Evaluating data-driven algorithms for predicting mechanical properties with small datasets: A case study on gear steel hardenability

Data-driven algorithms for predicting mechanical properties with small datasets are evaluated in a case study on gear steel hardenability. The limitations of current data-driven algorithms and empirical models are identified. Challenges in analysing small datasets are discussed, and solution is prop...

Full description

Saved in:
Bibliographic Details
Published in:International journal of minerals, metallurgy and materials metallurgy and materials, 2022-04, Vol.29 (4), p.836-847
Main Authors: Nenchev, Bogdan, Tao, Qing, Dong, Zihui, Panwisawas, Chinnapat, Li, Haiyang, Tao, Biao, Dong, Hongbiao
Format: Article
Language:English
Subjects:
Algorithms
Alloying elements
Case studies
Ceramics
Characterization and Evaluation of Materials
Chemistry and Materials Science
Composites
Corrosion and Coatings
Datasets
Empirical equations
Glass
Hardenability
Homogeneity
Inhomogeneity
Machine learning
Materials Science
Mechanical properties
Metallic Materials
Natural Materials
Neural networks
Performance evaluation
Steel
Surfaces and Interfaces
Thin Films
Tribology
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data-driven algorithms for predicting mechanical properties with small datasets are evaluated in a case study on gear steel hardenability. The limitations of current data-driven algorithms and empirical models are identified. Challenges in analysing small datasets are discussed, and solution is proposed to handle small datasets with multiple variables. Gaussian methods in combination with novel predictive algorithms are utilized to overcome the challenges in analysing gear steel hardenability data and to gain insight into alloying elements interaction and structure homogeneity. The gained fundamental knowledge integrated with machine learning is shown to be superior to the empirical equations in predicting hardenability. Metallurgical-property relationships between chemistry, sample size, and hardness are predicted via two optimized machine learning algorithms: neural networks (NNs) and extreme gradient boosting (XGboost). A comparison is drawn between all algorithms, evaluating their performance based on small data sets. The results reveal that XGboost has the highest potential for predicting hardenability using small datasets with class imbalance and large inhomogeneity issues.
ISSN:1674-4799
1869-103X
DOI:10.1007/s12613-022-2437-0