AGENS Neural Networks (NN)

Forecasting the energy demand of individual actors based on time series involves processing vast amounts of data due to the large number of energy consumers. Currently, every commercial customer consuming more than 100 megawatt hours per year is subject to registration power metering (RLM). Individual consumption forecasts are created for each customer. To manage this complexity using mathematical models, flexible data-driven solutions are essential. While the extensive data presents challenges for analysis, it also enables the data-driven modelling of complex behavioural patterns. The key challenge, however, lies in applying this model to accurately forcast each individual actor's energy demand.

The goal of AGENS is to develop flexible neural network (NN) based models that are able to model the overall complexity using large amounts of data. In order to enable a robust forecast per actor, an improvement of the data quality for each individual consumer is necessary.