ComRed – Complexity reduction for planning, simulation, optimisation and control of district heating systems
To achieve EU 2050 goal for greenhouse gas emission, special attention must be paid to the long-term planning and sustainability of our heating and cooling supply which is 51 % of the overall energy consumption in Europe. Intelligent coupling to the electricity and gas network and other infrastructures (wastewater, sewage treatment plants, industry) is crucial to establish efficient renewable and sustainable energy systems. District heating plays a decisive role in this context by enabling the integration of various heat sources such biomass, solar thermal energy, industrial and waste heat, large heat pumps and renewable combined heat and power generation. These integrations will provide a high degree of flexibility and efficiency to enable reliable and sustainable operation. In turn can have a significant impact on the district heating infrastructure.
As the number of technical components and potential interactions increase, future district heating systems become more complex. To deal with this complexity, the requirements for modelling and simulation in all steps have increased: Modelling and simulation of these systems is not only vital in the planning and design phase of such a system, yet also in the operational phase such as predictive monitoring of these systems as well as model-predictive control and merit-order optimization.
State-of-the-art tools and methods on simulation of district heating networks are limited to medium-scale networks. These tools have serious efficiency issues, when confronted with large-scale models of distributed systems. Limitation of these tools and methods hamper their practical use for the design and operation of future district heating networks.
The PhD project ComRed aims to overcome these limitations and develop a framework that allows to efficiently model and simulate large-scale, flexible district heating and cooling systems. There are several promising methods for a reduced model that can be effectively applied to multiple simulations of a network, with significant reduction of the computational time and resources. These are topological simplification, aggregation methods, model order reduction and data-driven models. Within ComRed, solutions to fully-dynamic simulation of large-scale district heating networks, combination and integration of these promising methods and approaches in an a) robust, b) time-efficient and c) modular and scalable way will be investigated.