New method quickly identifies rock types during drilling using sound waves

Jim Crocker
10th November, 2025

New method quickly identifies rock types during drilling using sound waves

Side view of the drilling rig used in the study.

Image adapted from: Jiang et al. / CC BY (Source)

Key Findings

  • This study, conducted in China’s coal mines, developed a new method to identify rock types automatically during drilling using sound
  • Researchers linked acoustic pressure—sound created during drilling—to measurable rock properties like density and strength, creating a new identification algorithm
  • The algorithm accurately identified six common rock types with 47-71% accuracy, and transition zones between rock layers didn’t hinder its performance
Identifying the types of rock encountered during drilling is a fundamental challenge in coal mine exploration. Accurate identification, known as lithology identification, is crucial for understanding the underground environment and improving the efficiency of resource extraction. Traditional methods rely on analyzing core samples brought to the surface, which is time-consuming and doesn’t provide real-time information. A research team from the China Coal Research Institute, China Coal Technology & Engineering Group, China University of Mining & Technology (Beijing), and Henan Polytechnic University have developed a new method for automatic lithology identification while drilling[1]. The core of this study lies in establishing a connection between the sound generated during the drilling process – specifically, acoustic pressure – and the physical properties of the rocks being drilled. Previous research has highlighted the importance of analyzing drilling parameters to understand rock mechanics[2]. This new study builds on that foundation by focusing specifically on acoustic pressure as a key indicator of rock type. To begin, the researchers collected core samples from an active mine to create both single-rock-type and mixed-rock-type specimens. These samples were then used in a specially designed laboratory drilling system, allowing for controlled experiments and the collection of detailed data. This system enabled a detailed time-frequency analysis of the acoustic pressure generated during drilling. A key aspect of the work was developing a mathematical model to quantify the relationship between acoustic pressure and rock physics parameters – measurable characteristics like density, strength, and elasticity. Understanding this relationship is vital for translating the sound signal into information about the rock being drilled. The research revealed the underlying physical mechanisms connecting acoustic pressure to these properties. Based on this established relationship, the team created an algorithm capable of automatically identifying rock types based on the acoustic pressure signal. The algorithm was tested on six common rock types found in coal mines: sandy mudstone, coal, mudstone, shale, limestone, and granite. The results showed varying degrees of accuracy, ranging from 47% for sandy mudstone to 71% for granite. Interestingly, the study also investigated the impact of “perforated transition zones” – areas where the rock structure is disrupted – on the accuracy of the identification process. The results indicated that these zones did not significantly interfere with the algorithm’s ability to correctly identify the lithology. This is a significant finding, as such zones are common in mining environments. The ability to automatically identify rock types while drilling has significant implications for the industry. It allows for real-time geological mapping, which can improve the accuracy of reservoir prediction and guide drilling operations. Furthermore, it contributes to the broader goal of automating drilling exploration, reducing the need for manual analysis and improving efficiency. Earlier work has also focused on using borehole imaging to understand rock structure[3][4]. While this study doesn’t directly utilize borehole images, it complements that research by providing a method for identifying the rock types within those structures. For example, understanding the lithology identified through acoustic pressure can help interpret the features observed in borehole images, providing a more complete picture of the underground environment. The intelligent segmentation of structural planes in borehole images, as demonstrated in previous studies[4], can be further enhanced by integrating lithological information obtained through acoustic analysis.

AgricultureEnvironment

References

Main Study

1) Study on automatic lithology identification method while drilling based on acoustic pressure-rock physics parameters mapping

Published 7th November, 2025

https://doi.org/10.1371/journal.pone.0330037

Related Studies

2) An experimental investigation into the borehole drilling and strata characteristics.

https://doi.org/10.1371/journal.pone.0253663

3) Monitoring method for borehole images of changes in overlying rock structure during coal mining.

https://doi.org/10.1038/s41598-025-00513-5

4) The segmentation and intelligent recognition of structural surfaces in borehole images based on the U2-Net network.

https://doi.org/10.1371/journal.pone.0299471


Related Articles

An unhandled error has occurred. Reload 🗙