Abstract
The growing energy demand in the world, together with the increasing challenges related to climate change, has sparked an ongoing restructuring of energy systems towards renewable energy. The variability of non-dispatchable renewable energy sources and the increasing power demand caused by increasing electrification have triggered the development of new smart energy systems. Currently, electricity covers a large part of the heating demand in Norway. Low-temperature District Heating Grids (DHGs) can contribute substantially to more efficient use of energy resources as well as better integration of renewable energy and surplus heat to cover heating demand. This thesis studies an optimization problem regarding the cost-effectiveness of utilizing waste heat, Demand Side Management (DSM), and Thermal Energy Storage (TES) in low-temperature DHGs. To solve the optimization problem, we first present a deterministic model, before expanding it to two stochastic models with uncertain Space Heating demand. Another significant contribution of this work is the comparison of the deterministic and stochastic models and assessing the value of including uncertainty in Space Heating demand. This thesis presents both a traditional scenario tree-based model and a multi-horizon structure model. The size of the traditional stochastic model increases exponentially with the number of periods with uncertainty. By decoupling the periods with uncertainty, the multi-horizon approach reduces the problem size extensively and overcomes the computational challenges faced by the traditional stochastic model. Calculation times are reduced from about 9 hours for each problem instance to 25 seconds. In most cases, the multi-horizon model provides a satisfactory solution close to the one provided by the traditional stochastic model. The methodology is evaluated in a planned residential area at Leangen, in Trondheim. Considering seasonal TES and DSM, the analysis in this thesis shows that a TES has the most significant impact on the annual operational cost as it allows the largest reduction of heat production from expensive heat technologies in winter months. If there is a large surplus of heat from waste incineration in the summer, the larger the TES capacity, the more production from the most expensive production technologies can be reduced, which results in a significant reduction in total operational cost. DSM is valuable both with and without TES, but with a moderately lower impact on operational cost. The availability of TES and application of DSM provides savings of up to 22% in CO2 emissions, 11% in peak production, and 9% lower operational cost. Even if the current development of the DHG at Leangen does not consider a TES, our results indicate that the payback time for storage may be as low as 9 to 11 years and suggest that further research should be carried out of including TES in the DHG.