Advanced Photonics, Volume. 6, Issue 5, 054001(2024)
Machine learning for perovskite optoelectronics: a review
Fig. 1. Flowchart of ML-assisted preparation of perovskite optoelectronic devices. This includes the fabrication of devices, data acquisition, feature engineering, selection of appropriate models, and performance evaluation. The models encompass a range of learning algorithms: supervised, unsupervised, and semi-supervised.
Fig. 2. (a) Schematic illustration of design rules for capping layer of PSCs. (b) The cross-validated root mean square error of various machine-learning models. (c) The feature importance ranking obtained from the RF regression algorithm and SHAP library, showing the chemical properties and processing conditions in descending order of importance (rank). The yellow and purple color indicates high and low values of a given feature, respectively. Reproduced with permission from Ref. 78 (CC-BY).
Fig. 3. Working principle of ML: relevance analysis—experimental verification—one with the best properties. Complex multivariable analysis by ML. PVK is perovskite. Reproduced with permission from Ref. 83 © 2022 Wiley‐VCH.
Fig. 4. Flow chart of the two-step ML in this work, which includes two steps, first-step ML (left side) and second-step ML (right side). Reproduced with permission from Ref. 80 © 2021 Royal Society of Chemistry.
Fig. 5. Schematics of Sn-PSC design ML recommendation. (a) Parameter setting of each feature for the recommendation process. (b) Schematic of recommended Sn PSCs. (c) Labeled number and corresponding materials used as other layers in (a). Reproduced with permission from Ref. 84 (CC-BY).
Fig. 6. ML flowchart for studying the variables that influence the performance of vapor-deposited PSCs, consisting of four main parts: building the dataset, feature engineering, SHAP interpretation model, and model prediction. Reproduced with permission from Ref. 85 © 2023 Royal Society of Chemistry.
Fig. 7. Machine vision workflow for quantifying large-area perovskite film morphology from optical images using the PerovskiteVision tool. Reproduced with permission from Ref. 86 (CC-BY).
Fig. 8. Traditional method versus ML-assisted method. The traditional method is to verify the effectiveness of the additive for PeLEDs, while the ML-assisted method is to fabricate the device with the “good” additive predicted by ML. The additive is added into the perovskite precursor solution to form the perovskite film. Reproduced with permission from Ref. 39 © 2022 Wiley-VCH.
Fig. 9. Illustration of lifetime prediction by 5EML. Reproduced with permission from Ref. 40 © 2024 Wiley-VCH.
Get Citation
Copy Citation Text
Feiyue Lu, Yanyan Liang, Nana Wang, Lin Zhu, Jianpu Wang, "Machine learning for perovskite optoelectronics: a review," Adv. Photon. 6, 054001 (2024)
Category: Reviews
Received: May. 23, 2024
Accepted: Aug. 1, 2024
Published Online: Aug. 29, 2024
The Author Email: Lin Zhu (iamlzhu@njtech.edu.cn), Jianpu Wang (iamjpwang@njtech.edu.cn)