A comparative study of various hybrid neural networks and regression analysis to predict unconfined compressive strength
- PDF / 1,396,053 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 14 Downloads / 283 Views
PRACTICE-ORIENTED PAPER
A comparative study of various hybrid neural networks and regression analysis to predict unconfined compressive strength of travertine Mohammad Ebdali1 · Emad Khorasani2 · Sohrab Salehin2 Received: 13 April 2020 / Accepted: 20 July 2020 © Springer Nature Switzerland AG 2020
Abstract In this paper, the relationships between engineering properties of travertine rock samples including uniaxial compressive strength, density, Brazilian tensile strength and compressional and shear wave velocities were evaluated. The Bukan travertine mine located in Iran was considered as case study here. Various data analysis approaches including simple regression method, multiple regression method and artificial neural network (ANN) have been used for finding optimum estimation model for uniaxial compression strength of travertine rocks. Rock sample preparations difficulties and conducting expensive tests such as UCS motivated many researchers to study different regression methods to estimate UCS from other rock mechanic tests. In this paper, different statistical methods as well as some ANN optimization algorithms that were used by several researchers are compared to find the optimum solution for UCS estimation problem of travertine rock samples. These optimization tools comprising genetic algorithm, particle swarm optimization and imperialist competitive algorithm were applied to improve the efficiency of ANN analysis. Finally, after comparing all of the presented methods, the best results were obtained by ANN-PSO algorithm. Keywords Rock properties · Travertine · Artificial neural network · ICA · PSO · GA
Introduction Unconfined compressive strength or shortly UCS is one of the critical parameters for almost any analysis on the subject of rock mechanics and related designs in civil and mining engineering. Moreover, there are several tests and derived indices for describing rocks mechanically; for example, tests such as point load, Brazilian tensile strength and ultrasonic wave velocity are some well-known rock mechanics tests to describe rock behavior in compressive or tensile statically or dynamically loading. Although these tests are easy
* Sohrab Salehin [email protected] Mohammad Ebdali [email protected] Emad Khorasani [email protected] 1
School of Civil Engineering, College of Engineering, University of Tehran, Tehran, Iran
School of Mining Engineering, College of Engineering, University of Tehran, Tehran, Iran
2
to conduct, still UCS test is the major necessary test for extracting rock strength as input for any design (Table 1). American Society for Testing and Materials (ASTM) and International Society for Rock Mechanics have standardized the UCS test. Preparing standard core samples and the time and money consuming nature of UCS test caused many researches to try to find a meaningful relations between other mentioned above easy to perform tests results (point load index ( Is(50) ) , Brazilian tensile strength ( Qt ), compressional and shear wave velocity ( VP and VS ), densi
Data Loading...
