A Framework-Oriented Approach for Determining Attribute Importance When Building Effective Predictive Models for Oil and Gas Data Analytics

Abstract

Upstream oil and gas industry services work to deliver success throughout the life cycle of the reservoir. However, conventional sources of oil and gas are declining; hence, operators are increasingly turning their attention to unexplored and underdeveloped regions, such as high-pressure/high-temperature (HP/HT) and deepwater areas, as well as working to increase recoveries in mature fields. As reservoirs become more complex and drilling operations become more expensive, there is a growing need to reduce inefficiencies and costs. Petroleum engineers are increasingly using optimized formation evaluation techniques, software with three-dimensional (3D) visualization, and multidisciplinary data interpretation techniques. In addition, data from new downhole equipment provides reliable, real-time information about downhole conditions. With these improved techniques and better data, operators can model, predict, and control their operations better in real-time, thereby reducing inefficiencies and cost. However, this massive integration of varied data moving in higher volumes and at increased speeds has created an increasing demand on the computation and use of actionable and predictive data-driven analytics. Furthermore, these analytics must work in real-time to quickly discover the important critical data attributes and features for use in forecasting. Attribute importance is a well-known statistical technique used to identify critical attributes and features within a set of attributes that could impact a specific target. The benefits of deriving attribute importance include improved data set comprehension, reduced analysis efforts, and reduced time and computational resources required for actionable predictive analytics. Furthermore, attribute importance aids in understanding the attribute space and interactions, and it enables dimensionality reduction; thereby, leading to comprehendible, accurate, "parsimonious" (i.e., simple) models that lead to actionable predictive analytics. Today, numerous standard and custom techniques—each having their own strengths and weaknesses—are available for performing attribute importance. Each technique applies a different function to evaluate the importance of an attribute and score it to produce a ranked subset of attributes. Hence, it is possible (and is fairly common) to arrive at different subsets of important attributes based on the choice and configurations of the various techniques. To solve this problem, a feasible and effective framework-based approach is proposed that uses multiple attribute importance techniques and then intelligently fuses the results of these methods to arrive at a fused subset of important attributes. This paper describes a framework/"ensemble" (i.e., combined) approach called Segmented Attribute Kerneling (SAK). In addition, it discusses the results obtained from applying this approach on a dataset incorporating real-time drilling surface data logs and drilling parameters. This approach runs multiple attribute importance algorithms simultaneously, finds the intersecting subset of important attributes across the multiple techniques, and then outputs a consolidated ranked set. In addition, the method identifies and presents a ranked subset of the attributes excluded from the union. This paper compares the results of this approach to a single-approach technique based on the output of the predictive models.

Digital transformation: a review on artificial intelligence techniques in drilling and production applications

“…In drilling, the problems related to stuck pipe, also called pipe sticking in the literature, include mechanical and differential pressure stuck pipe prediction [35,36,42,43,46], differential pressure stuck pipe [33,37], and mechanical stuck pipe [47].…”

“…Dursun et al [43] developed a new approach for selecting attributes, with the aim of identifying those of greatest importance and that are critical to certain problems. Attribute selection is often useful as a stage prior to the machine learning process and applies a classification algorithm using only the selected attributes.…”

“…A Naïve Bayes probabilistic classifier for selection of attributes for stuck pipe predictive models Dursun et al [43] Real-time pipe sticking prediction tools based on data analysis Salminen et al [46] Neural network and support vector machine (SVM)-based tools for stuck pipe prediction Miri et al [33] Comparison between neural network and support vector machine (SVM) techniques for better stuck pipe prediction Jahanbakhshi et al [37] Real-time classifier system for detection and prediction of mechanical stuck pipe Marques [47] Loss of circulation Neural network-based tools for stuck pipe and loss of circulation prediction Moazzeni et al [35] Monitoring real-time system for loss of circulation prevention • Drill string composition and well equipment configuration (bottom hole assembly); • Operational drilling parameters (rate of penetration, rotation, vibration, torque, drag and weight, internal and external pressure of the casing); • Drilling mud properties (weight, fluid flow, filtrate, solids content, gel strength); • Formation parameters (lithology, pore pressure, fracture pressure, discontinuities, and faults), and; • Well parameters and dimensions (shoe depth, well vertical depth, degree of inclination, well bottom pressure); • Depth and position of sensors.…”

“…The literature review allows finding relationships among the inputs (data and parameters available) and the main problems. For mechanical and differential stuck pipe, the AI techniques used data from drilling string composition, operational drilling parameters, and drilling mud properties [36,42,43,46,47].…”

“…Results show that AI techniques can identify and distinguish mechanical from differential stuck pipe [36,42,43,46].…”