Multi-Objective Optimisation of Expansion Trains in CAES: Incorporating Organic Rankine Cycles for Improved Efficiency

Abstract This paper is focused on the expansion train designed for the Compressed Air Energy Storage concept under development in the ASTERIx-CAESar project funded by the European Union. This project aims to boost the efficiency of existing diabatic CAES concepts by integrating concentrated solar thermal energy collected by an innovative high-temperature, volumetric central-receiver system. This concept renders a combined energy storage system exploiting the synergies between thermal energy storage and compressed air energy storage, in turn achieving significantly higher round-trip efficiency. In this hybrid system, electricity from the grid is used to power a series of compressors when prices are low, in particular when excess renewable energy not demanded by the consumers is curtailed. This electricity is thereby stored in the form of low-temperature compressed air in a Compressed Air Storage, whilst the thermal energy harvested from the compression process (i.e., intercoolers of the compression train) is stored simultaneously in a Low-Temperature Thermal Energy Storage unit. Independently from this storage system, a field of heliostats collects solar energy during sun hours, and this is converted into very high temperature (800°C) heat by means of an innovative volumetric air receiver atop the tower. This thermal energy at very high temperature is stored in a separate High-Temperature Thermal Energy Storage. When the demand for electricity in the grid rises, air is discharged from the compressed air storage unit and heated up with the energy stored in the low and high temperature energy storage systems. This heating and expansion process takes place across a tailored expansion train design using preheaters and reheaters to maximise the utilisation of the energies stored. The exhaust gases coming out from the low pressure turbine still carry a significant amount of energy, what makes a negative impact on round-trip efficiency. Therefore, the ASTERIx-CAESar project considers the incorporation of a bottoming waste heat recovery unit based on Organic Rankine Cycle power systems. Multiple air-cooled Rankine cycle configurations are developed for a wide range of Expander Exit Temperature (from approximately 300°C to 600°C), working fluids (organic, water/steam), and cycle configurations (subcritical, transcritical and supercritical). The impact of scale effects on expander design (screw expander or axial turbine) and, accordingly, isentropic efficiency is also considered, given that this falls within the scale of the project, whose industrialisation looks into potential applications in the range from 1 MWe to 100 MWe. Simulations of these cycles used Python code, benchmarked against literature data and Thermoflex simulations, which provide detailed specs of heat recovery steam generator components and steady-state performance. A multi-objective optimisation of the bottoming cycle is conducted, considering both technical and economic aspects, with the aim to maximise efficiency and heat recovery by adjusting the pressure/temperature levels in the vapour generator. A global optimisation of the expansion train is conducted to maximise CAES system efficiency. The approach identifies the optimal cycle configuration for each exhaust temperature and scale, and integrates the bottoming heat recovery system with a two-stage turbine. Recommendations to enhance CAES system efficiency are provided.