Energy prices, both on the energy producer side and on the consumer side, are becoming increasingly dynamic and difficult to predict. Energy-dependent operations therefore want to respond quickly and flexibly to input (energy demand) in order to achieve the best possible balance in relation to output of energy/heat.
In district heating networks, energy demand must be met at all times. In biomass (heating) power plants, the biomass boiler often cannot react quickly enough to fluctuations in demand and peak loads. To be able to meet the demand, a fossil fuel-fired peak load boiler is usually used. This is on the one hand cost-intensive and on the other hand causes high emissions.
The toolbox “Heat demand forecast” provides basically available data, for example external weather data or temperature measurements from plant data, in a suitable form for the specific location and further processing. This allows a suitable forecast model to be trained from historical data using machine learning to generate a data-based forecast model rolling into the near future. Based on the prediction, control values can be programmed in such a way that an optimal buffer performance adapted to the situation can take place.
This continuously adjusted forecast enables proactive response and smoothing of peaks. This, in turn, allows for smoother boiler operation, lower emissions, and cost savings from eliminating or reducing peak boiler operation times.
In the case of energy generators for heat and electricity with several boilers, the energy balance can be optimally balanced between energy self-consumption and energy sales according to the objectives.