Back-analysis for initial ground stress field at a diamond mine using machine learning approaches
- PDF / 1,496,604 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 8 Downloads / 203 Views
Back‑analysis for initial ground stress field at a diamond mine using machine learning approaches Yuanyuan Pu1,2 · Derek B. Apel2 · Stanislaw Prusek3 · Andrzej Walentek3 · Tomasz Cichy3 Received: 26 September 2019 / Accepted: 3 September 2020 © Springer Nature B.V. 2020
Abstract Exact knowledge for ground stress field guarantees the construction of various underground engineering projects as well as prediction of some geological hazards such as the rock burst. Limited by costs, field measurement for initial ground stresses can be only conducted on several measure points, which necessitates back-analysis for initial stresses from limited field measurement data. This paper employed a multioutput decision tree regressor (DTR) to model the relationship between initial ground stress field and its impact factor. A full-scale finite element model was built and computed to gain 400 training samples for DTR using a submodeling strategy. The results showed that correlation coefficient r between field measurement values and back-analysis values reached 0.92, which proved the success of DTR. A neural network was employed to store the global initial ground stress field. More than 600,000 node data extracted from the full-scale finite element model were used to train this neural network. After training, the stresses on any location can be investigated by inputting corresponding coordinates into this neural network. Keywords Initial ground stress field · Full-scale finite element model · Multioutput decision tree regressor · Feed-forward neural network
1 Introduction In many hard rock mines, ground stress field is one of the major concerns due to potential ground failure and rock bursts (Arjang and Herget 1997), and it is also an essential condition for numerical analysis of geotechnical engineering stability. A proper investigation for the initial ground stress is a premise of conducting a successful numerical analysis of geotechnical stability for three reasons. First, the analysis method for geotechnical engineering is mainly incremental analysis on account of characteristics of geotechnical engineering. * Derek B. Apel [email protected] 1
State Key Laboratory of Coal Mine Disaster Dynamics and Control, Chongqing University, Chongqing, China
2
School of Mining and Petroleum Engineering, University of Alberta, Edmonton, Canada
3
Central Mining Institute (GIG), Katowice, Poland
13
Vol.:(0123456789)
Natural Hazards
The stresses of each analysis step are equal to stresses in previous step plus stress increment, which means initial ground stress is a starting condition for incremental analysis (Yoon et al. 2016). Second, material characteristics of rock mass are closely related to its stress state. Moreover, for some dynamic geotechnical engineering such as excavations and backfilling, initial ground stress field is a precondition for simulating the construction process accurately (Di Donna and Laloui 2015). Considering the cost and construction difficulty, it is infeasible to conduct massive field measurements
Data Loading...
