Monitoring and Optimizing District Heating Networks With Digital Twin
The district heating of the future as a central component of a sustainable energy system requires new control concepts as well as communicative approaches. With our software tool »AD-Net District Heating«, we at the Fraunhofer Institute for Industrial Mathematics ITWM dynamically simulate and optimize the operation of heating networks in real time.
Our digital twin enables heat suppliers to ensure a consistently efficient network operation – even under increasingly complex operating conditions with decentralized feed-in – while ensuring its stability and security of supply.
»AD-Net District Heating« Simulates and Optimizes Operation of Heating Networks
Current software for operational support of combined heat and power plants either deals with the optimal use of local operating resources, treating the heating network only as a structureless sink. Or software is used for spatially resolved thermo-hydraulic simulation of the heating networks, which, however, usually neither captures dynamic runtime effects nor allows for operational optimization. This is where »AD-Net District Heating« comes in.
The dynamic simulation of the district heating network has enormous advantages: The network operator can use the software to monitor both the temperature and the flow conditions at any point in the district heating network at any time. This allows to control the feed temperature provided at the power plant and the mass flow pumped into the network and thus prevents, for example, gas boilers from having to be switched on at high cost to produce heat.
Mathematics Behind the Software Creates Digital Twin
Classical methods for solving the fluctuating thermo-hydraulic equations are too inaccurate or time-consuming to provide optimized controls for live operation of the district heating network.
For this reason, we have developed our own efficient numerical method for solving the equations on the pipeline network for »AD-Net District Heating«, which can handle optimization horizons of several days. Based on the software, we create a digital twin of real existing district heating networks. This couples efficient dynamic flow simulation with robust consumption forecasting using artificial neural networks, model predictive control and modern measurement sensor technology. Due to the model-based approach, the digital twin is able to propose optimal controls for live operation even with low sensor coverage, significantly exceeding the wealth of experience in previous control routines and consumption situations. Given the decentralization of feed-in locations and the on-going expansion of the grid, this capability is more important than ever.
Details about the software and the projects in which it is used can be found on our website at: www.itwm.fraunhofer.de/ad-net-fernwaerme
More Research on District Heating: Optimize Heating Plant Processes
Researchers of our »System Analysis, Prognosis and Control« department develop and implement methods to intelligently monitor and automate plants. Our innovative, AI-supported technologies for plant operation and drive technology help to plan ahead and optimally control energy requirements. To this end, our researchers are expanding plants in order to monitor and automate them smart. In this way, we save energy and costs. More information you can find here.