Prediction of uniaxial compressive strength of rock using machine learning
By: Dadhich, Sharad.
Contributor(s): Sharma, Jitendra Kumar.
Publisher: USA Springer 2022Edition: Vol.103(4), Dec.Description: 1209-1224p.Subject(s): Humanities and Applied SciencesOnline resources: Click here In: Journal of the institution of engineers (India): Series ASummary: The uniaxial compressive strength of rock is one of the most significant parameters required for analysis of rock mass, its characterization, and design of foundations. Direct determination of the uniaxial compressive strength of rock is time-consuming, expensive, and requires destructive laboratory or field testing. Therefore, indirect methods based on regression analysis are widely used for estimation of the uniaxial compressive strength of rock, which have less accuracy. In this study, machine learning algorithms are used to estimate the uniaxial compressive strength of rock using point load strength, porosity, Schmidt rebound hardness, block punch index, and specific gravity. The performance of each machine learning model is evaluated using statistical parameters, viz., mean absolute error, value account for, and coefficient of determination. It is found that random forest regression is the most suitable model for estimation of uniaxial compressive strength with the minimum mean absolute error of 8.68 MPa and r2-score of 0.94.Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
---|---|---|---|---|---|---|
Articles Abstract Database | School of Engineering & Technology Archieval Section | Not for loan | 2023-0356 |
The uniaxial compressive strength of rock is one of the most significant parameters required for analysis of rock mass, its characterization, and design of foundations. Direct determination of the uniaxial compressive strength of rock is time-consuming, expensive, and requires destructive laboratory or field testing. Therefore, indirect methods based on regression analysis are widely used for estimation of the uniaxial compressive strength of rock, which have less accuracy. In this study, machine learning algorithms are used to estimate the uniaxial compressive strength of rock using point load strength, porosity, Schmidt rebound hardness, block punch index, and specific gravity. The performance of each machine learning model is evaluated using statistical parameters, viz., mean absolute error, value account for, and coefficient of determination. It is found that random forest regression is the most suitable model for estimation of uniaxial compressive strength with the minimum mean absolute error of 8.68 MPa and r2-score of 0.94.
There are no comments for this item.