Unsupervised machine learning for data-driven classification of rock mass using drilling data: How can a data-driven system handle limitations in existing rock mass classification systems?
- URL: http://arxiv.org/abs/2405.02631v1
- Date: Sat, 4 May 2024 10:54:07 GMT
- Title: Unsupervised machine learning for data-driven classification of rock mass using drilling data: How can a data-driven system handle limitations in existing rock mass classification systems?
- Authors: T. F. Hansen, A. Aarset,
- Abstract summary: Rock mass classification systems are crucial for assessing stability and risk in underground construction globally.
Rock mass classification systems developed in the 1970s lack access to modern high-resolution data and advanced statistical techniques.
We show that it is possible to form well-defined clusters that can act as a foundational basis for various rock mass classification systems.
- Score: 0.0
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Rock mass classification systems are crucial for assessing stability and risk in underground construction globally and guiding support and excavation design. However, systems developed primarily in the 1970s lack access to modern high-resolution data and advanced statistical techniques, limiting their effectiveness as decision-support systems. Initially, we outline the limitations observed in this context and later describe how a data-driven system, based on drilling data as detailed in this study, can overcome these limitations. Using extracted statistical information from thousands of MWD-data values in one-meter sections of a full tunnel profile, thus working as a signature of the rock mass, we have demonstrated that it is possible to form well-defined clusters that can act as a foundational basis for various rock mass classification systems. We reduced the dimensionality of 48-value vectors using nonlinear manifold learning techniques (UMAP) and linear principal component analysis (PCA) to enhance clustering. Unsupervised machine learning methods (HDBSCAN, Agglomerative Clustering, K-means) were employed to cluster the data, with hyperparameters optimised through multi-objective Bayesian optimisation for effective clustering. Using domain knowledge, we experienced improved clustering and system tuning opportunities in adding extra features to core clusters of MWD-data. We structured and correlated these clusters with physical rock mass properties, including labels of rock type and rock quality, and analysed cumulative distributions of key MWD-parameters for rock mass assessment 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 grounded in this objective, data-driven methodology, free from human bias.
Related papers
- Fuzzy K-Means Clustering without Cluster Centroids [79.19713746387337]
Fuzzy K-Means clustering is a critical computation technique in unsupervised data analysis.
This paper proposes a novel Fuzzy K-Means clustering algorithm that entirely eliminates the reliance on cluster centroids.
arXiv Detail & Related papers (2024-04-07T12:25:03Z) - A comparative study on machine learning approaches for rock mass classification using drilling data [0.3749861135832073]
Current rock engineering design in drill and blast tunnelling relies on engineers' observational assessments.
Measure While Drilling (MWD) data, a high-resolution sensor dataset collected during tunnel excavation, is underutilised.
This study aims to automate the translation of MWD data into actionable metrics for rock engineering.
arXiv Detail & Related papers (2024-03-15T15:37:19Z) - Unsupervised Learning for Fault Detection of HVAC Systems: An OPTICS
-based Approach for Terminal Air Handling Units [1.0878040851638]
This study introduces an unsupervised learning strategy to detect faults in terminal air handling units and their associated systems.
The methodology involves pre-processing historical sensor data using Principal Component Analysis to streamline dimensions.
Results showed that OPTICS consistently surpassed k-means in accuracy across seasons.
arXiv Detail & Related papers (2023-12-18T18:08:54Z) - Robust and Automatic Data Clustering: Dirichlet Process meets
Median-of-Means [18.3248037914529]
We present an efficient and automatic clustering technique by integrating the principles of model-based and centroid-based methodologies.
Statistical guarantees on the upper bound of clustering error suggest the advantages of our proposed method over existing state-of-the-art clustering algorithms.
arXiv Detail & Related papers (2023-11-26T19:01:15Z) - Hard Regularization to Prevent Deep Online Clustering Collapse without
Data Augmentation [65.268245109828]
Online deep clustering refers to the joint use of a feature extraction network and a clustering model to assign cluster labels to each new data point or batch as it is processed.
While faster and more versatile than offline methods, online clustering can easily reach the collapsed solution where the encoder maps all inputs to the same point and all are put into a single cluster.
We propose a method that does not require data augmentation, and that, differently from existing methods, regularizes the hard assignments.
arXiv Detail & Related papers (2023-03-29T08:23:26Z) - Clustering individuals based on multivariate EMA time-series data [2.0824228840987447]
Ecological Momentary Assessment (EMA) methodological advancements have offered new opportunities to collect time-intensive, repeated and intra-individual measurements.
Advanced machine learning (ML) methods are needed to understand data characteristics and uncover meaningful relationships regarding the underlying complex psychological processes.
arXiv Detail & Related papers (2022-12-02T13:33:36Z) - Deep Clustering: A Comprehensive Survey [53.387957674512585]
Clustering analysis plays an indispensable role in machine learning and data mining.
Deep clustering, which can learn clustering-friendly representations using deep neural networks, has been broadly applied in a wide range of clustering tasks.
Existing surveys for deep clustering mainly focus on the single-view fields and the network architectures, ignoring the complex application scenarios of clustering.
arXiv Detail & Related papers (2022-10-09T02:31:32Z) - Detection and Evaluation of Clusters within Sequential Data [58.720142291102135]
Clustering algorithms for Block Markov Chains possess theoretical optimality guarantees.
In particular, our sequential data is derived from human DNA, written text, animal movement data and financial markets.
It is found that the Block Markov Chain model assumption can indeed produce meaningful insights in exploratory data analyses.
arXiv Detail & Related papers (2022-10-04T15:22:39Z) - Enhancing cluster analysis via topological manifold learning [0.3823356975862006]
We show that inferring the topological structure of a dataset before clustering can considerably enhance cluster detection.
We combine manifold learning method UMAP for inferring the topological structure with density-based clustering method DBSCAN.
arXiv Detail & Related papers (2022-07-01T15:53:39Z) - Scalable Hierarchical Agglomerative Clustering [65.66407726145619]
Existing scalable hierarchical clustering methods sacrifice quality for speed.
We present a scalable, agglomerative method for hierarchical clustering that does not sacrifice quality and scales to billions of data points.
arXiv Detail & Related papers (2020-10-22T15:58:35Z) - Progressive Cluster Purification for Unsupervised Feature Learning [48.87365358296371]
In unsupervised feature learning, sample specificity based methods ignore the inter-class information.
We propose a novel clustering based method, which excludes class inconsistent samples during progressive cluster formation.
Our approach, referred to as Progressive Cluster Purification (PCP), implements progressive clustering by gradually reducing the number of clusters during training.
arXiv Detail & Related papers (2020-07-06T08:11:03Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.