Author(s)

Nie Yu, Li Chao, He Fubao, Hu Kaihua

Corresponding Author

Nie Yu

Abstract

in Order to Describe the Relationship between Ceramic Grinding Surface Texture Features and Roughness, and Realize Rapid Evaluation and Prediction of Ceramic Grinding Surface Roughness, the Gray Level Co-Occurrence Matrix Based on Machine Learning is Used to Extract and Analyze Surface Texture Features. by Using Variable Correlation Analysis, Six Parameters Such as Contrast Are Determined as the Main Texture Features of Si3n4 Grinding Surface, and the Variation Laws of Each Texture Parameter with Ra, Ry, s, Tp Are Discussed, Thus Qualitatively Evaluating the Roughness of Si3n4 Grinding Surface. the Influencing Factors of Gray Level Co-Occurrence Matrix Are Determined According to the Change Curve of Sampling Point Spacing and Gray Level with Characteristic Value. the Gray Level Co-Occurrence Matrix is Established in Four Directions and the Mean Value of All Texture Characteristic Parameters is Calculated. by Analyzing the Correlation between the Characteristic Parameters, Four Parameters Are Determined as the Main Characteristic Parameters of Ceramic Grinding Surface Texture. the Relationship between Texture Characteristic Parameters and Roughness Evaluation Index Was Studied by Using Multiple Nonlinear Regression Method, and Four Regression Prediction Models Were Constructed. the Results Show That the Deviation between the Calculated Value and the Measured Value is Less Than 0.25, Which Has a Good Prediction Effect.

Keywords

Machine Learning; Engineering Ceramics; Texture Analysis; Surface Roughness; Gray Level Co-Occurrence Matrix