Improving the accuracy of density-functional theory calculation: The genetic algorithm and neural network approach

Li, Hui; Shi, Lili; Zhang, Min; Su, Zhong‐Min; Wang, XiuJun; Hu, Lihong; Chen, Guanhua

doi:10.1063/1.2715579

Cited by 44 publications

(33 citation statements)

References 11 publications

(10 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The biggest problem associated with the calculation of TM containing systems is the near degeneracy stemming from electrons partially occupying the d orbitals. Recently, several studies have been conducted to evaluate the performance of different DFT methods for predicting several molecular properties and studying reactions involving TMs3,4, also several studies have been carried out to improve the performance of DFT functionals5,6. However, in order to obtain very accurate and reliable calculations for the TM containing systems, one must implement multireference methods such as Multi-Reference Configuration Interaction (MRCI).…”

Section: Introductionmentioning

confidence: 99%

Assessment of the “6-31+G** + LANL2DZ” Mixed Basis Set Coupled with Density Functional Theory Methods and the Effective Core Potential: Prediction of Heats of Formation and Ionization Potentials for First-Row-Transition-Metal Complexes

2009

View full text Add to dashboard Cite

Computational chemists have long demonstrated great interest in finding ways to reliably and accurately predict the molecular properties for transition metal containing complexes. This manuscript is a continuation of our validation efforts of Density Functional Theory (DFT) methods when applied to transition metal containing systems (K. E. Riley; K. M. Merz, Jr. J. Phys. Chem. 2007, 111, 6044-6053). In our previous work we examined DFT using all-electron basis sets, but approaches incorporating effective core potentials (ECPs) are effective in reducing computational expense. With this in mind, our efforts were expanded to include evaluation of the performance of the basis set derived to approximate such an approach as well on the same set of density functionals. Indeed, employing an ECP basis such as LANL2DZ for transition metals, while using all-electron basis sets for all other non-transition-metal atoms has become more and more popular in computations on transition metal containing systems. In this study, we assess the performance of twelve different DFT functionals, from GGA, hybrid-GGA, meta-GGA and hybrid-meta-GGA classes respectively, along with the 6-31+G** + LANL2DZ (on the transition metal) mixed basis set on predicting two important molecular properties: heats of formation and ionization potentials, for 94 and 58 systems containing first row transition metals from Ti to Zn, which are all in the third row of the periodic table. An interesting note is that the inclusion of the exact exchange term in density functional methods generally increases the accuracy of ionization potentials prediction for the hybrid-GGA methods but decreases the reliability of determining the heats of formation for transition metal containing complexes for all hybrid density functional methods. The hybrid-GGA functional B3LYP gives the best performance on predicting the ionization potentials while the meta-GGA functional TPSSTPSS provides the most reliable and accurate results for heats of formation calculations. TPSSTPSS, a meta-GGA functional, which was constructed from first principles and subject to known exact constraints just like in an "ab initio" way, is successful in predicting both the ionization potentials and the heats of formation for transition metal containing systems.

show abstract

Section: Introductionmentioning

confidence: 99%

Assessment of the “6-31+G** + LANL2DZ” Mixed Basis Set Coupled with Density Functional Theory Methods and the Effective Core Potential: Prediction of Heats of Formation and Ionization Potentials for First-Row-Transition-Metal Complexes

2009

View full text Add to dashboard Cite

show abstract

“…Comparing the descriptors selected by different feature selection methods, it should be noted that most methods choose four descriptors (18,24,25,28) for J SC . Similarly, nine descriptors (1,4,8,11,16,23,24,25,26), eight descriptors (1,2,3,8,16,17,25,31), and three descriptors (3,9,16) are selected by most methods for V OC , FF, and PCE, respectively. This result agrees well with some experimental findings.…”

Section: Feature Selectionmentioning

confidence: 99%

“…Five descriptors (11,13,14,16,18) calculated by 631G are retained by more than four feature selection methods for J SC . Likewise, eight descriptors (4,9,13,16,19,24,26,31) for V OC , seven descriptors (3,4,7,12,13,16,26) for FF, and ten descriptors (4,6,7,9,14,15,18,20,24,25) for PCE are also retained by more than four feature selection methods. Although the selected descriptors based on STO and 631G are different, the most important molecular descriptors, such as descriptors (18) for J SC , (4,16,24) for V OC , (3,16) for FF and (9) for PCE are retained by most of the feature selection methods for both 631G and STO calculations.…”

Section: Feature Selectionmentioning

confidence: 99%

“…Quantum chemical methods can capture the physical essence of molecular systems, and machine learning methods can explore the central relationship between molecular structures and the target properties or observed activities through the quantum chemical molecular descriptors. [14][15][16][17] As a result, we suggest applying quantum chemical methods together with machine learning methods to thoroughly investigate the quantitative structure-activity relationship (QSAR) of all-organic dyes in DSSCs that use a semiconductor TiO 2 film and a liquid electrolyte. With the QSAR model, the PCE of new organic dye sensitizers can be predicted, which cannot only save numerous resources but also provide some guidance for the synthesis of highly efficient dye sensitizers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A cascaded QSAR model for efficient prediction of overall power conversion efficiency of all-organic dye-sensitized solar cells

Zhong

Lin

et al. 2015

J. Comput. Chem.

Self Cite

View full text Add to dashboard Cite

A cascaded model is proposed to establish the quantitative structure-activity relationship (QSAR) between the overall power conversion efficiency (PCE) and quantum chemical molecular descriptors of all-organic dye sensitizers. The cascaded model is a two-level network in which the outputs of the first level (JSC, VOC, and FF) are the inputs of the second level, and the ultimate end-point is the overall PCE of dye-sensitized solar cells (DSSCs). The model combines quantum chemical methods and machine learning methods, further including quantum chemical calculations, data division, feature selection, regression, and validation steps. To improve the efficiency of the model and reduce the redundancy and noise of the molecular descriptors, six feature selection methods (multiple linear regression, genetic algorithms, mean impact value, forward selection, backward elimination, and +n-m algorithm) are used with the support vector machine. The best established cascaded model predicts the PCE values of DSSCs with a MAE of 0.57 (%), which is about 10% of the mean value PCE (5.62%). The validation parameters according to the OECD principles are R(2) (0.75), Q(2) (0.77), and Qcv2 (0.76), which demonstrate the great goodness-of-fit, predictivity, and robustness of the model. Additionally, the applicability domain of the cascaded QSAR model is defined for further application. This study demonstrates that the established cascaded model is able to effectively predict the PCE for organic dye sensitizers with very low cost and relatively high accuracy, providing a useful tool for the design of dye sensitizers with high PCE.

show abstract

“…In addition, the X1 method was also used to evaluate the dissociation energies of 142 chemical bonds and an accuracy of 2.45 kcal/mol was reported. [8] Our group [9] developed electronic excitation energies of 150 organic molecules. GANN was compared with the standard NN approach on the same problems.…”

Section: Introductionmentioning

confidence: 99%

An effective method for accurate prediction of the first hyperpolarizability of alkalides

Wang

Sun

et al. 2011

J Comput Chem

Self Cite

View full text Add to dashboard Cite

The proper theoretical calculation method for nonlinear optical (NLO) properties is a key factor to design the excellent NLO materials. Yet it is a difficult task to obatin the accurate NLO property of large scale molecule. In present work, an effective intelligent computing method, as called extreme learning machine-neural network (ELM-NN), is proposed to predict accurately the first hyperpolarizability (β(0)) of alkalides from low-accuracy first hyperpolarizability. Compared with neural network (NN) and genetic algorithm neural network (GANN), the root-mean-square deviations of the predicted values obtained by ELM-NN, GANN, and NN with their MP2 counterpart are 0.02, 0.08, and 0.17 a.u., respectively. It suggests that the predicted values obtained by ELM-NN are more accurate than those calculated by NN and GANN methods. Another excellent point of ELM-NN is the ability to obtain the high accuracy level calculated values with less computing cost. Experimental results show that the computing time of MP2 is 2.4-4 times of the computing time of ELM-NN. Thus, the proposed method is a potentially powerful tool in computational chemistry, and it may predict β(0) of the large scale molecules, which is difficult to obtain by high-accuracy theoretical method due to dramatic increasing computational cost.

show abstract

Improving the accuracy of density-functional theory calculation: The genetic algorithm and neural network approach

Cited by 44 publications

References 11 publications

Assessment of the “6-31+G** + LANL2DZ” Mixed Basis Set Coupled with Density Functional Theory Methods and the Effective Core Potential: Prediction of Heats of Formation and Ionization Potentials for First-Row-Transition-Metal Complexes

Assessment of the “6-31+G** + LANL2DZ” Mixed Basis Set Coupled with Density Functional Theory Methods and the Effective Core Potential: Prediction of Heats of Formation and Ionization Potentials for First-Row-Transition-Metal Complexes

A cascaded QSAR model for efficient prediction of overall power conversion efficiency of all-organic dye-sensitized solar cells

An effective method for accurate prediction of the first hyperpolarizability of alkalides

Contact Info

Product

Resources

About