Multi-objective optimizations of solar receiver based on deep learning strategy in different application scenarios

Efficient and secure operation of solar receivers is key to the development of concentrated solar power (CSP). Its precise and quick optimization is essential for receiver to achieve optimum thermal performance and safety across diverse application scenarios. In this study, we propose a hybrid model based on numerical simulation and deep learning to achieve this target. The hybrid model accounting for conduction, convection and radiation shows a small relative error (<2%) in over 96 % cases. We apply the model to two different application scenarios considering multiple performance objectives simultaneously and demonstrate that it can identify optimal structure parameters and heliostat scale for each inlet flow rate of the receiver. In combination with gas turbines, maximum efficiencies of 48.9 %, 53.2 %, and 56.8 % are achieved for flow rates of 0.5, 0.75, and 1 kg/s, respectively. When integrated into heating systems, the receiver achieves a maximum thermal efficiency of 89.8 % with a heliostat field scale of 5.76 MW. Furthermore, for the operation optimization, the proposed aiming strategy results in a 76 degree celsius reduction in maximum tube wall temperature compared to the conventional aiming strategy. Our approach not only provides an advanced solution for rapid cavity receiver design and operation optimization but also has the potential for extension to other structures.

Automated summary

In this study, a hybrid model based on numerical simulation and deep learning is proposed for the optimization and operation of solar receivers. By applying the model to different application scenarios and considering multiple performance objectives, small errors are achieved and optimal structure parameters and heliostat scales are identified. This approach is not only applicable to gas turbines but also heating systems.