Application of Various Nonlinear Models to Predict the Uniaxial Compressive Strength of Weakly Cemented Jurassic Rocks

Weakly cemented rocks are distributed widely in Jurassic coal-bearing strata in western China. Their uniaxial compressive strength (UCS), as a crucial mechanical property, is generally used in various rock engineering practices. However, the mechanical weakness of weakly cemented rocks makes it chal...

Full description

Saved in:
Bibliographic Details
Published in:Natural resources research (New York, N.Y.) N.Y.), 2022-02, Vol.31 (1), p.371-384
Main Authors: Wang, Zhenkang, Li, Wenping, Chen, Jiangfeng
Format: Article
Language:English
Subjects:
Adaptive systems
Artificial neural networks
Chemistry and Earth Sciences
Coal
Compressive strength
Computer Science
Correlation coefficient
Correlation coefficients
Earth and Environmental Science
Earth Sciences
Feasibility studies
Fossil Fuels (incl. Carbon Capture)
Fuzzy logic
Fuzzy systems
Geography
Inference
Jurassic
Mathematical Modeling and Industrial Mathematics
Mechanical properties
Mineral Resources
Modulus of elasticity
Multiple regression analysis
Neural networks
Original Paper
P waves
Performance prediction
Physics
Predictions
Regression analysis
Rocks
Soft computing
Statistical analysis
Statistics for Engineering
Sustainable Development
Tensile strength
Wave velocity
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:Weakly cemented rocks are distributed widely in Jurassic coal-bearing strata in western China. Their uniaxial compressive strength (UCS), as a crucial mechanical property, is generally used in various rock engineering practices. However, the mechanical weakness of weakly cemented rocks makes it challenging to prepare standard samples of such rocks for testing. To avoid this problem, this study assessed the feasibility of predicting the UCS of weakly cemented rocks using their unit weight, P-wave velocity, tensile strength, and elastic modulus. Several statistical and soft computing methods, namely multiple regression analysis, artificial neural networks, and adaptive neuro-fuzzy inference systems were applied for predicting the UCS of weakly cemented rocks. In addition, their prediction performances were assessed in accordance with root-mean-squared error, value account for, and correlation coefficient. Results of comparison indicated that all of the models employed in this study had good prediction performances, but the adaptive neuro-fuzzy inference system was the most suitable approach for the prediction of the UCS of weakly cemented rocks.
ISSN:1520-7439
1573-8981
DOI:10.1007/s11053-021-09970-x