Performance assessment in district cooling networks using distributed cold storages : A case study

Abstract: District cooling is a technology that has been gaining traction lately due to increased demand from the commercial and industrial sectors, especially in dense urban areas such as Stockholm. A literature study found that customers such as hospitals, offices, malls and data centres all depend on both comfort cooling and process cooling. District cooling networks operate by producing centralized cooling energy and distributing it to consumers through underground pipes. The produced cold therein is transferred via the district cooling network as chilled water, which is pumped through the heat exchangers located in the consumer facilities which enables maintaining the desired temperatures of the consumers’ intended facilities, by removing the additional heat. The literature review showed that cold storages, which are thermal energy storages, are used to peak shave and to help reduce the output of expensive chillers and heat pumps during peak demand hours. The aim of this project is to evaluate the possibilities of using distributed cold storages in district cooling network as a means to reduce effects of distribution limitations the bottlenecks and increase distribution capacity. This project defines distribution limitations as areas with specifically low differential pressures. Additionally, the objective is to compare the costs between the scenarios. In this project, Norrenergi AB’s district cooling network is used as a case study. Norrenergi AB is an energy company located in Solna that supplies district heating and cooling to customers, mainly in the Solna and Sundbyberg region. The company delivers roughly 1 000 GWh of district heating and 70 GWh of district cooling annually. Three scenarios with various configurations of storage size and location are developed and calculated in the network simulation software NetSim, which is a software that allows complex, dynamic simulations of energy networks. According to Norrenergi AB, the criteria for acceptable network operation is that the differential pressure is required to stay between 100-800 kPa. In Scenario 1 & 2, a 15 MW cold storage is implemented in Sundbyberg and Frösunda, respectively. In Scenario 3, two smaller storages with a capacity of 3 MW each are installed in both Sundbyberg and Frösunda. For all scenarios, the energy need to fully charge the storages is calculated along with the charging/discharging profiles of the storages, which are later used as input in NetSim. In all scenarios, the storages charge during the night-time and discharge during peak hours. The main results that can be concluded from this thesis is that all scenarios led to cost savings in terms of daily production cost. The daily cost savings for each of the scenarios were 2.7%, 4.8% and 4.3%, respectively. In addition to this, the effects of distribution limitations in the network are analysed with regards to the differential pressures. The results indicate that although Scenario 3 displayed only the second lowest production cost, it greatly reduced the effects of distribution limitations in key areas compared to that of Scenario 2 which showed abnormally low differential pressures during peak hours, leading to cooling not being delivered. With these aspects in mind, the deduction is that a combination of the capacity size similar to those of scenarios 1 & 2, combined with the capacity distribution in Scenario 3 should be the optimal setup in the future. Furthermore, cold storages can help reduce the use of chillers and thus help reduce the use of harmful refrigerants in the system. Future iterations of this model should consider the possibilities of including new consumers and optimized charging/discharging profiles of the storages. Variations of the temperature difference should be included as well since an increase/reduction of the temperature difference can directly affect storage capacities.