The urgent need to transition to zero-carbon energy provision, enshrined in international agreements on climate change mitigation and adaptation, e.g., the United Nations Paris Agreement, requires a fundamental transformation of the current interconnected power system towards a system design that is distributed, multimodal, hybrid and smart (DMHS). A concrete challenge implied by this transformation lies in coordinating a diverse set of actors across many
scales to ensure the system is stable and resilient. By building on our existing work in the priority programme, we propose to develop a validated set of models for most fundamental grid components along with methodological frameworks both for their control as well as for
aggregating their behaviour to enable the provision of ancillary services. The resulting set of methods will be applicable to AC and DC active distribution networks (ADNs), high voltage DC (HVDC) lines, energy storage systems (ESSs) connected to the transmission system, and many more. The focus of the modelling will be on the grid-facing behaviour and capabilities of the power grid components to contribute to grid stability and to controlling the overall system. This approach will not rely on overly detailed high fidelity models. These components will often be grouped within larger organisational units, typically along grid layers, for example active distribution grids, but also within and across layers, as in the case of virtual power plants.
We will design how such units, especially active distribution grids, can be kept internally stable using a sophisticated Quality-of-Servicebased control architecture. In order to make sure that the organisational units are also contributing to the overall system stability, we will show how to tune them to provide ancillary services
for the overall system. This will build on probabilistic stability methods.
In order to ensure the effective use of energy and ancillary services between such organisational units, we will also design and study system-wide learning-based control approaches to coordinate the different organisational units and the ancillary services they can provide. By explicitly considering organisational units, we provide a consistent decomposition of the overall system. This will allow us to
experimentally validate our models and controllers in laboratory experiments using a power hardware in the loop approach. We can then further study the overall integrated system performance and its resilience using probabilistic methods and detailed case studies. Taken together, the set of models and methods will provide a consistent design for the coordination of distributed actors in a multilayered integrated grid system.
