Drilling Monitoring Dataset: 120-Hour Multi-Modal Sensor Recordings with Verified Fault Events

Read full text Summarise with VIRAT
Authors: guoye zhang
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This dataset contains 120 hours of continuous drilling monitoring data from three drill rigs (machine_A, machine_B, machine_C) in a sandstone mine. The dataset comprises multi-modal sensor recordings including acoustic data (96 kHz, 16-bit) and process data (RPM, Torque, WOB, ROP at 10 Hz). Key features: - 153 segments across 3 drill rigs - 24 maintenance-verified fault events (bit breakage, bearing fault, stick-slip, excessive wear) - Complete fault evolution sequences: healthy → progressive wear → early anomaly → severe anomaly - 48.49-minute curated subset with detailed annotations - All events confirmed by maintenance records with work orders File format: - Acoustic: WAV (96 kHz, 16-bit, mono) - Process: CSV (10 Hz) - Metadata: JSON Total size: 487 MB (compressed)

Fields of Study & Subjects

Machine Fault Diagnosis Techniques Drilling and Well Engineering Tunneling and Rock Mechanics

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
28 March 2026
Language
EN