Supercritical carbon dioxide power cycles for waste heat recovery applications

Abstract

The growing energy demand and the increasingly stringent regulations on pollutant and greenhouse gas emissions are driving academia and industry to seek new approaches to increase the overall energy efficiency of existing industrial facilities. Among them, the recovery and utilisation of industrial waste heat is currently considered as one of the most effective approaches to reduce the energy demand of industrial processes as they are characterised by thermal energy losses, through high temperature exhausts above 300°C, that account for nearly 11.4% of primary energy consumption. For these high temperature waste heat sources, the use of conventional heat to power conversion systems based on bottoming thermodynamic cycles is limited by technological and economic constraints. Most of the state-of-the-art working fluids are indeed not able to perform safely and efficiently at high temperatures. Supercritical Carbon Dioxide (sCO2) power systems allow to overcome these limitations because of the chemical stability of CO2 at high temperatures. Furthermore, the favourable CO2 thermo-physical properties in the supercritical state, including high density, allow to achieve superior performance and lower footprint and cost compared to Organic Rankine Cycles and other more conventional technologies. With the aim of giving a broad overview of the potential of sCO2 power cycles in high temperature waste heat recovery (WHR) applications, this research firstly investigates the theoretical capabilities of several Joule-Brayton cycle configurations. The analysis involves performance indicators and economic metrics, which are calculated using cost correlations and budgetary quotations to estimate the investment costs of equipment. This aspect represents one of the main novel contributions of the research. Among the investigated layouts, the simple regenerative cycle showed the highest competitiveness for industrial uptake of the sCO2 technology at small-scales (<0.5 MWe) in high-grade waste heat to power applications. For this reason, such cycle layout has been adopted as reference for the design and construction of a 50 kWe state-of-the-art experimental facility. The facility comprises an 830 kW process air heater able of providing an exhaust mass flow rate of 1.0 kg/s at 70 mbarg and maximum temperature of 800°C, and a water dry cooler of 500 kW heat rejection capacity as heat sink. The sCO2 heat to power conversion block utilises a single-shaft Compressor-Generator-Turbine unit and three types of heat exchanger technology. The main design features of the test facility as well as operation and safety considerations are discussed. This research activity allowed to retrieve accurate geometrical and performance data by component manufacturers which have been used to develop a detailed numerical model of the facility with the objective of investigating the steady-state and transient performance of the sCO2 system. Operating maps of the unit have been obtained which can form the baseline for the setting up of optimisation and control strategies. The dynamic analysis showed that the system is able to quickly adapt to transient heat load profiles, proving the flexible nature of the sCO2 unit investigated. Start-up and shut-down strategies able to achieve a safer build-up and decline of pressures and temperatures in the circuit, thus eliminating the risk of flow shocks and excessive mechanical stresses, have also been identified. A further novel contribution is assessment of the advantages of having the turbine and compressor driven independently as opposed to being mounted on the same shaft that dictates operation at the same speed. The results show only a small benefit at design conditions, but a power increase of 27% at 10% increase in heat source temperature, highlighting the advantage of independent drive at off design conditions. The adoption of an inventory control strategy to regulate the sCO2 system during transient operations showed that the imposed variation in the CO2 mass circulating in the loop allows to achieve a 30% variation in the turbine inlet temperature with lower penalties on system performance compared to turbomachinery speed control.