Quantitative risk assessment for deep tunnel failure based on clustering and logistic regression at the Ashele copper mine, China

ABSTRACT

Tunnel failure caused due to geologic heterogeneity and variability in both mining processes and tunnel arrangement in deep metal mines is very complex. In this paper, a quantitative risk assessment for deep tunnel failure was conducted using clustering and logistic regression. The study was performed by considering various representative evaluation indexes of geological conditions, mining disturbance, and microseismic (MS) data. Comprehensive Seismic Event Clustering (CSEC) methodology was used to analyse the MS data to filter noise data and identify seismic sources. Additionally, peak particle velocity (PPV) was used to quantitatively evaluate the influence of MS activities on the response of tunnel which would aid in determining the PPV distribution characteristics of a large regional tunnel. Furthermore, Excavation Vulnerability Potential (EVP) index was used to quantitatively evaluate the effect of site conditions on response of the tunnel. Thus, based on logistic regression, PPV and EVP were used to establish a risk assessment model to determine the tunnel damage levels and probability of damage occurrence. The successful application of the proposed method at the Ashele copper mine demonstrates that it is capable of conducting a risk assessment for tunnel failure in deep metal mines.

KEYWORDS:

• Deep metal mine
• tunnel failure
• microseismic (MS) monitoring
• logistic regression
• clustering
• peak particle velocity (PPV)

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Key Research and Development Program of China (2021YFC3001301), the Natural Science Foundation of Shandong Province, China (ZR2021ZD36) and the National Natural Science Foundation of China (52174142, 51974059).