5th generation district heating and cooling network planning: A Dantzig–Wolfe decomposition approach

The adequate planning of district energy systems is a key process to provide communities with proper heating and cooling networks. To find cost-effective planning solutions, this decision-making process is usually formulated as mathematical model, which is optimized to determine where and how many assets should be installed in the network. However, this optimization problem is becoming more complex as district energy systems evolve and become more elaborate. For the 5th generation district heating and cooling (5GDHC) networks, this planning framework comprises multiple building energy systems, a thermal and electrical network as well as central heating and cooling units. As a result, the optimization problem associated with these circumstances can easily become intractable as the number of elements of the network increases. To alleviate this tractability problem, in this paper, a Dantzig–Wolfe approach is devised to decompose a mixed-integer linear program into multiple subproblems (for every building) and a master problem (thermal and electrical network and central units). A realistic case study based on a 5GDHC system in Germany is considered. For this case study, it is demonstrated that the proposed decomposition approach yields the same results attained by the original not decomposed problem while achieving significant gains in terms of scalability and computational times. More specifically, the decomposition approach reduces the computational time for districts with more than 10 buildings by in average 94%. This corroborates the potential of the proposed approach to improve the computational efficiency of models that will deliver cost-effective investment plans for 5GDHC networks.