A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Bandgap Prediction of Hybrid Organic–Inorganic Perovskite Solar Cell Using Machine Learning
https://orcid.org/http://orcid.org/0000-0001-8938-7659
https://orcid.org/http://orcid.org/0000-0002-6774-4579
Perovskite solar cells (PSCs) have been emerged as most promising third-generation solar cell technology among solar cell industry. The fabrication processes of highly efficient PSCs with/ without different interfacial layers are time consuming, costly and it required inert atmosphere. In this aspect, we have used autonomous experimentation toolkits like linear regression or other modern machine learning algorithms by preparing a dataset of 155 data points collected from recently published literatures covering various constituent properties of different perovskites absorbers. Those datasets are analyzed using Linear Regression (LR), K Nearest Neighbor, Random Forest and Neural Network algorithms for predicting the bandgap of the perovskite absorber and stable perovskite model toward higher solar absorption and power conversion efficiency (PCE). LR algorithm gives the best result among the all four algorithms by predicting the bandgap range of 1.55–3.02 eV for the perovskite absorber with the formula of CsaFAbMA(1−a−b)Pb(ClxBryI(1−x−y))3 (FA = Formamidinium, MA = methylammonium) by considering wide range of compositions of organic–inorganic cations and halide anions. This model shows highest R2-value of 0.99 which indicates best accuracy of 99% for the prepared dataset with a least RMSE of 0.0617.
Bandgap Prediction of Hybrid Organic–Inorganic Perovskite Solar Cell Using Machine Learning
https://orcid.org/http://orcid.org/0000-0001-8938-7659
https://orcid.org/http://orcid.org/0000-0002-6774-4579
Perovskite solar cells (PSCs) have been emerged as most promising third-generation solar cell technology among solar cell industry. The fabrication processes of highly efficient PSCs with/ without different interfacial layers are time consuming, costly and it required inert atmosphere. In this aspect, we have used autonomous experimentation toolkits like linear regression or other modern machine learning algorithms by preparing a dataset of 155 data points collected from recently published literatures covering various constituent properties of different perovskites absorbers. Those datasets are analyzed using Linear Regression (LR), K Nearest Neighbor, Random Forest and Neural Network algorithms for predicting the bandgap of the perovskite absorber and stable perovskite model toward higher solar absorption and power conversion efficiency (PCE). LR algorithm gives the best result among the all four algorithms by predicting the bandgap range of 1.55–3.02 eV for the perovskite absorber with the formula of CsaFAbMA(1−a−b)Pb(ClxBryI(1−x−y))3 (FA = Formamidinium, MA = methylammonium) by considering wide range of compositions of organic–inorganic cations and halide anions. This model shows highest R2-value of 0.99 which indicates best accuracy of 99% for the prepared dataset with a least RMSE of 0.0617.
Bandgap Prediction of Hybrid Organic–Inorganic Perovskite Solar Cell Using Machine Learning
https://orcid.org/http://orcid.org/0000-0001-8938-7659
https://orcid.org/http://orcid.org/0000-0002-6774-4579
Perovskite solar cells (PSCs) have been emerged as most promising third-generation solar cell technology among solar cell industry. The fabrication processes of highly efficient PSCs with/ without different interfacial layers are time consuming, costly and it required inert atmosphere. In this aspect, we have used autonomous experimentation toolkits like linear regression or other modern machine learning algorithms by preparing a dataset of 155 data points collected from recently published literatures covering various constituent properties of different perovskites absorbers. Those datasets are analyzed using Linear Regression (LR), K Nearest Neighbor, Random Forest and Neural Network algorithms for predicting the bandgap of the perovskite absorber and stable perovskite model toward higher solar absorption and power conversion efficiency (PCE). LR algorithm gives the best result among the all four algorithms by predicting the bandgap range of 1.55–3.02 eV for the perovskite absorber with the formula of CsaFAbMA(1−a−b)Pb(ClxBryI(1−x−y))3 (FA = Formamidinium, MA = methylammonium) by considering wide range of compositions of organic–inorganic cations and halide anions. This model shows highest R2-value of 0.99 which indicates best accuracy of 99% for the prepared dataset with a least RMSE of 0.0617.
Bandgap Prediction of Hybrid Organic–Inorganic Perovskite Solar Cell Using Machine Learning
J. Inst. Eng. India Ser. D
Sadhu, Debmalya (author) / De, Debasis (author) / Dattatreya, Devansh (author) / Deo, Arjun (author) / Gupta, Subir (author)
Journal of The Institution of Engineers (India): Series D ; 105 ; 795-801
2024-08-01
7 pages
Article (Journal)
Electronic Resource
English
Bandgap Prediction of Hybrid Organic–Inorganic Perovskite Solar Cell Using Machine Learning
| Springer Verlag | 2024
Steps toward efficient inorganic-organic hybrid perovskite solar cells
| British Library Online Contents | 2015
| British Library Online Contents | 2016
| British Library Online Contents | 2016