Prediction Of Penetration Rate Research Articles

Hybrid Model of Machine Learning Method and Empirical Method for Rate of Penetration Prediction Based on Data Similarity

The rate of penetration (ROP) is an important indicator affecting the drilling cost and drilling performance. Accurate prediction of the ROP has important guiding significance for increasing the drilling speed and reducing costs. Recently, numerous studies have shown that machine learning techniques are an effective means to accurately predict the ROP. However, in petroleum engineering applications, its robustness and generalization cannot be guaranteed. The traditional empirical model has good robustness and generalization ability. Based on the quantification of data similarity, this paper establishes a hybrid model combining a machine learning method and an empirical method, which combines the high prediction accuracy of the machine learning method with the good robustness and generalization of the empirical method, overcoming the shortcomings of any single model. The AE-ED (the Euclidean Distance between the input data and reconstructed data from the autoencoder model) is defined to measure the data similarity, and according to the data similarity of each new piece of input data, the hybrid model chooses the corresponding single model to calculate. The results show that the hybrid model is better than any single model, and all the evaluation indicators perform better, making it more suitable for the ROP prediction in this field.

A Novel Rate of Penetration Model Based on Support Vector Regression and Modified Bat Algorithm

In the geological drilling process, predicting the rate of penetration (ROP) is significantly important for improving drilling efficiency and reducing non-drilling time. However, due to the drilling data pollution and the complex nonlinearity in the geological drilling process, a reliable and highly accurate ROP prediction model is not easy to construct, and the model accuracy is affected by the value of model hyperparameters. In order to overcome the difficulties in modeling, a novel ROP model is developed to handle abnormal data and achieve nonlinear fitting. First, a local outlier factor is introduced to automatically detect the abnormal data, and then replace it with the nearest normal data. Then, the support vector regression (SVR) method is applied to construct nonlinear prediction model for ROP, and a modified bat algorithm (MBA) is developed to solve the non-convex problem in determining optimal value of hyperparameters for SVR-based ROP model. The MBA has six modifications to improve the global search ability, which achieves better performance in global search ability compared with other nine algorithms based on the experiments of IEEE 2005 benchmark functions. The developed ROP prediction model that combines SVR and MBA methods is tested based on actual drilling data and a semi-physical system, and the simulation and application results show that the developed modeling method has higher ROP prediction accuracy compared with the other modeling methods.

Evaluation of the wellbore drillability while horizontally drilling sandstone formations using combined regression analysis and machine learning models

The rate of penetration (ROP) is an influential parameter in the optimization of oil well drilling because it has a huge impact on the total drilling cost. This study aims to optimize four machine learning models for real-time evaluation of the ROP based on drilling parameters during horizontal drilling of sandstone formations. Two well data sets were implemented for the model training–testing (Well-X) and validation (Well-Y). A total of 1224 and 524 datasets were implemented for training and testing the model, respectively. A correlation for ROP assessment was suggested based on the optimized artificial neural network (ANN) model. The precision of this equation and the optimized models were tested (524 datapoints) and validated (2213 datapoints), and their accuracy was compared to available ROP correlations. The developed ANN-based equation predicted the ROP with average absolute percentage errors (AAPE) of 0.3% and 1.0% for the testing and validation data, respectively. The new empirical equation and the optimized fuzzy logic and functional neural network models outperformed the available correlations in assessing the ROP. The support vector regression accuracy performance showed AAPE of 26.5%, and the correlation coefficient for the estimated ROP was 0.50 for the validation phase. The outcomes of this work could help in modeling the ROP prediction in real time during the drilling process.

Unsupervised Adversarial Domain Adaptation Regression for Rate of Penetration Prediction

Summary The rate of penetration (ROP) refers to the speed at which a drill bit breaks through rock and deepens the drill hole. ROP is of great significance for drilling optimization and drilling cost savings. In real-world settings, the ROP data available for learning and training in a new oil field are scarce or even completely missing. In this paper, we propose a novel unsupervised multisource domain adaptation (MSDA) regression method for ROP that considers transferring the knowledge learned from a well-labelled source domain to the target domain with few labeled ROP data. First, we build a multisource unsupervised domain adaptation framework based on adversarial learning (WD-MUDA) which uses a weighted combination of multiple source domains to realize the fine-grained alignment of different data distributions. Specifically, we define a new similarity metric for different domains based on the Wasserstein distance. Furthermore, considering the uneven distribution of real drilling data samples, a novel regression loss is introduced to minimize the gradient discrepancy between multisource and target samples and improve the prediction accuracy of target samples. Extensive experiments on real drilling data sets reveal that the proposed method is effective and outperforms the state-of-the-art domain adaptation methods for ROP prediction.

Machine Learning Application in Enhancing Drilling Performance

Drilling Oil and Gas wells is an expensive operation, where the cost for drilling a single shallow offshore HP-UHT well may exceed 30 million USD easily. Due to the huge cost involved, companies are eager to accomplish the drilling operations in a minimal time frame, by increasing the drilling rate or rate of penetration (ROP) and reducing the downhole tool failures. The Mechanical Specific Energy (MSE) concept addresses these two drilling performance measures in totality as it provides maximum ROP which can be achieved with the existing drilling system (i.e., Drill Bits, Drilling Tubular, Well Profile, and Mud weight) without causing downhole tool failures. ROP is a function of Weight on Bit, Rotation per Minute, and Flow rate for an existing drilling system, optimization of these parameters results in increased ROP, whereas unharmonious values lead to shock and vibration in drilling assembly. Machine Learning algorithms facilitate this optimization of drilling parameters for enhancing drilling performance, by the development of an accurate ROP prediction and optimization model. The ROP prediction model was developed using Artificial Neural Network and Random Forest, and the optimization model was developed using evolutionary optimization algorithms like particle swarm optimization and genetic algorithms. The result of the optimization model depends upon the accuracy of the ROP prediction model and the upper and lower limits of drilling parameters defined in the optimization model. Machine learning applications in improving the Drilling Performance can be easily quantified in terms of saving days, where one day of operational cost for the company is approximately 0.2 million USD per Offshore – HP-UHT well.

An Advanced Long Short-Term Memory (LSTM) Neural Network Method for Predicting Rate of Penetration (ROP).

Rate of penetration (ROP) is an essential factor in drilling optimization and reducing the drilling cycle. Most of the traditional ROP prediction methods are based on building physical model and single intelligent algorithms, and the efficiency and accuracy of these prediction methods are very low. With the development of artificial intelligence, high-performance algorithms make reliable prediction possible from the data perspective. To improve ROP prediction efficiency and accuracy, this paper presents a method based on particle swarm algorithm for optimization of long short-term memory (LSTM) neural networks. In this paper, we consider the Tuha Shengbei block oilfield as an example. First, the Pearson correlation coefficient is used to measure the correlation between the characteristics and eight parameters are screened out, namely, the depth of the well, gamma, formation density, pore pressure, well diameter, drilling time, displacement, and drilling fluid density. Second, the PSO algorithm is employed to optimize the super-parameters in the construction of the LSTM model to the predict ROP. Third, we assessed model performance using the determination coefficient (R 2), root mean square error (RMSE), and mean absolute percentage error (MAPE). The evaluation results show that the optimized LSTM model achieves an R 2 of 0.978 and RMSE and MAPE are 0.287 and 12.862, respectively, hence overperforming the existing methods. The average accuracy of the optimized LSTM model is also improved by 44.2%, indicating that the prediction accuracy of the optimized model is higher. This proposed method can help to drill engineers and decision makers to better plan the drilling operation scheme and reduce the drilling cycle.

Prediction of rate of penetration in directional drilling using data mining techniques

Rate of penetration (ROP) represents drilling speed and its productive time during drilling operations in oil and gas wells. A predictive model that links ROP to its influential parameters is essential to optimize ROP for minimizing drilling costs. This study implements a comprehensive data mining approach utilizing Python toolboxes to improve ROP prediction in directional wells, which has not been addressed as much as vertical wells with respect to the downhole weight on the bit (WOB) and cutting transport. To do so, seven functions, including influential parameters, were identified to impact ROP in directional drilling. Drilling data of seven directional wells from an offshore rig in a gas field was compiled to set up the input dataset. The data preprocessing methods, consisting of the modified Z-score and Savitzky–Golay (SG) smoothing filter, were utilized to remove outliers and reduce the noises in the input dataset. Multilayer perceptron (MLP) neural network and random forest regression models were employed comparatively to predict ROP, and their architectures were designed by tuning hyperparameters of the models. The models' accuracy was statistically and graphically assessed by using the K-fold cross-validation and statistical metrics. The random forest model was demonstrated to be superior to the MLP neural network model in terms of accuracy and speed. The results represent that using calculated downhole WOB instead of measured surface WOB in the input dataset reinforces the models’ accuracy in the prediction of ROP. Statistical investigations such as partial correlation, mutual information, and permutation feature importance revealed that the cutting transport function can affect ROP in directional drilling as significantly as other influential parameters, which has not been mainly accounted for in the literature when establishing models for ROP prediction.

Drilling Parameters Optimization for Horizontal Wells Based on a Multiobjective Genetic Algorithm to Improve the Rate of Penetration and Reduce Drill String Drag

With the development of China’s oil and gas exploration and development to complex oil and gas fields, the drilling efficiency and safety of complex formations with large hardness and strong abrasiveness have become increasingly significant. Optimizing drilling parameters is an effective means to increase the rate of penetration (ROP) and improve drilling efficiency. However, traditional drilling parameter optimization methods with only a single objective of increasing the ROP lack consideration of the drill string’s drag which may also be increased when drilling parameters change. When drilling a horizontal well, increased drag can reduce drilling efficiency. Aiming at this problem, this paper uses the logging data of the oil field as the data source, establishes an intelligent ROP prediction model through the random forest algorithm, and calculates the string drag using the “hard-string” model. Finally, the nondominant sorting genetic algorithm-II (NSGA-II), which is a domination-based multiobjective optimization algorithm, is used to optimize the drilling parameters to increase the ROP and reduce the drag at the same time. The optimized drilling parameters guide the drilling operations. We used the proposed method to optimize the parameters during the drilling of a new horizontal well. The results show that the ROP of the horizontal section of the new well increases by 10.3%, and the drag reduces by 4.5% on average compared with the adjacent well.

Editorial: Special Issue on Smart and Green Infrastructures with Optimum Life Cycle Solutions

Editorial: Special Issue on Smart and Green Infrastructures with Optimum Life Cycle Solutions

TBM penetration rate prediction ensemble model based on full-scale linear cutting test

Accurate and effective prediction of TBM penetration rate (PR) is significant for ensuring TBM tunneling safety, improving tunneling efficiency and optimizing construction cost. Establishing a prediction model is usually an effective method to solve this critical problem. The type of TBM penetration rate prediction models can be divided into two categories: theoretical models and empirical models. Each model has its advantages, but there are also shortcomings. To make the advantages of the two models complement each other and further improve the accuracy and application scope of the prediction model, this paper proposes a prediction ensemble model of PR based on a full-scale linear cutting test and machine learning algorithm. A semi-theoretical TBM penetration rate prediction model is established through the full-scale linear cutting test and the rock-breaking analysis of granite and sandstone with different strengths. Based on the XGBoost intelligent algorithm, an empirical model of the penetration rate prediction is also established. On this basis, the Bagging algorithm combines these two models to form an ensemble prediction model of TBM penetration rate. We collected rock mass and operational parameters in the Hangzhou Second Water Source Water Transfer Channel Project to verify the semi-theoretical, empirical, and ensemble models, respectively. The results show that the three models' prediction errors are all less than 15%, and the ensemble model has a smaller error of 10.2%. This result confirms that the ensemble model can effectively improve prediction accuracy and has good engineering applicability.

Towards drilling rate of penetration prediction: Bayesian neural networks for uncertainty quantification

Drilling rate of penetration (ROP) prediction has long been a part of any drilling activity. By accurate prediction of ROP, optimization can be done that maximizes ROP and reduces drilling costs. ROP prediction relies on good quality data from nearby wells and is mostly performed through multiple-regression models. Recently, machine learning and artificial intelligence tools have been suggested as an alternative. On the other hand, the lack of knowledge about the uncertainty in these data-driven models can limit their applicability. A trained machine learning model can perform satisfactorily on the data on which it was trained, yet fail when applied to new data. In this work, we explore the application of Bayesian neural networks and the notion of uncertainty in ROP prediction. We compare the prediction of several deterministic and probabilistic models on a real drilling dataset. Despite accurately predicting the ROP in a 5000ft section of a horizontal well, the models are shown to have large uncertainties, potentially caused by the size of the training dataset. The results and analysis show how the adoption of a probabilistic framework can be leveraged to pinpoint the root cause of an ROP model’s lack of accuracy.

Rate of Penetration Prediction Method for Ultra-Deep Wells Based on LSTM–FNN

The drilling process is complex, especially for ultra-deep wells, which face the problems of high temperature, high pressure and poor drilling resistance in their formation. In order to establish an ROP (the Rate of Penetration) prediction model for ultra-deep wells, the characteristics of ultra-deep well drilling operations, such as formation temperature and formation pressure, are fully considered in the process of parameter optimization. Combined with the drilling mechanism and mutual information correlation coefficient, the final input parameters are determined. The powerful nonlinear fitting ability of the artificial intelligence method is very suitable for predicting the ROP. Considering the time sequence of multi-source data, this paper combines the powerful timing information-based mining ability of the LSTM (Long Short-Term Memory Neural Network) with the nonlinear fitting ability of FNN (Fully Connected Neural Network), and establishes an intelligent prediction model of the ROP based on a LSTM–FNN. The results show that the average relative error and R2 of the LSTM–FNN model on the data of well 1 and well 2 are better than the FNN and LSTM models. In addition, the accuracy of the LSTM–FNN model on the data of adjacent wells is reduced by only 5%, which further verifies the good mobility of the model.

Research on Prediction of TBM Performance of Deep-Buried Tunnel Based on Machine Learning

Based on the relevant data in the construction process of the south of the Qinling tunnel of the Hanjiang-to-Weihe River Diversion Project, this article obtains the main influencing factors of the tunnel boring machine (TBM) performance of the deep-buried tunnel. According to the characteristics of deep-buried tunnel excavation, the random forest algorithm is used to select the features of the factors affecting the TBM penetration rate, and the four factors with large influence weights including total thrust, revolutions per minute, uniaxial compressive strength and volumetric joint count, are used as TBM penetration rate prediction models input parameters, which can improve the prediction accuracy and convergence speed of the model, and enhance the engineering practicality of the prediction model. Three types of TBM penetration rate prediction models are established: multiple regression model (MR), back propagation neural network model (BPNN) and support vector regression model (SVR). The prediction accuracy of the three models is compared and analyzed. The BPNN prediction model exhibits better prediction performance and generalization ability than the multiple regression model and SVR model, which manifest higher prediction accuracy and prediction stability.

Introducing Tree-Based-Regression Models for Prediction of Hard Rock TBM Performance with Consideration of Rock Type

Prediction of machine performance is a fundamental step for planning, cost estimation/control and selection of the machine type when using a tunnel boring machine (TBM). Penetration rate (PR) and machine utilization (U) are the two principal measures of TBM performance for evaluating the feasibility of using a machine in a given ground condition. However, despite the widespread use of TBMs and established track records, accurate estimation of machine performance could still be a challenge, particularly in complex geological conditions. Since different types of rocks have varied texture (cementation and grain size), and respond differently to cutting forces in the TBM tunnelling, incorporating the effects of rock type in performance prediction models can improve the accuracy of the estimates. The aim of this study was to develop models for predicting penetration rate of hard rock TBMs in different types of rock based on field penetration index (FPI), using multivariable regression analysis and machine learning algorithm, including classification and regression tree (CART). The proposed models offer estimated FPIs in different rock types, rock strength, and rock mass properties in the form of graphs (diagrams), which can be used to estimate TBM penetration rate. The proposed models have been developed based on the analysis of a comprehensive database of TBM performance in various rock types and offers more accurate estimates of machine performance by incorporating many of the key parameters available in typical geotechnical reports and contract documents. The models also exhibit sensitivity to rock mass parameters for predicting the penetration rate.

ROP Prediction with Supervised Machine Learning; a Case Study

Optimum drilling penetration rate, known as the rate of penetration (ROP) has played a big role in drilling operations. Planning the well ROP always becomes a challenge for drilling engineers to calculate the drilling time needed for the section. Optimum ROP is achieved when the time to drill the section is as planned.
 Many empirical approaches were develop to model the ROP based on the drilling parameters, and might not always match the actual ROP. In some cases, the actual ROP was slower than planned, which may increase the drilling cost, which needs to be avoided.
 Hence, some approaches using artificial intelligent (AI), and supervised machine learning have been develop to overcome it. Supervised machine learning is used to develop a ROP model and ROP prediction for one of the development fields, based only on two wells drilling parameters data. The model was train using Gradient Boosting, Random Forest, and Support Vector Machine. Drilling parameter test data then is used to validate the model. The model of Random Forest shows a good or promising result with R2 of 0.90, Gradient Boosting shows R2 of 0.86, and Support Vector Machine with R2 0.72. Based on the models generated, the Random Forest has shown good trend which could be used for modeling ROP in the future development wells.

Deep Neural Network Prediction of Mechanical Drilling Speed

Rate of penetration (ROP) prediction is critical for the optimization of drilling parameters and ROP improvement during drilling. However, it is still challenging to accurately predict ROP based on traditional empirical formula methods. This is usually the case for the development of the Wushi 17-2 oilfield block in the South China Sea. The Liushagang Formation is complex and the ROP is relatively low and difficult to increase. Ordinary data-driven ROP prediction models are not applicable because they do not take into account the complexity of formation conditions. In this work, we characterize the formation with acoustic transit time and build a data-driven ROP prediction model based on a deep neural network approach. By using the exploratory well data of the Wushi 17-2 oilfield for training and testing, the matching degree of the established model with the real data can reach 82%. In addition, we have developed a drilling parameter optimization process based on the ROP prediction model to improve ROP. Through on-site simulation, we found that the process can well meet the construction requirements. The established models and process flow are also applicable to the development of other formations and fields.

Modeling and Simulation of Rock Bits Based on Mega Drilling Data

Rock bits directly undertake the task of fracturing rock under earth during drilling operation. Accurate prediction of Rate of penetration (ROP) play an important role in reducing drilling costs and shortening the drilling cycle. This paper introduces ROP prediction methods including empirical mathematical modeling, computer simulation of bit-rock interaction and BP neural network model based on mega drilling data. The most significant factors affecting ROP are summarized as follows: weight on bit(WOB), rotation speed, rock formation properties, bit cutting structure, etc., which have a complex multi-parameter nonlinear relationship with ROP. The drilling data features complexity and big volume. Therefore machine learning is a best way to model the relationship between main factors and ROP. The advantages and disadvantages of the different ROP modeling methods are analyzed and compared. It shows that the BP neural network model based on mega drilling data demonstrates high accuracy, and is the future development direction in ROP prediction.

Half a century experience in rate of penetration management: Application of machine learning methods and optimization algorithms - A review

Rate of penetration (ROP) management is a matter of importance in drilling operations and it has been considered in different studies. Different machine learning methods such as simple and optimized artificial neural networks (ANNs), support vector machines (SVMs), fuzzy logic (FL) or adaptive neuro-fuzzy inference system (ANFIS), ensemble methods of machine learning and meta-heuristic algorithms have been used for this purpose so far. In this article, some of the studies by using these methods as the main approach for ROP management are reviewed to achieve a better understanding of this concept, its economic benefits and also its research capacities. Results indicate that ANNs are the most popular machine learning method in ROP management, while simple ANNs excel the modified types in this regard. Still, modified ANNs outperform simple ones in terms of prediction accuracy, but as ANNs fall short in superior prediction performance, other machine learning approaches of ROP management such as linear regression (LR), random forest (RF) and gradient boosting method (GBM) have compensated this shortcoming and proved their efficiency and applicability. • Applications of different machine learning methods and optimization algorithms for ROP management are reviewed. • Application of ANN, FL, ANFIS and SVM along with other data-driven methods are compared based on ROP prediction precision. • Ensemble methods of machine learning have found a wide verity of applications in ROP management projects and studies. • Application of meta-heuristic optimization methods are investigated based on the ROP increase from the base case. • In the literature, MLP-ANN and PSO algorithms have been used for ROP prediction and optimization more repeatedly.

Rate of penetration prediction while drilling vertical complex lithology using an ensemble learning model

The rate of penetration (ROP) accounts for a substantial portion of the overall drilling cost. The drilling optimization process, which mostly involves the adjustment of the mechanical drilling parameters, is therefore of prime importance in ensuring efficient drilling. However, drilling formations with assorted types of lithology necessitate the involvement of more parameters to reduce uncertainty and enhance confidence when predicting the ROP. The objective of this paper is to introduce an ensemble model based on random forest (RF), in which artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS) are the base learner models, to predict the ROP across different lithological formations In this study, two types of actual field data of Well-1 were employed to build the model: (i) mechanical drilling parameters collected from real-time sensors allocated at the rig site, and (ii) petrophysical properties obtained from conventional well logs. Well-2 with more than 2300 unseen data points was used to compare the capability of the base learners (i.e., ANN and ANFIS), standalone RF, and RF-meta model in predicting the ROP with two of the earliest published ROP empirical models (Maurer's and Bingham's models). The results showed that the RF-meta model outperformed the base learners and Maurer's and Bingham's empirical models in predicting the ROP in Well-2 with a low absolute average percentage error (AAPE) of 7.8 % and a high coefficient of determination (R2) of 0.94.

Rock mass characteristics model for TBM penetration rate prediction– an updated version

Rock mass characteristics (RMC) model is the one of serval prediction models for TBM penetration rate that considers the interaction between the rock mass and cutters. The first version was developed in 2009. However, it was a relative site specific model and the machine parameters were ignored. To update the RMC model, a comprehensive data base which included the results of the laboratory cutting tests and the in site penetration tests was compiled. Firstly, the correction factors of the cutter spacing and cutter shape were set up based on the laboratory tests, which were then used to normalize the data with the given machine parameters. Finally, two critical parameters in RMC model, i.e., the specific rock mass boreability index (SRMBI) and the exponent c, were estimated by regression analysis using the rock mass parameters. Now SRMBI in the updated model represents the boreability of a rock mass excavated by the given machine setup at penetration rate of 1 mm/rev. A higher SRMBI implies the rock mass is more difficult to be bored. The exponent c decides the increment in cutter normal force for the increasing penetration rate, c together with the rock mass strength reflect the different TBM cutting modes. Since the rock specimens in the laboratory tests was intact and the rock masses in the in site penetration tests contain the joints, the updated model reflects the reduction in the required thrust force caused by the joints in nature. The updated RMC model covered a wide range of the geological context and machine parameters. It could be used to predict TBM penetration rate and optimize the operating parameter in the given stroke. The equations in the updated model need further modification when more test data is collected, and the in situ stress needs to be taken into consideration.