Percussive and rotary percussion drilling (RPD) has been deployed extensively in shallow drilling applications in mining operations to improve the rate of penetration (ROP) through hard rock, such as granite. However, for deep drilling, where depths exceed 3 km, through hard rocks, the limited understanding of rotary percussion drilling tools, fluid hammers, and drill bits, has been a major drawback. Field experience and past experimental work have shown a non-linear relationship between bit RPM, hammer frequency and rate of drilling and the incidence of damaging vibrations. Therefore, it is important to understand the coupling between rotary and axial motion of the hammer bits to extend bit life and improve the rate of penetration. We present a novel approach to compare and validate the effectiveness of different hammer bits and operation in RPD by introducing a metric we call the Bit Performance Index (BPI). The BPI uses the shape and layout of inserts on a bit face and the relationship between bit RPM and hammer frequency to evaluate the overlap between hammer blows, and thus drilling efficiency. The two main applications of the BPI are to compare and evaluate the configuration and layout of diamond inserts on hammer bits and then to obtain the optimal operating parameters during drilling with a given bit. The most important aspect of the BPI is that it uses only the drill bit geometry, bit RPM, and hammer frequency with ROP and delivered energy as the performance metric. The BPI is first validated using data collected with a laboratory-scale experiment and then validated with field data from shallow test wells drilled in Europe and North America. The reward function used in the current work is inspired by those used in machine learning contexts, such as Q-learning. Usually, reward functions either reward or penalize an action depending on the desired outcome, and here the reward function considers the overlap between impact locations of the individual inserts on a hammer bit face. In this work, the reward function is based on the impact location of the current hit with respect to a previous impact. If the current impact is on the edge of any previous impact, it would lead to greater rock destruction due to higher stress load and thus is rewarded. However, if it is on the same location as any previous impact, it would lead to less rock destruction due to lower axial force transmission and lower stress and thus penalized. The reward function is combined with the geometry of the hammer bit and yields an overall reward for a particular rpm and hammer frequency. This approach is validated with a single in-sert experiment to evaluate rock failure as a function of impact overlap. An experimental rig has been developed which applies an identical amount of energy to every impact. Diamond percussion insert(s) with different shapes are used to impact granite and impact size, force, and coefficient of restitution are measured. The BPI is also compared with field data showing preliminary agreement. The proposed metric will provide insights into the drilling performance, especially the bit wear, and help in obtaining optimal combinations of the hammer frequency and the bit RPM.

Press the Back button in your browser, or search again.

Copyright 2022, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.