1. Prediction of bending strength of Si3N4 using machine learning
- Author
-
Ping Yang, Haonan Wu, Shuangshuang Wu, Donglin Lu, Wenjing Zou, Yuanzhi Shao, Luojing Chu, and Shanghua Wu
- Subjects
Materials science ,Sintering ,02 engineering and technology ,Machine learning ,computer.software_genre ,01 natural sciences ,chemistry.chemical_compound ,Flexural strength ,0103 physical sciences ,Materials Chemistry ,Ceramic ,Extreme gradient boosting ,010302 applied physics ,business.industry ,Process Chemistry and Technology ,021001 nanoscience & nanotechnology ,Surfaces, Coatings and Films ,Electronic, Optical and Magnetic Materials ,Silicon nitride ,chemistry ,visual_art ,Ceramics and Composites ,visual_art.visual_art_medium ,Artificial intelligence ,Experimental methods ,0210 nano-technology ,business ,computer - Abstract
The bending strength of silicon nitride (Si3N4) plays a vital role in its application and is influenced by various process factors. Current experimental methods for investigating Si3N4 ceramics exhibiting low efficiency and high cost are incapable of systematically analysing the effect of process factors on the bending strength of Si3N4 ceramics and quantitatively predicting the optimum process parameters. In this study, machine learning (ML) approaches based on extreme gradient boosting (XGBoost) were applied to predict and analyse the bending strength of Si3N4 ceramics. Because the classification model of XGBoost is easily interpretable, the factors affecting the bending strength could be quantitatively evaluated. The current model can provide a suitable order of adding sintering additives to obtain excellent bending strength in Si3N4 ceramics. Although this study focuses on the bending strength of Si3N4 ceramics, the new approach reported herein is applicable for the in silico design and analysis of other ceramic materials.
- Published
- 2021