Study on bandgap predications of ABX3-type perovskites by machine learning

Liu, Yiming; Yan, Wensheng; Zhu, Heng; Tu, Yiteng; Guan, Li; Tan, Xinyu

doi:10.1016/j.orgel.2021.106426

Cited by 25 publications

(31 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Interestingly, the addition of Cl can significantly increase the band gap value. 10,11 Feature Engineering and Model Construction. In order to validate the accuracy of the data set and comparison with the subsequent model with physical features as input, we simultaneously trained models with composition (ML-C) as features using multiple algorithms, including GBRT, ANN, k-nearest neighbor (KNN), support vector regression (SVR), and the 10-fold cross-validation grid search was used to find the optimal parameters for all the ML models.…”

Section: Genetic Algorithm-based Symbolic Regression (Gasr)mentioning

confidence: 99%

“…The model performances are shown in Table S3, where GBRT-C has the highest accuracy, and the RMSE of the test set is 0.046 eV, which is currently the highest accuracy compared to the work of Li and Liu mentioned earlier. 10,11 To further enhance the physical interpretability, we used physical features to replace the simple atomic ratio features as the model inputs (ML-P). Previous works have shown that the excellent model's performance can be obtained by employing electronic propertiesrelated and structure-related features such as electronegativity, Goldschmidt tolerance factor (τ), and octahedral factor (μ).…”

Section: Genetic Algorithm-based Symbolic Regression (Gasr)mentioning

confidence: 99%

“…It is reported that it is difficult to control the Cl content in the annealing process and the band gap is easy to increase in the presence of Cl elements, so all Cl-containing compositions are excluded. 10,11 We therefore constructed the prediction set optimize the computational efficiency, a random selection of 100,000 compositions which has the same feature space distribution as the initial prediction set was made for prediction using GBRT-P. Then, we selected compositions that are not in the original data set and whose band gaps are distributed in the range of 1.3−1.4 and 1.7− 2.1 eV. Finally, we randomly selected three groups among the remaining compositions for the experimental synthesis.…”

Section: Genetic Algorithm-based Symbolic Regression (Gasr)mentioning

confidence: 99%

See 2 more Smart Citations

Accelerating the Discovery of Hybrid Perovskites with Targeted Band Gaps via Interpretable Machine Learning

Yang,

Chong,

et al. 2023

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

The band gap of hybrid organic–inorganic perovskites (HOIPs) is a key factor affecting the light absorption characteristics and thus the performance of perovskite solar cells (PSCs). However, band gap engineering, using experimental trial and error and high-throughput density functional theory calculations, is blind and costly. Therefore, it is critical to statistically identify the multiple factors influencing band gaps and to rationally design perovskites with targeted band gaps. This study combined feature engineering, the gradient-boosted regression tree (GBRT) algorithm, and the genetic algorithm-based symbolic regression (GASR) algorithm to develop an interpretable machine learning (ML) strategy for predicting the band gap of HOIPs accurately and quantitatively interpreting the factors affecting the band gap. Seven best physical features were selected to construct a GBRT model with a root-mean-square error of less than 0.060 eV, and the most important feature is the electronegativity difference between the B-site and the X-site (χB–X). Further, a mathematical formula (E g = χB–X 2 + 0.881χB–X) was constructed with GASR for a quantitative interpretation of the band gap influence patterns. According to the ML model, the HOIP MA0.23FA0.02Cs0.75Pb0.59Sn0.41Br0.24I2.76 was obtained, with a suitable band gap of 1.39 eV. Our proposed interpretable ML strategy provides an effective approach for developing HOIP structures with targeted band gaps, which can also be applied to other material fields.

show abstract

Section: Genetic Algorithm-based Symbolic Regression (Gasr)mentioning

confidence: 99%

Section: Genetic Algorithm-based Symbolic Regression (Gasr)mentioning

confidence: 99%

Section: Genetic Algorithm-based Symbolic Regression (Gasr)mentioning

confidence: 99%

See 1 more Smart Citation

Accelerating the Discovery of Hybrid Perovskites with Targeted Band Gaps via Interpretable Machine Learning

Yang,

Chong,

et al. 2023

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

show abstract

“…Data purification means that the data in the database are retained as much as possible of the same type or similar to the prediction target, which can greatly improve the accuracy of the ML model. For example, Liu et al [76] built a bandgap database of ABX 3 perovskites and performed bandgap prediction, which further reduced the generalization error of the ML models by refining the types of ABX 3 perovskite (i.e., all-inorganic perovskites, hybrid organic perovskites, Tin-free perovskites). Kim et al [77] used the Light GBM algorithm model to predict the properties (formation energy, convex hull energy, bandgap) of A 2 BB'X 3 double perovskite.…”

Section: Data Sourcesmentioning

confidence: 99%

Machine Learning for Perovskite Solar Cells and Component Materials: Key Technologies and Prospects

Liu

Tan

Liang

et al. 2023

Adv Funct Materials

Self Cite

View full text Add to dashboard Cite

Data-driven epoch, the development of machine learning (ML) in materials and device design is an irreversible trend. Its ability and efficiency to handle nonlinear and game-playing problems is unmatched by traditional simulation computing software and trial-error experiments. Perovskite solar cells are complex physicochemical devices (systems) that consist of perovskite materials, transport layer materials, and electrodes. Predicting the physicochemical properties and screening the component materials related to perovskite solar cells is the strong point of ML. However, the applications of ML in perovskite solar cells and component materials has only begun to boom in the last two years, so it is necessary to provide a review of the involved ML technologies, the application status, the facing urgent challenges and the development blueprint.

show abstract

“…In recent years, machine-learning algorithms have been widely used as a data-driven modeling method in many fields, such as geotechnical engineering [ 27 ], traffic safety [ 28 ], material engineering [ 29 ], and biomedicine [ 30 ]. To improve computational accuracy and computational efficiency, Breima proposed the random forest (RF) algorithm in 2001 [ 31 ].…”

Section: Introductionmentioning

confidence: 99%

Prediction of Both E-Jet Printing Ejection Cycle Time and Droplet Diameter Based on Random Forest Regression

et al. 2023

View full text Add to dashboard Cite

Electrohydrodynamic jet (E-jet) printing has broad application prospects in the preparation of flexible electronics and optical devices. Ejection cycle time and droplet size are two key factors affecting E-jet-printing quality, but due to the complex process of E-jet printing, it remains a challenge to establish accurate relationships among ejection cycle time and droplet diameter and printing parameters. This paper develops a model based on random forest regression (RFR) for E-jet-printing prediction. Trained with 72 groups of experimental data obtained under four printing parameters (voltage, nozzle-to-substrate distance, liquid viscosity, and liquid conductivity), the RFR model achieved a MAPE (mean absolute percent error) of 4.35% and an RMSE (root mean square error) of 0.04 ms for eject cycle prediction, as well as a MAPE of 2.89% and an RMSE of 0.96 μm for droplet diameter prediction. With limited training data, the RFR model achieved the best prediction accuracy among several machine-learning models (RFR, CART, SVR, and ANN). The proposed prediction model provides an efficient and effective way to simultaneously predict the ejection cycle time and droplet diameter, advancing E-jet printing toward the goal of accurate, drop-on-demand printing.

show abstract

Study on bandgap predications of ABX3-type perovskites by machine learning

Cited by 25 publications

References 41 publications

Accelerating the Discovery of Hybrid Perovskites with Targeted Band Gaps via Interpretable Machine Learning

Accelerating the Discovery of Hybrid Perovskites with Targeted Band Gaps via Interpretable Machine Learning

Machine Learning for Perovskite Solar Cells and Component Materials: Key Technologies and Prospects

Prediction of Both E-Jet Printing Ejection Cycle Time and Droplet Diameter Based on Random Forest Regression

Contact Info

Product

Resources

About