Combining machine learning algorithm to improve prediction performance of ab initio method for vibrational energy spectra of HF/HBr/H³⁵Cl/Na³⁵Cl

Abstract

Halides play an important role in atmospheric chemistry, corrosion of steel, and in controlling the abundance of O<sub>3</sub>.Moreover high-precision vibrational energy spectra contain a large amount of quantum information of molecular system and is essential basic data for people to understand and manipulate molecules. At present, ab-initio methods have achieved great results in the calculation of the potential energy surfaces and corresponding vibrational energies of molecules, but they still face challenges in terms of accuracy and computational cost. Recently, data-driven machine learning methods have demonstrated a very strong ability to extract high-dimensional functional relationships from massive data and have been widely used in spectroscopy studies. Therefore, a theoretical approach combining ab-initio methods and machine learning algorithms is presented here to predict the vibrational energy of diatomic systems, which improves the accuracy and simultaneously reduces the computational cost. Firstly, the vibrational energy levels of 42 diatomic molecules were obtained with different CCSD(T) method calculation configurations form simple to complex and the corresponding experimental results are also collected. A machine learning algorithm is then used to learn the difference between the calculated CCSD(T) and experimental vibrational results, and a high-dimensional error function is finally constructed to improve the original CCSD(T) computational accuracy. The results for HF, HBr, H<sup>35</sup>Cl and Na<sup>35</sup>Cl (they did not appear in the training set) and other halogen molecules show that the current method reduces the prediction error by more than 50% and the computational cost by nearly one order of magnitude compared with the CCSD(T)/cc-pV5Z calculation method alone. It is worth noting that the method proposed in this paper is not only limited to the energy level prediction of diatomic systems, but also can be used in other fields where data can be obtained by ab initio methods and experimental methods simultaneously, such as the energy spectrum properties of macromolecular systems.