Novel quantum chemical molecule performance prediction method and system

Abstract

The invention discloses a novel quantum chemical molecule performance prediction method and system. The method comprises the following steps: obtaining structure parameters of chemical molecules to be tested; obtaining parameter data related to the prediction performance; performing geometric structure optimization on the chemical molecules to be predicted by using quantum chemistry software Gaussian to obtain quantum chemistry parameters; the obtained parameter data, related to the prediction performance and representing the same performance, of the chemical molecules to be tested are sorted according to the numerical value size and divided into a verification set and a training set according to the proportion of 1: 3, and the training set is used for establishing a prediction model; and taking the obtained molecular descriptors as independent variables, taking parameter data basis which is related to the predicted performance and represents the same performance as dependent variables, establishing a quantitative relation model of various structure descriptors and the predicted performance, and substituting verification set data into an obtained regression equation to obtain a predicted value. According to the method, various performances of molecules can be quickly predicted through a quantum chemical calculation method, and the method has the characteristics of high prediction speed and high prediction efficiency.

Description

technical field [0001] The invention relates to the technical field of temperature controllers, in particular to a method and system for predicting the properties of novel quantum chemical molecules. Background technique [0002] In 1900 Planck proposed quantum theory. In 1926, the Schrodinger equation came out, marking the establishment of quantum mechanics. In 1927, Heitler and London used the basic principles of quantum mechanics to study the molecular structure of hydrogen, marking the birth of a new discipline of quantum chemistry. With the development of theoretical methods of quantum chemistry and the advancement of computer technology, quantum chemistry has been applied to various disciplines and fields for more than 80 years. Notable achievements, in the prediction of molecular properties, most of the existing methods are deep learning methods to predict the properties of molecules, the prediction effect is average, a large amount of sample data is required, and t...

AI Technical Summary

Problems solved by technology

With the development of theoretical methods of quantum chemistry and the advancement of computer technology, quantum chemistry has been applied to various disciplines and fields for more than 80 years. Notable achievements. In the prediction of molecular properties, most of the existing methods are deep learning methods to predict the properties of molecules. The prediction effect is average, requiring a large amount of sample data, and cannot quickly predict molecular properties.

Images