基于贝叶斯优化的选择性热光伏辐射器设计

Designing Selective Thermophotovoltaic Emitter through Bayesian Optimization

通过控制热光伏系统辐射器的发射光谱,可以有效降低光伏接收器的热损失能量。本文证明了在机器学习的指导下可以实现高效的非周期性选择性热光伏辐射器的设计。对于锑化镓光伏电池,在超过5.23×10^(9)个每层包含不同材料和不同层厚的多层膜候选结构中,仅计算了不超过0.67%的候选结构就可以得到最大特征因子为82.16%对应的结构参数。测量利用优化的结构参数加工的样品发射光谱得到的特征因子为81.35%,并对优化的选择性辐射器进行了系统理论效率分析和热稳定性测试。结果表明采用贝叶斯优化算法可以高效设计选择性热光伏辐射器,进一步促进了基于机器学习的超材料设计方法在其他领域的应用。

Controlling the thermal emission characteristics of a thermophotovoltaic(TPV)emitter can effectively reduce the thermal losses in a photovoltaic cell.In this work,we demonstrate that a highly selective,aperiodic TPV emitter with high figure of merit(FOM)can be achieved with the guidance of machine learning.As for the gallium antimonide photovoltaic cell,the design of selective TPV emitter is optimized over 5.23×10^(9) candidate structures in multilayers consisting of multiple components to maximize the FOM.The maximum FOM could be realized within calculations less than 0.67%candidate structures,the resulting optimal structure is an aperiodic multilayer structure with a FOM of 82.16%.The optimal structure is then fabricated,and experimental result of the emission characteristics is achieved with a FOM of 81.35%.What’s more,we have analyzed the efficiency of the TPV system and measured the thermal stability of the fabricated samples.The results demonstrate the Bayesian optimization is efficient to design selective TPV emitters,and the machine-learning-based design of metamaterials can be extended for the expensive black-box global optimization problems in other field applications.