Advantage
Precise forecast of network loss for the coming day
Data
Transmission system operator database
Methods
Data mining, cluster analysis, random forest, generalized additive model (GAM) and gradient boosting
Challenge
The transmission system operator 50Hertz ensures that electricity in northern and eastern Germany reaches where it's needed: the customers. However, physical losses occur during the electricity transmission process. These grid losses must be compensated by purchasing the forecasted amount of lost electricity through an electricity exchange.
Goal
To ensure that the client knows exactly how much electricity they will have to compensate for grid losses, precise forecasts are essential. 50Hertz therefore calculates the potential loss the day before using various statistical models. To further optimize grid loss forecasts, data sets containing actual grid losses are evaluated in combination with the forecasted feed-in of wind and photovoltaic energy.
Three steps to your successful data product
Solution
To identify potential patterns in the data of observed grid losses, eoda uses data mining techniques: Information is filtered from data sets, new variables are extracted, and aggregated using cluster analysis.
For the client's specific use case, eoda selected the appropriate method set and tested the forecasting performance of Random Forest, Generalized Additive Model (GAM), and Gradient Boosting. The models with the highest accuracy were presented to the client.
Result
Using the models provided, 50Hertz can more accurately forecast the magnitude of grid losses for the coming day and thus compensate for the electricity volume more cost-effectively on the exchange.
The solution lies in the data: A seemingly sparse database is analyzed using data mining from eoda – the result is reliable forecast models that enable customers to plan and therefore cost-effectively purchase the amount of electricity lost due to grid losses.
Get started now:
We look forward to exchanging ideas with you.
Your expert on Data-Science-Projects:
Manfred Menze
projects@eoda.de
Tel. +49 561 87948-370
