Real-time prediction of Poisson’s ratio from drilling parameters using machine learning tools

Abstract

Rock elastic properties such as Poisson’s ratio influence wellbore stability, in-situ stresses estimation, drilling performance, and hydraulic fracturing design. Conventionally, Poisson’s ratio estimation requires either laboratory experiments or derived from sonic logs, the main concerns of these methods are the data and samples availability, costs, and time-consumption. In this paper, an alternative real-time technique utilizing drilling parameters and machine learning was presented. The main added value of this approach is that the drilling parameters are more likely to be available and could be collected in real-time during drilling operation without additional cost. These parameters include weight on bit, penetration rate, pump rate, standpipe pressure, and torque. Two machine learning algorithms were used, artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS). To train and test the models, 2905 data points from one well were used, while 2912 data points from a different well were used for model validation. The lithology of both wells contains carbonate, sandstone, and shale. Optimization on different tuning parameters in the algorithm was conducted to ensure the best prediction was achieved. A good match between the actual and predicted Poisson’s ratio was achieved in both methods with correlation coefficients between 0.98 and 0.99 using ANN and between 0.97 and 0.98 using ANFIS. The average absolute percentage error values were between 1 and 2% in ANN predictions and around 2% when ANFIS was used. Based on these results, the employment of drilling data and machine learning is a strong tool for real-time prediction of geomechanical properties without additional cost.

Real-time prediction of formation pressure gradient while drilling

“…AI has been broadly applied in oil and gas industry because it has not only the capability to solve complicated issues, but it also represents them with a high accuracy 27 . Intelligent models were developed for various targets such as estimating the equivalent circulation density in real-time 28 – 30 , pore pressure estimation while drilling 31 , 32 , porosity prediction 33 , resistivity prediction 34 , predicting mud rheological properties 35 – 39 , predicting the unconfined compressive strength 40 , estimating the oil recovery factor 41 , bulk density log prediction 42 , 43 , well planning 44 , lithology classification 45 , fracture density estimation 46 , estimating the static elastic moduli 47 , 48 , Poisson’s ratio prediction 49 – 51 , and prediction of formation tops 52 .…”

A Deep Learning Method for the Prediction of the Index Mechanical Properties and Strength Parameters of Marlstone

“…The results indicated that the predictability of the applied model led to RMSE = 0.0612/MAE = 0.0442 and RMSE = 0.0806/MAE = 0.0660 error rates in the training and testing stages, respectively. Siddig et al [ 43 ] utilized the ANFIS and MLP on extensive data logs recovered from a drilled well to predict the Poisson’s ratio variation on host sedimentary rocks. Both methods provided an R 2 , MLP, and ANFIS of 0.97, 0.98, and 0.97, respectively.…”

Machine learning-based prediction and inverse design of 2D metamaterial structures with tunable deformation-dependent Poisson's ratio

“…31,32 As abundant data are collected, one well-trained neural network is capable of the real-time 33 prediction of mechanical properties, such as Poisson's ratio. 34 Moreover, ML can be taken as an agent, reflected by the convolutional neural network (CNN) 33 model formulated to predict the thermal conductivity of porous graphene. 5 Noticeably, an inverse design methodology based on ML is also performed to determine the optimal choice of porous graphene with the lowest thermal conductivity, revealing the relationship between the hole distribution and thermal conductivity reduction in monolayer graphene.…”