Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Cited by Google
- Similars in SciELO
- Similars in Google
Share
Tecnura
Print version ISSN 0123-921X
Abstract
FERRARI, Guido Claudio and GOMEZ, Martín Pedro. Characterization of drilling stages in multilayer specimens through the Acoustic Emission generated during the drilling process. Tecnura [online]. 2020, vol.24, n.63, pp.26-39. ISSN 0123-921X. https://doi.org/10.14483/22487638.16193.
Objective:
This work is based on the Acoustic Emission (EA) monitoring of the cutting tool in a geological material drilling process and the use of Kohonen type Neural Networks for the classification of information.
Methodology:
The methodology consisted in the realization of a series of perforations on a specimen built with sandstone, limestone and slate rocks, arranged in layers and consolidated with a cement mixture. The tool used for drilling consisted of a double-edged Tungsten Carbide (CT) cutter, 65 mm in diameter. The entire process was monitored by an AE system coupled to the rotating drill and test specimen. Subsequently, the EA was correlated with the stratigraphy of the perforation, the information was processed and adapted to train and simulate a Kohonen type neural network, which classified the process information according to the type of rock that was traversed with the cutter.
Results:
The results show that the Acoustic Emission technique is sensitive to stratum changes during the drilling process of these geomaterials and that the instrumentation of the rotating drill provides a good monitoring channel for this process. In this, the change of interfaces of the interfaces and the stable drilling process can be observed through the analysis of acoustic emission parameters such as rms, MARSE energy, rise time and average frequency.
Conclusions:
The Acoustic Emission technique can be used to monitor the drilling process on this scale. The processing of the Acoustic Emission parameters allowed to train and simulate a Kohonen neural network that can classify different stages of the drilling process with a mixing error of less than 5%.
Keywords : Acoustic Emission; Drilling; Process Monitoring; Sandstone; Neural Networks.