ENTROPIA
In an increasingly dynamic environment, district heating operators are controlling their systems with outdated technology, resulting in new problems. Rabensteiner Engineering GmbH is therefore developing a model predictive control system for heating (and power) plants, which should be universally applicable in the future.
Low efficiency
Unnecessarily high system temperatures lead to high grid and storage losses. Control without forecasting means that boilers often operate in the inefficient load ranges.
Difficult connection with other systems
High system temperatures and the inability to use flexibilities lead to difficult coupling with renewable energy (e.g. solar thermal), the dynamic electricity market, and other sectors (e.g. industry).
Lack of qualified boiler operators*
The increasing demands on system control due to the general conditions are exacerbated by the lack of qualified boiler operators
* Bundesministerium für Inneres, 2024. Bundesweite Mangelberufe. www.migration.gv.at
In the past, district heating systems always had a central heat generation plant that supplied all heat consumers in the system. However, the integration of distributed renewables leads to decentralization and variability, as we already know from the electricity market. Unlike the electricity market, which is already well equipped to handle the dynamics of renewable energy, the heating market is lagging behind. The main reason for this is that the heating sector has not yet been digitized, and systems are often operated in a very static manner, leading to increased losses in heat generation and distribution. The ENTROPIA software counteracts these trends and the associated disadvantages by proactively controlling all controllable components using Model Predictive Control (MPC).
Next generation of Digital twin
Network simulations can be used to solve detailed thermo-hydraulic problems that are not typically requested by operators. ENTROPIA is the only MPC optimization software that does not require detailed network simulation and offers comparable or higher savings through its new control strategies.
Up to 20%* efficiency increase
* Jacobsen, T., 2022. Make building heating more energy efficient by Artificial intelligence. APUEA Magazine, Issue 7, 22-25
Forecasting
Clearly structured data preparation in combination with artificial intelligence tools enables precise forecasts (e.g. heat demand) even for complex systems.
Digital twin
The physical representation of the technical system, consisting of heat generation and distribution as well as sector coupling options, can be used to calculate different operating schedules.
Optimization
A genetic algorithm calculates the optimal schedule of all controllable components (boiler, storage, pumps, …). Both fuel savings and possible revenue maximization are considered.
