We propose a new model and workflow to predict, quantify and mitigate undesired flowback of proppant from created hydraulic fractures. We demonstrate several field cases in which we predict significant proppant flowback and propose options for mitigation. Mitigation of proppant flowback is based on case- specific changes in the fracturing treatment design and modifications in the well startup schedule that preserve near-wellbore conductivity. The presented workflow integrates four key components of proppant flowback study: (A) simulation of the fracturing job with a high-resolution model of proppant placement inside a fracture; (B) subsequent simulation of flowback from the created fractures equipped by a validated proppant flowback model; (C) a series of laboratory experiments, which quantify the proppant flowback for a wide range of commercial proppants; and (D) an accurate mathematical model, which is validated by the results of laboratory experiments and integrated into a flowback simulator to predict the behavior of injected proppant. Each component is presented with sufficient details to demonstrate its necessity for accurate modeling of a coupled solid-and-fluid flow inside a fracture. The presented theory of proppant pack mobilization is based on the concept of a proppant pack erosion process evolving from free boundaries of proppant packs. The theory confirms that proppant flowback critically depends on flow rate, proppant, fracture, and reservoir parameters. Laboratory experiments on proppant flowback in a cell support these theoretical predictions. The theoretical model of proppant flowback is integrated into the numerical simulator of early-time production from a fractured reservoir and predicts flowback of both injected solids and fluids from a fracture. We show that the combination of the proppant flowback model, laboratory experiments, hydraulic fracturing design tool, and early-time production simulator result in a useful workflow for prediction and mitigation of issues with proppant flowback and production decline. The entire workflow was validated using field cases where proppant flowback was observed. Modeled amounts of flowed back proppant are in good agreement with amounts of proppant observed in the field. Hydraulic fracturing design optimization was performed to minimize or eliminate proppant flowback. The novelty of the proposed study is related to the model of proppant flowback, which accounts for erosion of the proppant pack and is calibrated against unique laboratory experiments. The presented model and proppant flowback mitigation workflow can assist in understanding and mitigating proppant flowback events that can occur during wide range of oilfield operations.

We demonstrate the advantages of a new hydraulic fracturing simulator comprising a fine-scale fracture hydrodynamics and in-situ kinetics model. In contrast to existing commercial modeling tools, it has a sufficient resolution and other functionality for adequate representation of modern stimulation technologies: pulsing injection of proppant, mixtures of multiple fracturing materials (fluids, proppants, fibers, etc.), materials degradation, etc. This simulator accounts for the influence of materials distribution on fracture propagation and calculates fracture conductivity distribution. We coupled it with a production simulation model and established a complete framework for hydraulic fracturing treatment design. In addition to the selection of the pumping schedule, this model can be used to define specifications for novel hydraulic fracturing materials. This is a step change tool for wellbore stimulation and production forecast.

It is often thought that in ultra-low-permeability unconventional reservoirs, proppant does not play a significant role in productivity because any proppant effectively results in a relatively infinite fracture conductivity. Although with over 75% of the treatment jobs pumped with slickwater in the shale reservoirs due to its cost benefit, it is also thought that slickwater has limited solids carrying capacity. Overflushing may compromise productivity by creating near-wellbore pinchouts. The study presented here aims to test some of these conventions through the use a new high-resolution slurry transport model in unconventional reservoirs. Hydraulic fracture propagation and solids transport are simulated across multiple wells and reservoir settings to measure the production performance of unconventional reservoirs. Wells completed in the Eagle Ford formation are studied using an integrated earth model built to capture the reservoir geology, petrophysics and geomechanics. A pseudo 3D model for fracture propagation has been coupled with a fine grid numerical simulation of proppant and fluid transport in hydraulic fracture. The transport model can distinguish and demarcate the corresponding bridging resolution. This allows capturing the effect of slurry with proppant bypassing bridged 2D elements. Multivariate analysis of over 50 cases with various combinations of hydraulic fracturing fluid with viscosity ranging from 1.5 cP (slickwater) to 362 cP (crosslinked gel) and various proppants ranging from 100 mesh to 20/40 proppants are evaluated using the 2D transport model to determine the impact on production. Additionally, various pumping schedules ranging from 750 lbm/ft to 3,000 lbm/ft are evaluated. Parametric sensitivity of the overflush fluid type and volume has been studied to measure the impact on proppant dislodgement in the near-wellbore area. Production performance for all the scenarios is studied through numerical simulation and an economic analysis workflow to evaluate the matrix for fracturing fluid and proppant selection.

Delays in the hydraulic fracturing stimulation process due to equipment issues may result in Non-Productive Time (NPT) and often associated with high costs. Both operators and service companies are interested in implementing a system to predict these events in advance so that they can proactively react and prevent failures before they occur. Our objective was to develop, implement, and demonstrate the capabilities of such a system. Based on historical data, we identified the most NPT-prone subcomponents of stimulation equipment. We combined failure reports and maintenance records with sensors data to train Machine Learning (ML) models. The final solution was deployed on the cloud to predict remaining useful time (RUL) in real-time and to start specific notifications sending whenever RUL dropped below the preset value. We analyzed 20,000+ failure records and identified the most NPT-prone equipment, subcomponents, failure modes, and root causes. We found out that root causes are very scattered. We evaluated the applicability of physics-based and data analytics approaches for Top-10 root causes by NPT duration and NPT count. When NPT count per root cause is low (<10), the trained ML models had insufficient accuracy. In this case, only subject matter experts and understanding of physics behind performance deterioration allowed us to boost data-driven models’ accuracy. We started with testing on historical data. Then we continued testing on real-time data without field intervention. This stage was required to check and to build trust in the predictive ability of the models. After initial testing, we eliminated some of the models and switched to monitoring with field interventions. For six months of operating, the model helped to decrease the NPT related to the selected subcomponent by 20 times. This work presents the process of enabling Intelligent Diagnostics for hydraulic fracturing equipment by implementing failure identification and prediction models. It combines both domain knowledge (failure modes of the equipment, sensors data) and recent advances in ML (time series analysis, data processing, and feature engineering). The developed system to be further extended to different equipment, as the workflow described in the present study is not specific to particular equipment.

No abstract