Identification of the optimal operational strategy of a large district heating network through POD modeling Sara Cosentino a , Elisa Guelpa b , Roberto Melli c , Adriano Sciacovelli d , Enrico Sciubba e , Claudia Toro f and Vittorio Verda g a Politecnico di Torino Dipartimento Energia, Torino, Italy, sara.cosentino@polito.it b Politecnico di Torino Dipartimento Energia, Torino, Italy, elisa.guelpa@polito.it c University of Rome La Sapienza Dept of Mechanical & Aerospace Engineering, Rome, Italy, roberto.melli@uniroma1.it d Politecnico di Torino Dipartimento Energia, Torino, Italy, adriano.sciacovelli@polito.it e University of Rome La Sapienza Dept of Mechanical & Aerospace Engineering, Rome, Italy, enrico.sciubba@uniroma1.it f University of Rome La Sapienza Dept of Mechanical & Aerospace Engineering, Rome, Italy, claudia.toro@uniroma1.it g Politecnico di Torino Dipartimento Energia, Torino, Italy, vittorio.verda@polito.it Abstract: District heating is expected to play a major role in the supply of low-carbon heating in urban areas. This result is obtained through heat generation from CHP systems, residual heat from industries or waste-to- energy plants, and is, whenever possible, integrated by additional recourse to renewable energy sources. The constrained minimization of pumping losses is an important aspect of district heating systems design, especially in large networks where it is mandatory to achieve high efficiency and security of supply. In this paper a large district heating system, which supplies heating to a total volume of buildings of about 50 million of cubic meters, is considered. The aim of the work is to identify an optimal operational strategy for the network. An optimization entirely based on fluid-dynamic model requires large computational resources, therefore the use of a reduced model based on proper orthogonal decomposition (POD) is investigated. Various operating conditions corresponding to partial load operation are analyzed using a fluid-dynamic model of the network. The results are used to obtain the characteristic eigenvalues and eigenvectors that constitute the basis for the POD model. The latter significantly reduces the simulation time, and therefore it can be conveniently applied to the optimization of the set points of the pumping system. The results show that significant reductions in primary energy consumption can be achieved with respect to a conventional control strategy. Keywords: POD, Genetic Algorithm, District heating system, Optimization, Pumping cost 1. Introduction District Heating (DH) is an important option to provide domestic heat and domestic hot water to buildings and it plays an important role in densely populated areas [1]. This technology consists of the heat generation in efficient plants, which is then supplied to the connected users through a distribution network of insulated pipes where hot water, superheated water or steam flows. District heating networks (DHNs) are generally fed by cogeneration power plants, heat recovered from industrial processes, biomass boilers and other renewable energy systems. DHNs allow one to reduce the use of domestic boilers, with big advantages in terms of primary energy consumption, which is a major task to reach the EU 202020 objectives. The Turin district heating network, which is considered in this paper as the case study, is primarily based on cogeneration from combined cycles. The primary energy consumption per unit of heat supplied with district heating network is about 0.52 MJ/MJ, while the use of condensing boilers would involve a specific primary energy consumption of about 0.95 MJ/MJ [2]. The option of using geothermal heat