1 Department of Electrical Engineering, Technical University of Denmark2 Automation and Control, Department of Electrical Engineering, Technical University of Denmark3 Center for Electric Power and Energy, Department of Electrical Engineering, Technical University of Denmark4 Centre for oil and gas – DTU, Center, Technical University of Denmark
Uncontrollable power generation, distributed energy resources, controllable demand, etc. are fundamental aspects of energy systems largely based on renewable energy supply. These technologies have in common that they contradict the conventional categories of electric power system operation. As their introduction has proceeded incrementally in the past, operation strategies of the power system could be adapted. For example much more wind power could be integrated than originally anticipated, largely due to the flexibility reserves already present in the power system, and the possibility of interregional electricity exchange. However, at the same time, it seems that the overall system design cannot keep up by simply adapting in response to changes, but that also new strategies have to be designed in anticipation. Changes to the electricity markets have been suggested to adapt to the limited predictability of wind power, and several new control strategies have been proposed, in particular to enable the control of distributed energy resources, including for example, distributed generation or electric vehicles. Market designs addressing the procurement of balancing resources are highly dependent on the operation strategies specifying the resource requirements. How should one decide which control strategy and market configuration is best for a future power system? Most research up to this point has addressed single isolated aspects of this design problem. Those of the ideas that fit with current markets and operation concepts are lucky; they can be evaluated on the present design. But how could they be evaluated on a potential future power system? Approaches are required that support the design and evaluation of power system operation and control in context of future energy scenarios. This work addresses this challenge, not by providing a universal solution, but by providing basic modeling methodology that enables better problem formulation and by suggesting an approach to addressing the general chicken/egg problem of planning and re-design of system operation and control. The dissertation first focuses on the development of models, diagrams, that support the conceptual design of control and operation strategies, where a central theme is the focus on modeling system goals and functions rather than system structure. The perspective is then shifted toward long-term energy scenarios and adaptation of power system operation, considering the integration of energy scenario models with the re-design of operation strategies. The main contributions in the first part are, firstly, by adaptation of an existing functional modeling approach called Multilevel Flow Modeling (MFM) to the power systems domain, identifying the means-ends composition of control levels and development of principles for the consistent modeling of control structures, a formalization of control-as-a-service; secondly, the formal mapping of fluctuating and controllable resources to a multi-scale and multi-stage representation of control and operation structures; and finally the application to some concrete study cases, including a present system balancing, and proposed control structures such as Microgrids and Cells. In the second part, the main contributions are the outline of a formation strategy, integrating the design and model-based evaluation of future power system operation concepts with iterative energy scenario development. Finally, a new modeling framework for development and evaluation of power system operation in context of energy-storage based power system balancing is introduced.