Distribution system operators (energy trade)
Optimization of the management of a balancing group through the use of smart meter data
Distribution system operators are responsible for actively managing their balancing group. This means that they must ensure that production and consumption are in balance on a quarter-hourly basis. Deviations are costly because they need to be compensated by expensive balancing energy. Therefore, precise forecasts for the difference between planned and actual energy flows (differencetime series or DBA) are necessary.
An important factor for forecasting the DBA time series is the actual state of the balancing group. However, the actual measurements of the DBA time series are only available three to four days later. Therefore, a „Balancing Group Online Estimator” has been developed in cooperation with EnBW Trading department.
![](/wp-content/uploads/2024/07/boxen13-4.png)
Target Group
- Power Company
- Energy distribution companies
- Energy portals
- App and applicati
![](/wp-content/uploads/2024/07/boxen11-4.png)
Added values for your company
- Fast, simple and cost-effective integration into existing systems
- All advantages of “Software as a Service”, e.g. high availability, automatic updates, no maintenance effort
- Flexible scaling with growing corporate structure
- High degree of security due to secured and encrypted interfaces
SANDY AI Solution
- Based on several 10,000 smart meters, the deviation of the planned and actual load is predicted. This is used to generate day-ahead and intraday forecasts that are significantly better than forecasts without this information.
- Every minute a new forecast is computed to incorporate all smart meter data that is available already. This allows the DBA manger to work always with the best possible forecasts.
- The Balancing Group Online Estimator provides a good estimate of the current state of the DBA within 9 minutes. This results in an improvement of day-ahead and intraday forecasts of 10-20%
- Changes in the balancing group are taken into account by fully automatic training of the model parameters
- Automatic training, historicizing of models, permanent updates of forecasts and monitoring of results
- Improvement of the operating forecast by 10-20 % and sustainable reduction of the balancing energy costs for the balancing group