A data- and model-driven strategy for the evaluation of the experimental transition lines: Theoretical prediction for the ground state of 12C16O

An analytical formula that relates the molecular constants of the Herzberg expression and experimental transition lines is developed herein with a difference algebraic approach (DAA) model. Based on the datadriven strategy, the DAA model is able to deal with the tiny uncertainties that exhibit behind the experimental transition lines, which is applied to the P branch emission spectra of some first overtone bands of the ground electronic state of (CO)-C-12-O-16. The relationship can be used to generate transition lines with sufficient accuracy, as evident from the high J of agreement with the HITRAN database, Velichko data, Goorvitch data and quantum-mechanical data. In addition, line intensities, absorption oscillator strengths and Einstein A coefficients of these lines, which are introduced to enhance the dataset and are in good agreement with those of other authors, are also reported to validate our results. These various comparative results show that the proposed data-driven strategy based on the DAA model is expecting to be a good algorithm that relies on relatively limited data for training. (C) 2021 Elsevier B.V. All rights reserved.

Automated summary

In this study, an analytical formula relating molecular constants with experimental transition lines was developed using a difference algebraic approach (DAA) model. The data-driven strategy based on the DAA model was successfully applied to generate transition lines with sufficient accuracy, effectively handling tiny uncertainties behind the experimental transition lines. Comparisons with various databases and quantum mechanical data showed high agreement, validating the results and the potential of the proposed algorithm relying on limited training data.