Energy Systems

Power Grid Technologies

Research Group
Institute of Networked Energy Systems

Energy Meteorology

Research Group
Institute of Networked Energy Systems