Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Unsupervised machine learning for data-driven rock mass classification: addressing limitations in existing systems using drilling data

T. F. Hansen, A. Aarset

TL;DR

The ability of MWD data to form distinct rock mass clusters suggests substantial potential for future classification systems using this objective, data-driven methodology, minimising human bias.

Abstract

Rock mass classification systems are crucial for assessing stability and risk in underground construction globally and guiding support and excavation design. However, these systems, developed primarily in the 1970s, lack access to modern high-resolution data and advanced statistical techniques, limiting their effectiveness as decision-support systems. We outline these limitations and describe how a data-driven system, based on drilling data, can overcome them. Using statistical information extracted from thousands of MWD-data values in one-meter sections of a tunnel profile, acting as a signature of the rock mass, we demonstrate that well-defined clusters can form a foundational basis for various classification systems. Representation learning was used to reduce the dimensionality of 48-value vectors via a nonlinear manifold learning technique (UMAP) and linear principal component analysis (PCA) to enhance clustering. Unsupervised machine learning methods (HDBSCAN, Agglomerative Clustering, K-means) clustered the data, with hyperparameters optimised through multi-objective Bayesian optimisation. Domain knowledge improved clustering by adding extra features to core MWD-data clusters. We structured and correlated these clusters with physical rock properties, including rock type and quality, and analysed cumulative distributions of key MWD-parameters to determine if clusters meaningfully differentiate rock masses. The ability of MWD data to form distinct rock mass clusters suggests substantial potential for future classification systems using this objective, data-driven methodology, minimising human bias.

Unsupervised machine learning for data-driven rock mass classification: addressing limitations in existing systems using drilling data

TL;DR

The ability of MWD data to form distinct rock mass clusters suggests substantial potential for future classification systems using this objective, data-driven methodology, minimising human bias.

Abstract

Rock mass classification systems are crucial for assessing stability and risk in underground construction globally and guiding support and excavation design. However, these systems, developed primarily in the 1970s, lack access to modern high-resolution data and advanced statistical techniques, limiting their effectiveness as decision-support systems. We outline these limitations and describe how a data-driven system, based on drilling data, can overcome them. Using statistical information extracted from thousands of MWD-data values in one-meter sections of a tunnel profile, acting as a signature of the rock mass, we demonstrate that well-defined clusters can form a foundational basis for various classification systems. Representation learning was used to reduce the dimensionality of 48-value vectors via a nonlinear manifold learning technique (UMAP) and linear principal component analysis (PCA) to enhance clustering. Unsupervised machine learning methods (HDBSCAN, Agglomerative Clustering, K-means) clustered the data, with hyperparameters optimised through multi-objective Bayesian optimisation. Domain knowledge improved clustering by adding extra features to core MWD-data clusters. We structured and correlated these clusters with physical rock properties, including rock type and quality, and analysed cumulative distributions of key MWD-parameters to determine if clusters meaningfully differentiate rock masses. The ability of MWD data to form distinct rock mass clusters suggests substantial potential for future classification systems using this objective, data-driven methodology, minimising human bias.
Paper Structure (34 sections, 1 equation, 17 figures, 9 tables)

This paper contains 34 sections, 1 equation, 17 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Limitations in existing rock mass classification systems
  3. The need for data-driven rock mass classification systems
  4. Advances in data-driven classification outside the tunnel profile
  5. Unsupervised machine learning for clustering and pattern detection in spatial data
  6. Laying the groundwork for data-driven classification
  7. Dataset and feature combinations
  8. Methodology
  9. Dimension reduction to improve clustering
  10. Clustering approaches for rock mass data
  11. Representation learning: a pipeline of unsupervised dimension reduction and clustering
  12. Evaluating clustering results
  13. Evaluating established clustering scores
  14. Evaluation beyond standard clustering metrics
  15. Visual inspection of clustering
  16. ...and 19 more sections

Figures (17)

  • Figure 1: Collection process for MWD-data and extraction of statistical information. X represents the features, and Y represents the labels. The two labels, Q-class D and rock type Granite are examples of values.
  • Figure 2: Distribution of mean values for all MWD and geometric parameters in the dataset
  • Figure 3: Schematic illustration of UMAP, adapted from oide_protein_2022
  • Figure 4: Illustrating density function and clusters for HDBSCAN, adapted from campello2013density
  • Figure 5: Illustrating the optimal selection of clusters for a cluster tree, adapted from campello2013density
  • ...and 12 more figures