Distribution system operators

Generation and load forecasts of power grids

Since last year, all European transmission system operators have been obliged to provide generation and load data according to the Generation and Load Data Provision Methodology (GLDPM) for a pan-European network model. The aim of the procedure is a transparent, pan-European overview of the capacities of the European electricity grid down to the last detail level of the individual electricity generator. The respective distribution grid operators are responsible for the data collection of the generation and load data as well as their forecasts. 

The common grid model is intended to show the changes in the high-voltage or high-voltage grid in order to make the resulting effects on other grids easier and more reliable to identify. Together with and for networks BW, SANDY developed the corresponding forecast models and made the 24/7 operational data available. 

In order to meet the European requirements of a standardized grid model for the calculation of power grid capacity, the „Generation and load data provision methodology“ (GLDPM), distribution system operators are required to provide a daily forecast of the grid status in order to pass it on to the transmission system operator. To ensure forecast quality, both changes in the grid switching state as well as grid topologies must be taken into account. Furthermore, the forecast results must be reproducible
Target Group

 Power Company

Energy distribution companies

Energy portals

App and application developers

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

no use of personal data

flexible scaling with growing corporate structure

high degree of security due to secured and encrypted interfaces

SANDY AI Solution

Fully automatic provision of day-ahead and 7-day-ahead forecasts for 2,000 transformers under consideration of the individual grid topology

› Forecasts broken down into a total of 17 load types and energy types, thus up to 36.000 forecasts per hour

› Fully automated self-learning model parameter training for best forecast results under changing framework conditions

Storage of the forecasts, incl. input data and models for 28 days

Deployment of operational AI in critical infrastructure with high data throughput

Get this information as PDF!

We are happy to assist you!

Take advantage of the innovative SANDY concept and contact us today! We look forward to your inquiry: