Introduction
Blasthole drilling is a preliminary step in surface mining, therefore, delays caused by bit failure could impose significant losses on the production. Tricone bit design is relatively complex (Figure 1). Cones are connected to the bit body by bearings elements. Progressive bit wear results in failure of rolling elements which is known as the dominant catastrophic failure of tricone bits and usually happens as locked or detached cones. Some research works are conducted on wear detection of tricone bits employing a variety of techniques including drilling performance measurement, study of load distribution, vibration analysis and image processing approaches (Cooper, Reference Cooper2002; Ghosh et al., Reference Ghosh, Schunnesson and Kumar2016; Naganawa, Reference Naganawa2012; Rafezi & Hassani, Reference Rafezi and Hassani2017; Reference Rafezi and Hassani2018; Saeidi et al., Reference Saeidi, Rostami, Ataei and Torabi2014; Sheppard & Lesage, Reference Sheppard and Lesage1988). This article discusses the effectiveness of ROP monitoring and image processing methods for the purpose of tricone wear monitoring based on real-world drilling data and feasibility restrictions.
Figure 1. Tricone bit installed on the rig.
Table 1. Operation life of four tricone bits in one mine site
Method
Blasthole drill rigs were instrumented with data acquisition units and accelerometers in two mines in Canada for measurement while drilling (MWD). Drilling vibration, control signals as well as the ROP were recorded during the complete lifecycle of tricone bits. For data labeling, bit wear condition was manually inspected, recorded, and digital images were acquired while operating in various geological formations.
Results
In this study capability of ROP assessment and imaging approaches for tricone bit wear monitoring and failure prediction were investigated. The ROP trend during the bit lifecycle, starting from a brand new bit until the bit is worn out and replaced, has a fluctuating trend (Figure 2). Consequently, as the bits become worn out, a decremental trend in the ROP is not necessarily identified in a dynamic subsurface condition. Direct assessment of bit condition through digital imaging provides an insight into the shape of the bit and the missing or broken teeth are detectable using this approach.
Figure 2. The rate of penetration (ROP) trend during the complete lifecycle of a bit.
Discussion
It was observed that the ROP trend is significantly influenced by geological variations and as presented in the table, the lifetime of tricone bits can significantly vary (in terms of hours of operation and length of drilling) depending on the working condition for the identical bit brand, model, and diameter in the same mine site. For the bits listed in the table, the bit lives are ranging from 21.8 hrs. to 68 hrs. and from 746.1 m to 2805.4 m. On the other hand, processing of digital images for condition monitoring of tricone bits has some essential drawbacks; infeasibility of real-time monitoring while drilling down the hole and inability to assess the bearings condition which leads to the dominant failure mode of tricone bits.
Conclusions
The effectiveness of ROP monitoring and vision-based assessments for automated tricone bit condition monitoring was investigated in fieldworks. It is concluded that the ROP is mainly determined by geological condition and extraction of bit wear status from ROP trend requires a precise knowledge subsurface condition. Analysis of tricone bit wear through visual techniques, could provide an insight about the teeth wear state, however, it is not able to cover the bearings condition. In comparison, bit vibration frequency signatures reveal information about bit rolling elements condition and have the possibility of real-time measurement without interruption of the operation. Therefore, drilling vibration analysis is a promising and practical approach for tricone wear monitoring and failure perdition in industrial applications for fully autonomous drilling.
Acknowledgement
The authors would like to thank Teck Resources and ArcelorMittal companies for enabling the mine fieldworks as well as McGill University Engine centre.
Funding
This work was supported by the Natural Sciences and Engineering Research Council of Canada NSERC (CRD grant# 461514, I2I grant# 516232).
Conflicts of Interest Declaration
Hamed Rafezi and Ferri Hassani declare none.
Data Availability
The data are presented in the manuscript.
Related article
Tricone bit health monitoring using wavelet packet decomposed vibration signal. Rafezi, H., & Hassani, F., https://doi.org/10.1109/codit.2018.8394922
Open access
Comments
Comments to the Author: This paper does not provide clear guidelines on how the research will proceed.
The author did not give detailed instructions on how to implement (or has implemented) a tool condition monitoring system.
The author introduced the term “rate of penetration (ROP)”, which corresponds with feed rate (m/min) in cutting theory. I didn’t see which type of signal (in the tool wear monitoring system) will be compared to feed rate to get information about tool wear.
What about the mechanical properties of the rocks which is used in experimental? Does it affect to the wear rate on the cutting edge of the tools…?