以卤化物双钙钛矿材料为研究对象,利用机器学习方法高速、高精确预测卤化物双钙钛矿材料的带隙和相对稳定性。使用贝叶斯岭回归、梯度提升回归、支持向量回归和XGBoost这4种算法建立模型,分析得出：梯度提升回归可为相对稳定性提供最高性能预测（R2=0.9161,MAE=0.2061）,XGBoost可为带隙提供最高性能预测（R2=0.9899,MAE=0.0542）;采用SHAP方法解释模型后,对元素替换后的新样本进行筛选,最终获得18种光吸收范围理想且稳定性良好的卤化物双钙钛矿。结果表明,相比传统方法,基于数据驱动的机器学习方法可有效加速功能材料的发现,提高设计效率。
Abstract
Taking halide double perovskite materials as the research object, the machine learning method is used to predict the band gap and relative stability of halide double perovskite materials with high speed and high accuracy. Four distinct algorithms, namely Bayesian ridge regression, gradient boosting regression, support vector regression, and XGBoost, are employed to construct predictive models. The results show that gradient boosting regression can provide the highest performance prediction for relative stability （R2=0.9161, MAE=0.2061）, XGBoost can provide the highest performance prediction for band gap （R2=0.9899, MAE=0.0542）, and after using the SHAP method to explain the model, the new samples after element substitution are screened, and finally 18 halide double perovskites with ideal light absorption range and exceptional stability are obtained. These outcomes indicate that compared with traditional methods, data-driven machine learning can effectively accelerate functiona material discovery and improve design efficiency .
关键词
卤化物双钙钛矿 /
机器学习 /
特征工程 /
材料发现 /
太阳电池
Key words
halide double perovskite /
machine learning /
feature engineering /
material discovery /
solar cells
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] WANG J Y, XU P C, JI X B, et al.Feature selection in machine learning for perovskite materials design and discovery[J]. Materials, 2023, 16(8): 3134.
[2] LIU Z Y, YANG H J, WANG J Y, et al.Synthesis of lead-free Cs2AgBiX6 (X=Cl, Br, I) double perovskite nanoplatelets and their application in CO2 photocatalytic reduction[J]. Nano letters, 2021, 21(4): 1620-1627.
[3] FORTKAMP F P, NAKASHIMA A T D, DOS SANTOS V M A, et al. Computationally-efficient optimization of the remanence angles of permanent magnet circuits for magnetic refrigeration[J]. Journal of magnetism and magnetic materials, 2023, 569: 170429.
[4] IVANOV N, ZHURAVLEV S, ZANEGIN S, et al.Calculation, design, and winding preliminary tests of 90 kW HTS machine for small-scale demonstrator of generating system for future aircraft with hybrid propulsion system[J]. IEEE transactions on applied superconductivity, 2023, 33(2): 5200105.
[5] WEI J, CHU X, SUN X Y, et al.Machine learning in materials science[J]. InfoMat, 2019, 1(3): 338-358.
[6] LU S H, ZHOU Q H, GUO Y L, et al.On-the-fly interpretable machine learning for rapid discovery of two-dimensional ferromagnets with high curie temperature[J]. Chem, 2022, 8(3): 769-783.
[7] WAN X Y, ZHANG Y H, LU S H, et al.Machine learning accelerated search for new double perovskite oxide photocatalysis[J]. Acta physica sinica, 2022, 71(17): 177101.
[8] NISHIMURA S, LE S D, MIYAZATO I, et al.High-throughput screening and literature data-driven machine learning-assisted investigation of multi-component La2O3-based catalysts for the oxidative coupling of methane[J]. Catalysis science & technology, 2022, 12(9): 2766-2774.
[9] MORELOCK R J, BARE Z J L, MUSGRAVE C B. Bond-valence parameterization for the accurate description of DFT energetics[J]. Journal of chemical theory and computation, 2022, 18(5): 3257-3267.
[10] GUO Z M, LIN B.Machine learning stability and band gap of lead-free halide double perovskite materials for perovskite solar cells[J]. Solar energy, 2021, 228: 689-699.
[11] KIM Y, KIM E, ANTONO E, et al.Machine-learned metrics for predicting the likelihood of success in materials discovery[J]. NPJ computational materials, 2020, 6: 131.
[12] 王妮妮, 马萍, 张宏立, 等. 基于多尺度深度卷积网络特征融合的滚动轴承故障诊断[J]. 太阳能学报, 2022, 43(4): 351-358.
WANG N N, MA P, ZHANG H L,et al.Fault diagnosis of rolling bearing based on feature fusion of multi- scale deep convolutional network[J]. Acta energiae solaris sinica, 2022, 43(4): 351-358.
[13] KONG Q Z, HE C, LIAO L J, et al.Hyperparameter optimization for interfacial bond strength prediction between fiber-reinforced polymer and concrete[J]. Structures, 2023, 51: 573-601.
[14] MARAPELLI B, CARIE A, ISLAM S M N. Software effort estimation with use case points using ensemble machine learning models[C]//2021 International Conference on Electrical, Computer and Energy Technologies(ICECET).Cape Town, South Africa, 2021: 333-338.
[15] ZHA L L, MA K, LI G Q, et al.An improved extreme learning machine with self-recurrent hidden layer[J]. Advanced engineering informatics, 2022, 54: 101736.
[16] MIENYE I D, SUN Y X.A survey of ensemble learning: concepts, algorithms, applications, and prospects[J]. IEEE ACCESS, 2022, 10: 99129-99149.
[17] ZHENG W D, CHENG H, LIU Y R, et. al. Machine learning for imbalanced datasets: application in prediction of 3d-5d double perovskite structures[J]. Computational materials science, 2022, 209: 111394.
[18] ZHI C Y, WANG S, SUN S J, et al.Machine-learning-assisted screening of interface passivation materials for perovskite solar cells[J]. ACS energy letters, 2023, 8(3): 1424-1433.
[19] HAO J G, HO T K.Machine learning made easy: a review of scikit-learn package in python programming language[J]. Journal of educational and behavioral statistics, 2019, 44(3): 348-361.
[20] SUCHOPAROVA G, NERUDA R.Genens: an AutoML system for ensemble optimization based on developmental genetic programming[C]//2020 IEEE Symposium Series on Computational Intelligence (SSCI). Canberra, ACT, Australia, 2020: 631-638.
[21] LU S H, ZHOU Q H, OUYANG Y X, et al.Accelerated discovery of stable lead-free hybrid organic-inorganic perovskites via machine learning[J]. Nature communications, 2018, 9: 3405.
[22] ZHANG L, LI N.Material analysis and big data monitoring of sports training equipment based on machine learning algorithm[J]. Neural computing and applications, 2022, 34(4): 2749-2763.
[23] PAN Z, ZHOU Y G, ZHANG L.Photoelectrochemical properties, machine learning, and symbolic regression for molecularly engineered halide perovskite materials in water[J]. ACS applied materials & interfaces, 2022, 14(7): 9933-9943.
[24] 胡扬, 张胜利, 周文瀚, 等. 基于机器学习探索钙钛矿材料及其应用[J]. 硅酸盐学报, 2023, 51(2): 452-468.
HU Y, ZHANG S L, ZHOU W H, et al.Studies on perovskite material and its applications via machine learning[J]. Journal of the Chinese Ceramic Society, 2023, 51(2): 452-468.
[25] CHOUBISA H, TODOROVIĆ P, PINA J M, et al.Interpretable discovery of semiconductors with machine learning[J]. NPJ computational materials, 2023, 9: 117.
[26] HUANG Y R, LI S Y, ZHANG L.Accelerated multisolvent prediction for aqueous stable halide perovskite materials[J]. ACS applied materials & interfaces, 2023, 15(41): 48771-48784.
[27] HU W G, ZHANG L.High-throughput calculation and machine learning of two-dimensional halide perovskite materials: formation energy and band gap[J]. Materials today communications, 2023, 35: 105841.
[28] OBADA D O, OKAFOR E, ABOLADE S A, et al.Explainable machine learning for predicting the band gaps of ABX3 perovskites[J]. Materials science in semiconductor processing, 2023, 161: 107427.
[29] KIM S, JEONG Y, HAN D W, et.al. Machine learning-assisted defect analysis and optimization for P-I-N-structured perovskite solar cells[J]. Journal of electronic materials, 2023, 52(9): 5861-5871.
[30] LI Z Z, XU Q C, SUN Q D, et.al. Thermodynamic stability landscape of halide double perovskites via high-throughput computing and machine learning[J]. Advanced functional materials, 2019, 29(9): 1807280.
基金
云南省教育厅科学研究基金（2024Y152）
PDF(5616 KB)
