Material structure classification method and system based on machine learning clustering algorithm

Abstract

The invention provides a material structure classification method and system based on a machine learning clustering algorithm, and the method comprises the following steps: building a data set: collecting material structure sample data of a multi-layer composite film, and obtaining discrete dimension features and continuous dimension features of the sample data; feature preprocessing: carrying out standardization processing on the feature values; performing clustering analysis on the data set after feature preprocessing through a clustering algorithm, then performing modeling, and evaluating the model according to a clustering result output by the model to obtain an optimal clustering algorithm; and outputting a clustering result of the model obtained based on the optimal clustering algorithm, counting the percentage of the data volume of the material structure classified into each category in the total data volume of the material structure according to the clustering result, and selecting the category with the maximum proportion as a classification result. According to the method, the dimension features of the composite film are analyzed according to an artificial intelligence technology, and accurate classification of various material structures is realized by processing five dimension features of sample data.

Description

technical field [0001] The invention relates to the technical field of material structure classification in the process of formulating material product standards, and in particular, relates to a material structure classification method and system based on a machine learning clustering algorithm. Background technique [0002] In the process of formulating material product standards, it is often necessary to classify the material structure of a series of products within the scope of the standard, and to specify different technical parameter requirements for products with different material structures. In order to enable product standards to cover a wider variety of material structures, and to enable the technical requirement parameters of each material structure to more accurately determine the degree of product qualification, a series of product material structures within the standard range are accurately classified at the initial stage of standard formulation. classes are es...

AI Technical Summary

Problems solved by technology

[0005] The object of the present invention is to provide a material structure classification method and system based on a machine learning clustering algorithm, aiming to solve the problems pointed out in the background technology

Images