AI-Based Well Ranking System for Hydraulic Fracturing Events

We are presenting a new, highly intelligent AI-based ranking system for selecting the most appropriate candidates for well treatment. The system is trained to predict flow rates after hydraulic fracturing (HF) and rank wells by the expected effect of the event with machine learning techniques. We demonstrate a significant effort for preprocessing the available field data to create a dataset for training machine learning (ML) models. The dataset included information about geology, transport and storage properties, depths, oil/liquid rates before fracturing for target and neighboring wells. Each ML model has been trained to predict monthly production of oil and liquid right after fracturing and after flow stabilization. Also, confidence intervals of the prediction have been provided. To study the dynamics of future oil rate decline after HF on a stable regime, we have trained several regression models to make predictions at each future point (6 next months after fracturing). To estimate the effect due to HF, we defined expected production "without fracturing." Typically, wells behave with a stable decline trend of production that is approximated by Arps function. The function is defined before HF, then extrapolated to the period after the event where it shows expected production without fracturing. One may conclude about the effectiveness of HF by calculating areas difference under the extrapolated curve (cumulative production without HF), and ML predicted cumulative production for future six months. Reservoir engineers could calculate these differences for each well and create a ranking list from the highest effect to the lowest. The developed system does this automatically for the required oilfield or its part. Therefore, one may easily define the list of best candidates for HF. Gradient Boosting algorithm has been applied to obtain results. Feature selection and tuning of hyperparameters have been provided with the application of cross-validation technique. To test the developed approach, we have divided the dataset from 8 conventional oil fields at a ratio of four to one. The total dataset included 700+ well interventions. Then we have trained and validated models for flow rate prediction on the major part and tested on the holdout part. For different oil field determination coefficients (R2) and normalized root mean square errors (n-RMSE) for oil rate predictions were around R2=0.8 and n-RMSE=0.35 correspondently. The proposed technique is a new approach for fast, accurate, and objective selection of the candidates for hydraulic fracturing based on real-time state of a field. Such AI-based system could become very handy assistant for reservoir engineer in addition to hydraulic fracturing and hydrodynamic simulators. The presented solution computationally efficient and does not require detailed information about HF design.

CALCULATIONS OF SHORT PIPELINES TAKING INTO ACCOUNT THE FLOW STABILIZATION

Analyzing the existing dependences on the calculation of short pressure pipelines in turbulent mode, significant shortcomings of these recommendations were identified. A physical model of motion is proposed, which explains the processes occurring in the area of stabilization after local resistance and allows to establish the factors influencing its length.Experiments on pipes with different roughness and different pipeline fittings made it possible to analyze the change in kinematic characteristics in the area after resistance. Studies have shown that in short pipes, the length of which is less than the length of the stabilization section, there are less pressure losses than in the calculations by the usual method, when there are simply local and length losses. Dependencies and graphs are given that take into account the mutual influence of local resistances in the case of their location at a distance less than the stabilization area. The proposed recommendations make it possible to make more informed economic decisions when designing short pressure pipelines of various water supply systems.

Unlocking Opportunities for Gas Lift Well Surveillance - Building the Framework for Consolidated Data Capture and Processing

Gas lift operations are highly dependent on data quality and team competence to operate the asset efficiently. Traditional methods for gas lift well surveillance and diagnostics rely on wireline services, a method with growing constraints to adapt to constantly evolving well and operational challenges. The Well Intervention-less Tracer Surveillance System (WITSS) provides a cost effective, comprehensive approach to well surveillance without the reliance on tools entering the well. This results in reduced HSE risks and no associated deferred production. This paper describes a pilot implementation to evaluate the adequacy and accuracy of this technology in the context of ADNOC Onshore gas lift producers. The objective is to evaluate its performance against conventional method data sets and assess the reproducibility of data where no reference existed. The 10 well pilot included both accessible and obstructed wells. Data from the custom designed modular portable kit used for executing the surveillance activities, was analyzed and compared against conventional flowing gradient surveys with full data consumption in well models for comprehensive nodal analysis and opportunity identification. For this pilot, ten wells were surveyed twice using the WITSS method. Results were compared to traditional methods acquired through wireline surveys for accessible wells, and against established multi-phase flow correlations for obstructed wells. The pilot confirmed the WITSS method is as accurate as conventional gauge measurements in mapping pressure and temperature profiles in gas lifted wells. The WITSS method provided additional insight on accurate gas consumption based on the assessment of total gas lift utilization per well and allowed comprehensive model calibration and well performance definition. It also identified potential integrity issues via identification of primary injection at designed stations and secondary unwanted injection sites. Continuous compositional gas analysis of both injected and produced gas streams provided additional verification for analyzing gas lift injection performance. It also highlighted a change in fluid compositional analysis opening discussions for material selection review of the assets. Production uplift identified from 50% of wells was compliant with the reservoir management strategy. The value proposals of flow stabilization through gas lift valve re-calibrations and replacements, adjustment of injection flow rate and further controls on injection pressure management are under process for implementation. Full field scale up scenario is under preparation.

Field experiment on flow stabilization of working fluid in a top-heat-type thermosyphon

To address the problem of global warming, increasing efforts are being made to use renewable sources of energy, such as solar energy, wind energy, and geothermal energy. However, the effective use remains a major challenge for its sustainable development. In this study, we used a top-heat-type thermosyphon to heat water using solar energy and transport the low-density hot water from the source to the sink (high to low elevation) without an external power source. The transported hot water can be used for cooking, bathing, underfloor heating, and heating homes and buildings, and warming cold springs. However, a disadvantage of top-heat-type thermosyphon is the intermittent flow of the circulating working fluid under low solar radiation. To address this issue, the authors proposed and developed a control system to stabilize the intermittent flow and prevent equipment damage and failure due to the sudden boiling of water. Field experiments were conducted to assess the practicability of the developed controller. The results showed that the controller efficiently converted the intermittent flow of working fluid to continuous flow by reducing the pressure in the buffer chamber and thus lowering the boiling point of the working fluid in the header of the solar collector.

Numerical simulation of a stabilizing Poiseuille-type polymer fluid flow in the channel with elliptical cross-section

This work is devoted to the numerical analysis of stabilization of the incompressible viscoelastic polymer fluid flow in the channel with elliptical cross-section. To describe the flow, mesoscopic rheological relations are used, and resolving non-stationary equations are derived. For solving them a special pseudo-spectral method is developed and implemented. As time increases, under certain conditions on the parameters of flow the solution to the non-stationary problem stabilizes and converges to the one of three branches of the solution to the corresponding stationary problem. It is shown that the variation of the parameters describing polymer microstructure leads to the switch of stabilized solution between these branches. The work provides the results of simulation of the flow stabilization and the analysis of the threshold values of parameters at which the switching occurs.

Numerical Simulation of Disturbance Evolution in the Supersonic Boundary Layer over an Expansion Corner

Abstract— The linear and nonlinear stages of disturbance development in the supersonic boundary layer over a 10° expansion corner is investigated numerically within the framework of Navier—Stokes equations for Mach number 3. The effect of sudden flow expansion on the disturbance evolution is analyzed. The flow stabilization effect observable in the aerodynamic experiment is also discussed.

Development of an Automated Tracer Testing System for UASB Laboratory-Scale Reactors

Residence time distribution (RTD) curves play an essential role in the hydraulic characterization of reactors. Current approaches for obtaining RTD curves in laboratory-scale reactors are time-consuming and subject to large errors. Thus, automated systems to obtain RTD curves in laboratory-scale reactors are of great interest for reducing experimental errors due to human interaction, minimizing experimentation costs, and continuously obtaining experimental data. An automated system for obtaining RTD curves in laboratory-scale reactors was designed, built, and tested in this work. During the tests conducted in a cylindrical upflow anaerobic sludge blanket (UASB) reactor, the system worked properly using the stimulus–response pulse technique with sodium chloride as a tracer. Four main factors were found to affect the representativeness of the RTD curves: flow stabilization time, test water conductivity, temperature, and surface tension. A discussion on these factors and the corresponding solutions is presented. The RTD curves of the UASB reactor are left-skewed with a typical tank reactor’s flow shape with channeling and dead zones. A transitory flow behavior was evidenced in the reactor, which indicates the influence of internal turbulent flow structures. The system proposed herein is expected to help study the hydraulics of reactors using laboratory-scale models more efficiently.

Radio-echo sounding and waveform modeling reveal abundant marine ice in former rifts and basal crevasses within Crary Ice Rise, Antarctica

Crary Ice Rise formed after the Ross Ice Shelf re-grounded ~1 kyr BP. We present new ice-penetrating radar data from two systems operating at center frequencies of 7 and 750 MHz that confirm the ice rise is composed of a former ice shelf buried by subsequent accumulation. Stacks of englacial diffraction hyperbolas are present almost everywhere across the central ice rise and extend up to ~350 m above the bed. In many cases, bed reflections beneath the diffraction hyperbolas are obscured for distances up to 1 km. Waveform modeling indicates that the diffraction hyperbolas are likely caused by marine ice deposits in former basal crevasses and rifts. The in-filling of rifts and basal crevasses may have strengthened the connection between the ice rise and the surrounding ice shelf, which could have influenced local and regional ice dynamics. Three internal reflection horizons mark the upper limit of disturbed ice and diffraction hyperbolas in different sections of the ice rise, indicating at least three stages of flow stabilization across the ice rise. A surface lineation visible in MODIS imagery corresponds spatially to deepening and strong deformation of these layers, consistent with the characteristics of former grounding lines observed elsewhere in Antarctica.