Energy Transport Networks and Model Reduction

Energy Transport Networks

Admitting renewable production of electric power and gas has a massive impact on the operation of transport networks in electricity, gas and district hea-ting. A strongly fluctuating feed-in into medium and lower network levels requires increasingly communicative, model-based concepts for optimal control under uncertainties.

District heating networks, for instance, serve mainly to supply heat and hot water. However, a significant portion of the revenues is achieved by selling electricity from cogeneration of heat and power.

Taking into account the spacial and transient behavior of the district hea-ting network can help to use the network as large energy storage reacting the volatility on the electricity market and thus increase the profits of the company. However, optimizing the operating sequences of the heating plant requires that all its components be mapped as accurately as possible.

For the district heating network, this means that one must consider not only static states, but also its entire temporal dynamics. This is the only way to opti-mally operate the network as a component of the heating plant.

Irrespective of the type of energy (electricity, natural gas, district heating), there are recurrent mathematical problems in modeling, simulation, optimization and/or control.

Our competence in the field of energy networks covers the entire spectrum:

  • Mathematical modeling
  • Algorithm development
  • Optimization and control
  • Software implementation 

Due to the size and complexity of the mathematical models, methods of model reduction have to be developed and applied in order to speed up the simulati-on and to enable a model-based control at all. This model reduction must be designed in such a way that the essential processes are still accurately re-produced.

Model Reduction

Due to highly developed software and computer power technical products and processes may be simulated in very detail nowadays. However, developers typically want more: variants have to be checked, quickly evaluated, and optimized or the interaction of structure, fluid, and electronic components have to be investigated. The key technology in this environment is parametric model reduction: the original objects, e.g. large finite element models of a car body or non-linear differential-algebraic equations describing an electrical power supply network, are converted to parametric reduced state space models. These can be evaluated much quicker as the original models and, due to the uniform structure, be coupled, even if the original objects were modelled by different simulators.

The new parametric approach developed at the Fraunhofer ITWM solves two classical problems of model reduction:

  • It is no longer necessary to start a new reduction, if design parameters change. Rather, a few pro-calculated reduced models are interpolated, which usually takes only fractions of a second.
  • Non-linearities can finally be treated by using operating points as parameters.

The concept for dealing with new applications consists of implementing new interfaces to our universal model reduction toolbox (MRT). Up to now, there are available interfaces to ANSYS (multi-physics FE) and PSAT (electric power transport). In the preceding years projects from very different domains have been realized: power and gas networks, active noise and vibration control in vehicles, real-time simulation of low frequency sound fields in indoor acoustics, optimization of a subwoofer, and confidential industrial projects in the area of thermal deformation.

Example Projects for Energy Transport Networks

 

Project »ENERDIG«

Digitalization and artificial intelligence should reduce energy consumption in production and buildings, adapt it to fluctuating generation and control energy conversion processes more efficiently.

 

Ammonia as a »Green Hydrogen« Storage

As part of the project »AMMONPAKTOR«, we are researching the possibilities of using ammonia as a »green hydrogen storage« together with the Fraunhofer IMM.

 

Digital Twin Monitors and Controls District Heating Networks

In the project »DingFESt« we are developing a digital twin for controlling decentralized district heating networks.

 

Better Understanding Hydrogen Electrolysis Through Simulation

In the project »H2-D« we focus on the shape design of bipolar plates.

 

Efficiency of District Heating Grids

The aim of the research project »DYNEFF« is to sustainably improve the efficiency of the operating processes of district heating cogeneration plants by means of simulations.

 

MathEnergy

Control Concepts for Power Grids: Multi-network model-based monitoring and control for planning and operation.

 

Gas Pipeline Networks

The research work described here represents a contribution to the EU project »e_GASGRID«.

Example Projects for Model Reduction

 

Root Cause Analysis of Measurement Issues

In the project »RoMI« we have developed an algorithm that analyzes potential root causes of aberrations in components or joints with the help of measurement data.

 

Equalization of Bass Speakers

The project presented here has been carried out in close cooperation with the company KS-Beschallungstechnik, one of the most renowned addresses for expert sound systems.

 

Simulation of Mixing Processes for Granular Material

In the project a software for regulation and simulation of mix beds was simulated.

 

Radiation Transport in Scattering Media

The production, processing, and also the use of glass and ceramic materials as fireproof materials are considerably influenced by processes of radiative transfer.

 

»PAMOR«

Parametric model order reduction for computational design optimization. Goal of the project was the further development of a software (Model Reduction Toolbox).

3D-Reconstruction of Fractured Hip

We support the development of a computer assisted surgery system for minimally invasive treatment of hip fractures by Stryker Navigation.

Local Application of Drugs to the Inner Ear

The goal of the project was the development of a model for the computer simulation of agent distribution in the liquids of the inner ear.