Data-driven Analysis and Optimization of Low Voltage Networks

The increased use of renewable energy sources and the shift towards electromobility are leading to greater volatility in the energy system. Grid operators are faced with the challenge of a lack of high-resolution, real-time information on grid utilization. Forecasts are often based on incomplete or incorrect data - especially in connection with PV systems, charging stations and battery storage. This leads to inefficient use of resources and avoidable costs.

The aim of the DAWN project is to create a high-quality data basis through digitalization and the targeted use of data. This should both improve network calculations and identify optimization potential. In addition, the first data-driven forecasting models are being developed, which are important for both grid operators and energy suppliers.

Our work package focuses on analyzing real consumption and generation data. We are investigating key issues:

- Automated determination of customer data: How can information about PV systems, charging points for electric vehicles or the use of heat pumps be obtained from smart meter data?
- Improving synthetic load profiles: What representative consumption and generation profiles can be created for different customer groups?
- Determining flexibility: To what extent could customers stabilize the grid with battery systems and what capacities would be required for this?

Procedure and methodology

- Data collection and ensuring data quality: Definition of the required data and its visualization and quality check.
- Data analysis: Determination of consumption profiles and customer master data using innovative classification methods.
- Simulation and optimization: Development and testing of various battery strategies and determination of their impact on the grid.