Prediction and Analysis of PDC Bit Wear in Conglomerate Layer with Machine Learning and Finite-Element Method

Polycrystalline diamond compact (PDC) bits experience a serious wear problem in drilling tight gravel layers. To achieve efficient drilling and prolong the bit service life, a simplified model of a PDC bit with double cutting teeth was established by using finite-element numerical simulation technology, and the rock-breaking process of PDC bit cutting teeth was simulated using the Archard wear principle. The numerical simulation results of the wear loss of the PDC bit cutting teeth, such as the caster angle, temperature, linear velocity, and bit pressure, as well as previous experimental research results, were combined into a training dataset. Then, machine learning methods for equal-probability gene expression programming (EP-GEP) were used. Based on the accuracy of the training set, the effectiveness of this method in predicting the wear of PDC bits was demonstrated by verifying the dataset. Finally, a prediction dataset was established by a Latin hypercube experiment and finite-element numerical simulation. Through comparison with the EP-GEP prediction results, it was verified that the prediction accuracy of this method meets actual engineering needs. The results of the sensitivity analysis method for the gray correlation degree show that the degree of influence of bit wear is in the order of temperature, back dip angle of the PDC cutter, linear speed, and bit pressure. These results demonstrate that when an actual PDC bit is drilling hard strata such as a conglomerate layer, after the local high temperature is generated in the formation cut by the bit, appropriate cooling measures should be taken to increase the bit pressure and reduce the rotating speed appropriately. Doing so can effectively reduce the wear of the bit and prolong its service life. This study provides guidance for predicting the wear of a PDC bit when drilling in conglomerate, adjusting drilling parameters reasonably, and prolonging the service life of the bit.

Bi-Directional Long Short-Term Memory Variational Autoencoder for Real-Time Bit-Wear Estimation

Drilling operations rely on learned expertise in monitoring the drilling performance data and the rock data to assess the dull condition of the drill bit. While human learning can subjectively pick up the indicators based on rig surface data streams, this information is highly convoluted with changes in rock and drilling data. Recent approaches for bit wear estimation also include model-based and traditional supervised machine learning methods, which are usually costly and time-consuming. In this study, we developed a bi-directional long short-term memory-based variational autoencoder (biLSTM-VAE) to project raw drilling data into a latent space in which the real-time bit-wear can be estimated. The proposed deep neural network was trained in an unsupervised manner, and the bit-wear estimation is demonstrated as an end-to-end process.

Quantifying PDC Bit Wear in Real-Time and Establishing an Effective Bit Pull Criterion Using Surface Sensors

It is useful during drilling operations to know when bit failure has occurred because this knowledge can be used to improve drilling performance and provides guidance on when to pull out of hole. This paper presents a simple polycrystalline diamond compact (PDC) bit wear indicator and an associated methodology to help quantify wear and failure using real-time surface sensor data and PDC dull images. The wear indicator is used to identify the point of failure, after which corresponding surface data and dull images can be used to infer the cause of failure. It links rotary speed (RPM) with rate of penetration (ROP) and weight-on-bit (WOB). The term incorporating RPM and ROP represents a "sliding distance", i.e. the number of revolutions required to drill a unit distance of formation, while the WOB represents the formation hardness or contact pressure applied by the formation. This PDC bit wear metric was applied and validated on a data set comprised of 51 lateral production hole bit runs on 9 wells. Surface electric drilling recorder (EDR) data alongside bit dull photos were used to interpret the relationship between the wear metric and observed PDC wear. All runs were in the same extremely hard (estimated 35 – 50 kpsi unconfined compressive strength) and abrasive shale formation. Sliding drilling time and off-bottom time were filtered from the data, and the median wear metric value for each stand was calculated versus measured hole depth while in rotary mode. The initial point in time when the bit fails was found to be most often a singular event, after which ROP never recovered. Once damaged, subsequent catastrophic bit failure generally occurred within drilling 1-2 stands. The rapid bit failure observed was attributed to the increased thermal loads seen at the wear flat of the PDC cutter, which accelerate diamond degradation. The wear metric more accurately identifies the point in time (stand being drilled) of failure than the ROP value by itself. Review of post-run PDC photos show that the final recorded wear metric value can be related to the observed severity of the PDC damage. This information was used to determine a pull criterion to reduce pulling bits that are damaged beyond repair (DBR) and reduce time spent beyond the effective end of life. Pulling bits before DBR status is reached and replacing them increases overall drilling performance. The presented wear metric is simple and cost-effective to implement, which is important to lower-cost land wells, and requires only real-time surface sensor data. It enables a targeted approach to analyzing PDC bit wear, optimizing drilling performance and establishing effective bit pull criteria.

Bone drilling with internal gas cooling: Experimental and statistical investigation of the effect of cooling with CO2 on reduction of temperature rise due to drill bit wear

Bone drilling is a major stage in immobilization of the fracture site. During bone drilling operations, the temperature may exceed the allowable limit of 47 °C, causing irrecoverable damages of thermal necrosis and seriously threatening the fracture treatment. One of the parameters affecting the temperature rise of the drilling site is the frequency of applying the drill bit and its extent of wear. The present study attempted to mitigate the effect of drill bit wear on the bone temperature rise through the internal gas cooling method via CO2 and to reduce the risk of incidence of thermal necrosis. To this end, drilling tests were conducted at three rotational speeds 1000, 2000, and 3000 r·min-1 in two states of without cooling and with internal gas cooling by CO2 through an internal coolant carbide drill bit, along with six drill bit states (new, used 10, 20, 30, 40, and 50 times) on a bovine femur bone. The results indicated that in the internal gas cooling state, as the number of drill bit applications increased from the new state to more than 50 times, the temperature of the hole site increased on average by ΔT = 2-3 °C (n = 1000 r·min-1), ΔT = 5-8 °C (n = 2000 r·min-1), and ΔT = 5-7 °C (n = 3000 r·min-1). Furthermore, the internal gas cooling method was able to significantly reduce the effect of the drill bit wear on the temperature rise of the drilling site and to resolve the risk of incidence of thermal necrosis regardless of the process parameters for drilling operations.

Rethinking the evidence for early horse domestication at Botai

Despite its transformative impact on human history, the early domestication of the horse (Equus caballus) remains exceedingly difficult to trace in the archaeological record. In recent years, a scientific consensus emerged linking the Botai culture of northern Kazakhstan with the first domestication of horses, based on compelling but largely indirect archaeological evidence. A cornerstone of the archaeological case for domestication at Botai is damage to the dentition commonly linked with the use of bridle mouthpieces, or “bit wear.” Recent archaeogenetic analyses reveal, however, that horse remains from Botai are not modern domesticates but instead the Przewalski’s horse, E. przewalskii—warranting reevaluation of evidence for domestication. Here, we compare osteological traits hypothesized to have been caused by horse transport at Botai with wild Pleistocene equids in North America. Our results suggest that damage observed in Botai horse teeth is likely generated by natural disturbances in dental development and wear, rather than through contact with bridle equipment. In light of a careful reconsideration of the mid-Holocene archaeological record of northern Eurasia, we suggest that archaeological materials from Botai are most effectively explained through the regularized mass harvesting of wild Przewalski’s’ horses—meaning that the origins of horse domestication may lie elsewhere.

Real-Time Classification of Drilled Lithology from Drilling Data Using Deep Learning with Online Calibration

Decision making to optimize the drilling operation is based on a variety of factors, among them real-time interpretation of drilled lithology. Since logging while drilling (LWD) tools are placed some meters above the bit, mechanical drilling parameters are the earliest indicators, although difficult to interpret accurately. This paper presents a novel deep learning methodology using mechanical drilling parameters for lithology classification. A cascade of multilayered perceptrons (MLPs) are trained on historical data from wells on a field operated by Equinor. Rather than an end-to-end approach, the drilling parameters are utilized to estimate LWD sensor readings in an intermediate step using the first MLPs. This allows continuous updates of the models during operation using delayed LWD data. The second MLP takes the virtual LWD estimates as input to predict currently drilled lithology, similar to manual expert interpretation of logs. This configuration takes into account case dependent (mud, BHA, wellbore geometry) and time varying (bit-wear, wellbore friction) relationships between drilling parameters and LWD readings, while assuming a constant rule when utilizing LWD data to classify lithology. Upon completion of training and validation, the system is tested on a separate, unseen wellbore, for which results are presented. Visualizations for true lithology alongside the estimates are given, along with confusion matrices and model accuracy. The system achieves high accuracy on the test set and presents low confusion between classes, meaning that it distinguishes well between the lithologies present in the wellbore. It can be seen that the borders between successive layers of lithology are detected rapidly, which is crucial seen from an optimization standpoint, so the driller may adjust accordingly immediately. It shows promise as an advisory system, capable of accurately classifying currently drilled lithology by continuously adapting to changing downhole conditions. Although we cannot expect perfect estimates of lithology purely based on drilling parameters, we can obtain a preliminary map of the subsurface this way. This novel configuration gives a real-time interpretation of the currently drilled lithology, allowing the driller to take proper actions to optimize the drilling operation in terms of rate of penetration (ROP) and best practices for different lithologies.

Evaluating Data-Driven Techniques to Optimize Drilling on the Moon

Several companies and countries have announced plans to drill in the lunar South Pole in the next five years. The drilling process on the Moon or any other planetary body is similar to other exploration drilling by using rotary drills, for example the oil and gas drilling. However, the key performance indicators (KPIs) for this type of drilling are significantly different. This work aimed to develop the drilling optimization algorithms to optimize drilling on the Moon based on the experiences with the terrestrial drilling in related industries. A test drilling unit was designed and fabricated under a NASA Early Stage Innovation (ESI) grant; A high-frequency data acquisition system was used to record drilling responses at 1000 Hz. Parameters like weight on bit (WOB), torque, RPM, rate of penetration (ROP), mechanical specific energy (MSE), field penetration index (FPI), and the uniaxial compressive strength (UCS) were recorded for 40 boreholes in the analog formations. This work utilizes the large dataset comprising of more than 1 billion data points recorded while drilling into various lunar analogous formations and cryogenic lunar formations to optimize power consumption and bit wear during drilling operations. The dataset was processed to minimize the noise. The effect of drilling dysfunctions like auger choking and bit wear was also removed. Extensive feature engineering was performed to identify how each of the parameter affects power consumption and bit wear. The data was then used to train various regression algorithms based on the machine learning approaches like the random forest, gradient boosting, support vector machines, logistic regression, polynomial regression, and artificial neural network to evaluate the applicability of each of these approach in optimizing the power consumption using the control variables like RPM and penetration rate. The best performing algorithm based on ease of application, runtime, and accuracy of the algorithm was selected to provide recommendations for ROP and RPM which would result in minimum power consumption and bit wear for a specific bit design. Since the target location for most lunar expeditions is in permanently shadowed regions, the power available for a drilling operation is extremely limited. The bit wear will significantly affect the mission life too. Algorithms developed here would be vital in ensuring efficient and successful operations on the Moon leading to more robust exploration of the targeted lunar regions.