Text

  • Study location Mälardalen University in Västerås
Date
  • 2020-09-29 09:15–11:15

The public defense of Nathan Zimmerman's doctoral thesis in Energy and Environmental Engineering

The public defence of Nathan Zimmerman's doctoral thesis ”Modelling Towards Control Applications Within the Heat and Power Industry” will take place at Mälardalen University, Västerås, on September 29, 2020, at 9.15 AM.

Title: Modelling Towards Control Applications Within the Heat and Power Industry

Serial number: 319

The examining committee consists of Professor Magnus Genrup, Lund University, Professor Johan Ölvander, Linköping University and Professor Andrew Nix, West Virginia University. Professor Natasa Nord, Norwegian University of Science and Technology, has been appointed the faculty examiner.

Reserve is Mikael Lundh, Senior Principal Scientist, ABB.


Abstract

For the development of future energy systems it is not only pertinent to analyse the production of energy, but to also consider when and how this energy will be consumed. In this work, primary energy production is investigated by analysing the transient behaviour of refuse derived fuels fired in circulating fluidized bed boilers. The inherent nature of refuse derived fuels contributes to a fluctuation in the fuel’s composition and moisture content, and in effect, its heating value. This fluctuation can lead to swings in the boiler’s combustion temperatures, and if not mitigated, can contribute to unwanted emissions and even corrosion or fouling. In regards to the consumption side, this work focuses on the dynamics of how temperatures within district heating networks propagate over time. By gaining an understanding of how heat flows through a network, the investigation of how to reduce peak load production, identify network bottlenecks, and the implications of non-conventional thermal heating networks are investigated.

The integration of a holistic perspective on primary energy production and its consumption provides the ability to shift towards more efficient energy systems. This can be achieved by understanding the transient factors associated with combustion of refuse derived fuels and district heating through the development of dynamic physics based models. In the case of production, feed-forward model predictive control can be implemented by predicting the transient causing factors such as a fuel’s moisture content and heating value. In the case of consumption, feed-forward model predictive control can be enabled by predicting the heating requirements of the end-users in a district heating network. From a feed-forward control perspective, information on the transient causing factors can be introduced to the controller in advance before it has a chance to disrupt the process. The controller is then able to make control predictions forward in time, choose an optimal action, and along with feedback, assist in improving system performance.

The development and utilization of modelling libraries based on first principle modelling techniques afford a higher level of durability in energy system modelling. The approach described in this thesis unlocks a degree of flexibility which eliminates the need for unambiguous modelling and simulations by allowing model components to be reused. The ex-portability of these models further distinguishes them as they can be utilized to test new control approaches within an energy system as real-time predictions within each subsystem of the energy system become more accessible.