The integration of machine learning (ML) with perovskite solar cells (PSCs) signifies a groundbreaking era in photovoltaic (PV) technology. The traditional iterative approaches in PSC research are often time‐consuming and resource‐intensive. In contrast, ML leverages available data and sophisticated algorithms to quickly identify properties and optimize parameters for novel materials and devices. This review explores how ML‐driven approaches are improving various facets of PSCs research, including the rapid screening of novel compositions, enhancing stability, refining device architectures, and deepening the understanding of underlying physics. The paper is structured to gradually familiarize readers with essential terminologies and concepts, ensuring a solid foundation before delving into more intricate topics. A concise workflow and various introductory toolkits for ML are also briefly discussed. Through a detailed analysis of compelling case studies, a basic research framework within ML‐PSC‐integrated research is provided. This comprehensive review can serve as a valuable reference for researchers aiming to understand and leverage ML‐driven approaches in PSCs research, advancing the path for more efficient and sustainable PV technologies.

中文翻译：

---

推进钙钛矿太阳能电池研究的机器学习方法


机器学习 （ML） 与钙钛矿太阳能电池 （PSC） 的集成标志着光伏 （PV） 技术的开创性时代。PSC 研究中的传统迭代方法通常耗时且耗费资源。相比之下，ML 利用可用数据和复杂算法来快速识别新材料和设备的特性并优化参数。本综述探讨了 ML 驱动的方法如何改进 PSC 研究的各个方面，包括快速筛选新成分、增强稳定性、改进器件架构和加深对基础物理学的理解。本文的结构旨在让读者逐渐熟悉基本术语和概念，确保在深入研究更复杂的主题之前打下坚实的基础。此外，还简要讨论了 ML 的简明工作流程和各种入门工具包。通过对令人信服的案例研究的详细分析，提供了 ML-PSC 综合研究中的基础研究框架。本综述可为旨在理解和利用 ML 驱动方法在 PSC 研究中的研究提供有价值的参考，从而推动更高效和可持续的 PV 技术的发展。