Integrating off-design performance in designing CO2 power cycle systems for engine waste heat recovery

Abstract CO2 transcritical power cycle (CTPC) systems are a promising solution to improving engine efficiency through waste heat recovery. However, variable engine conditions often force the CTPC systems operate at their off-design conditions, where the performance may degrade due to the inefficiency of the components. Matching the system with the expected heat source variations is crucial for the commercialization of this technology. In this work, a novel design optimization framework is proposed, which considers the off-design performance under possible heat source variations at the design stage. Therefore, both the design and off-design models of the components and the system are integrated in detail in the framework, allowing simultaneous optimization of the components and the cycle parameters, and robust design of the system that could deal with the probable fluctuations on the heat sources. In particular, a detailed double-pipe model is used for the heat exchanger sizing and evaluation, and a radial-inflow turbine model based on the 1-D mean-line method for the expander sizing and performance prediction. The heat source variations are generated from the actual likely engine conditions and featured by the probability of occurrence. The results of a case study prove that the proposed framework can effectively integrate the system off-design performance when designing a system, and downsizing the equipment to match the probability of occurrence of the possible off-design operating conditions can lead to a medium-sized system that is much more favorable in terms of economic performance over its whole lifetime. The systems can not only be 50% more compact than that designed at the maximum power condition, reducing the initial investment cost to as low as 52%, but also maintain a duty ratio of as high as 90%. This design framework can be extended to the applications with transient heat sources and other cycle configurations.