Hybrid evolution method for global optimization of surface catalytic structure accelerated by machine learning

Abstract

The invention discloses a hybrid evolution method for global optimization of a surface catalytic structure accelerated by machine learning. A hybrid evolution algorithm in which a genetic algorithm and a difference algorithm are coordinated is introduced. When the structure is generated, a part of the structure is generated by using a genetic algorithm (interlace operation), and a part of the structure is generated by using a differential algorithm (differential operation). Compared with a genetic algorithm or an evolutionary algorithm which is singly used, the hybrid evolutionary algorithm provided by the invention has a better effect when being used for processing a surface catalytic structure with a relatively large number of atoms. According to the method, a Gaussian process regression model in the field of machine learning is introduced and used, and (Gaussian) posterior distribution is constructed by establishing a corresponding relation between atomic coordinates of a structure and energy of the atomic coordinates, so that targeted sampling on a potential energy surface is realized, and the efficiency of an optimization process is enhanced.

Description

technical field [0001] The invention relates to a method for optimizing a supported surface catalytic structure. In particular, it involves a hybrid evolutionary approach for the global optimization of surface catalytic structures accelerated by machine learning. Background technique [0002] A potential energy surface represents the functional relationship between the energy of a chemical system and related parameters (usually atomic coordinates). Matter usually exists in its lowest-energy structure, which is the global minimum point on the potential energy surface. Understanding the energy minimum structure (and its atomic coordinates) of surface catalytic systems is crucial for predicting catalyst structures, analyzing adsorption properties, studying heterogeneous catalytic reaction mechanisms, and explaining experimental phenomena, and provides a basis for rational design of catalysts. [0003] The potential energy surface is highly complex, and as the number of atoms ...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a method that can overcome the disadvantages of the existing method that the prediction ability of the complex system of the high-dimensional potential energy surface is not strong, efficiently optimize the complex surface catalytic system, and obtain the lowest energy of the system. A Hybrid Evolutionary Approach for Global Optimization of Surface Catalytic Structures Accelerated by Structural Machine Learning

Images