Search

Mud logging data, which quantify the energy expended in rock breaking during drilling, are routinely acquired during drilling. These data offer a practical means to estimate the geomechanical properties of rocks, particularly confined compressive strength (CCS), avoiding the need for costly and destructive laboratory tests. Thus, in this study, predictive models for CCS were developed utilizing mudlogging data and employing various machine learning (ML) algorithms, including multi-layer perceptron neural network (MLPNN), least squares support vector machine (LSSVM), Gaussian process regression (GPR), and random forest (RF). Data were compiled from reservoir sections of two vertical wells (Well A and Well B) of an oil field in southwest Iran. CCS values were estimated using petrophysical logs and empirical correlations, calibrated with laboratory test results, and organized into a database of input and output features. 80% of Well A data were allocated for training, with 20% reserved for testing. The Tukey method was then used to remove outliers from the training subset, ensuring a high-accuracy and generalizable model. Subsequently, tuned ML algorithms based on the problem condition were applied to develop predictive CCS models using the training data. These models were then rigorously assessed using the test data. The outcomes of these steps revealed that the GPR-based model exhibited the lowest Root Mean Square Error (RMSE, train: 0.8499 MPa and test: 6.3886 MPa) and the highest coefficient of determination (R2, train: 0.9993 and test: 0.9597) among developed ML models. Further scrutiny employing over-fitting index and regression error characteristic curve techniques underscored the high accuracy and the generalizability of the GPR model on unseen data. Application of the GPR model on Well B demonstrated remarkable accuracy (RMSE = 2.3107 MPa) and its real-time predictive capability from minimal mudlogging data variables, providing a reliable and cost-effective solution for CCS prediction in vertical wells. Highlights: 4 machine learning models predict confined compressive strength (CCS) from mud-log data 17 outliers detected with high mechanical specific energy values in CCS prediction Gaussian process regression (GPR) delivers the most precise CCS predictions Depth and weight on bit are the most and least influential input features on CCS Partial-dependence plots reveal key input influences on CCS predictions" [ABSTRACT FROM AUTHOR]

Mechanical brittleness index ( B I mech ) of the rock is a necessary parameter for selecting appropriate drilling bits and proper depth intervals for hydraulic fracturing. The B I mech measurement is possible through continuous coring followed by laboratory tests, which are extremely cost- and time-intensive. Although petrophysical logs provide us with some continuous pieces of information, the complex nonlinear relationship between the logs and the B I mech calls for implementing intelligent approaches before one can predict B I mech from the petrophysical logs. Therefore, the present research is an attempt to develop B I mech prediction models using least-squares support-vector machine (LSSVM) and multilayer perceptron (MLP) neural network (NN) as well as their hybrid forms with cuckoo optimization algorithm (COA), particle swarm optimization (PSO), and genetic algorithm (GA) on data from wells penetrating Maroon Oilfield. For this purpose, we began by approximating the B I mech from the Poisson's ratio and static Young's modulus – with the later obtained from laboratory test results-calibrated petrophysical logs. Next, the entire set of available data was split into two subsets, namely modeling and validation subsets. Application of the second version of the nondominated sorting genetic algorithm (NSGA-II) combined with the MLP-NN for feature selection showed that, among the eight features considered, five made the best set for developing estimator models, including P- and S-wave velocities, depth, density (RHOB), and gamma-ray (GR) readings. Accordingly, intelligent algorithms were developed by means of these five features on the basis of the modeling data. Results of the training and testing phases showed that the hybrid algorithms were more accurate than the simple forms of either MLP or LSSVM. Among the hybrid algorithms, the highest levels of accuracy and generalizability were achieved with the LSSVM-COA. Application of the developed models on the validation data and a well in the Azadegan oil field further confirmed the high accuracy and generalizability of this model for predicting the B I mech . Therefore, at a high level of confidence, the proposed model can be recommended for predicting B I mech at wells penetrating similar formations. [ABSTRACT FROM AUTHOR]

Permeability is an important parameter in the petrophysical study of a reservoir and serves as a key tool in the development of an oilfield. This is while its prediction, especially in carbonate reservoirs with their relatively lower levels of permeability compared to sandstone reservoirs, is a complicated task as it has larger contributions from heterogeneously distributed vugs and fractures. In this respect, the present research uses the data from two wells (well A for modeling and well B for assessing the generalizability of the developed models) drilled into a carbonate reservoir to estimate the permeability using composite formulations based on least square support vector machine (LSSVM) and multilayer extreme learning machine (MELM) coupled with the so-called cuckoo optimization algorithm (COA), particle swarm optimization (PSO), and genetic algorithm (GA). We further used simple forms of convolutional neural network (CNN) and LSSVM for the sake of comparison. To this end, firstly, the Tukey method was applied to identify and remove the outliers from modeling data. In the next step, the second version of the nondominated sorting genetic algorithm (NSGA-II) was applied to the training data (70% of the entire dataset, selected randomly) to select an optimal group of features that most affect the permeability. The results indicated that although including more input parameters in the modeling added to the resultant coefficient of determination (R2) while reducing the error successively, yet the slope of the latter reduction got much slow as the number of input parameters exceeded 4. In this respect, petrophysical logs of P-wave travel time, bulk density, neutron porosity, and formation resistivity were identified as the most effective parameters for estimating the permeability. Evaluation of the results of permeability modeling based on root-mean-square error (RMSE) and R2 shed light on the MELM-COA as the best-performing model in the training and testing stages, as indicated by (RMSE = 0.5600 mD, R2 = 0.9931) and (RMSE = 0.6019 mD, R2 = 0.9919), respectively. The generalizability assessment conducted on the prediction of permeability in well B confirmed the MELM-COA can provide reliable permeability predictions by achieving an RMSE of 0.9219 mD. Consequently, the mentioned methodology is strongly recommended for predicting the permeability with high accuracy in similar depth intervals at other wells in the same field should the required dataset be available. [ABSTRACT FROM AUTHOR]

Accurate and frequent monitoring of the solid content (SC) of drilling fluids is necessary to avoid the issues associated with improper solid particle concentrations. Conventional methods for determining SC, such as retort analysis, lack immediacy and are labor-intensive. This study applies machine learning (ML) techniques to develop SC predictive models using readily available data—Marsh funnel viscosity and fluid density. A dataset of 1290 data records was collected from 17 wells drilled in two oil fields located in southwest Iran. Four ML models—least squares support vector machine (LSSVM), multilayered perceptron neural network, extreme learning machine, and generalized regression neural network—were developed to predict SC from the compiled dataset. Multiple assessment techniques were applied to attentively evaluate the models’ prediction performances and select the best-performing, SC prediction model. The LSSVM model generated the least errors, exhibiting the lowest root-mean-square error values for the training (1.80%) and testing (1.84%) subsets. The narrowest confidence interval, 0.18, achieved by the LSSVM model confirmed its reliability for SC prediction. Leverage analysis revealed minimal influence of outlier data on the LSSVM model's SC prediction performance. The trained LSSVM model was further validated on unseen data from another well drilled in one of the studied oil fields, demonstrating the model’s generalizability for providing credible close-to-real-time SC predictions in the studied fields. [ABSTRACT FROM AUTHOR]

An important requirement of reservoir management is to understand the properties of reservoir fluids and dependent phase behaviors. This makes it possible to determine the properties of reservoir fluids in laboratory pressure-volume-temperature (PVT) tests. Such laboratory tests are costly and time consuming, and reservoir data are sometimes not available. When estimating the in-place fluid volumes and/or designing enhanced recovery processes information on bubble point pressure (BPP) and oil formation volume factor (OFVF) is required. Predicting these two parameters is therefore one of the priorities of reservoir engineers. It is becoming increasingly beneficial to be able to predict BPP and OFVF with efficient machine-learning algorithms based on underlying variables that are more easily measured directly in the field. In this study, a dataset of 638 data records of published crude oil fluid samples from around the world is evaluated. After filtering the dataset for outliers, 591 data records for BPP and 599 datasets for OFVF are evaluated with efficient hybrid machine-learning algorithms (multi-layer extreme learning machine --MELM-- and least squares support vector machine -- LSSVM) optimized using a genetic algorithm –GA-- and a particle swarm optimizer –PSO-- to improve their prediction performance. Four underlying variables are considered for each data record: temperature (T), solution gas oil ratio (Rs), gas specific gravity (γ g) and oil gravity (API). The PSO- MELM-PSO hybrid algorithm achieved the highest prediction accuracy measured in terms of root mean square errors (RMSE) of 33.5 psi for BPP and 0.0199 for OFVF. The four-hybrid machine-learning-optimization algorithms evaluated all outperform the empirical relationships used for many decades in the oil industry to predict BPP and OFVF. • Hybrid machine-learning with optimization achieve high prediction accuracy. • Bubble point pressure & formation volume factor predicted for 567 crude oils. • Multi-layer extreme learning machine with particle swarm optimizer works best. • MELM-PSO outperforms traditional empirical correlations in prediction accuracy. • Solution gas oil ratio is the most influential of four input variables. [ABSTRACT FROM AUTHOR]