ISI Logo

Dynamic Shale Analytics

Managing & Mitigating Frac-Hit; AI-Based Shale Reservoir Management

Comprehensive, Dynamic, Full Field, Coupled Reservoir-Wellbore Simulation of the Fluid Flow in Shale Plays Using Artificial Intelligence and Machine Learning

Shahab D. Mohaghegh - December 2018

Frac-Hit, also referred to as Dynamic active-fracture interaction, and Fracture-driven interaction, is defined as the interaction between multiple, simultaneously growing active fractures (SPE Technical Report “Fracture-Driven Interaction Between Horizontal Wells”) . Today Frac-Hit is one of the most important issues that needs to be addressed in shale assets. This is due to the increase in the number of shale wells that results in reduction in the distances between wells. The most important contributors to Frac-Hit are (a) Reservoir Characteristics, (b) Well Spacing & Stacking, and (c) Completion Design and Implementation. It is impossible to maximize production and recovery of hydrocarbon from shale wells without addressing the interaction and interference between multiple Parent and Child wells (Frac-Hit).

Dynamic Shale Analytics is a new technology that combines pattern recognition capabilities of Artificial Intelligence and Machine Learning with the domain knowledge related to reservoir, completion, and production engineering in order to address, manage, and mitigate Frac-Hit. Our decades of research and development in reservoir engineering and reservoir modeling that has been augmented by the state of the art in Artificial Intelligence and Machine Learning has recently resulted in a major conclusion. We have come to the conclusion that using well productivity snapshots in time as output of the models or analyses cannot provide the type of tools that are required for addressing issues such as Frac-Hit.

Almost all current applications of Artificial Intelligence and Machine Learning use oil and gas production from shale wells in the form of productivity indices such as 30, 90, 180 … days of cumulative production, as output. Such production characteristics have proven to be insufficient for the prediction, management, and mitigation of Frac-Hit. We have concluded that Frac-Hit prediction, management, and mitigation require modeling the dynamics of each individual well in the context of reservoir dynamics, also known as reservoir simulation and modeling. However, the problem is that such models cannot be developed using traditional numerical reservoir simulation and modeling technology that is currently being used in our industry.

Traditional Numerical Reservoir Simulation have been contributing to the oil and gas industry for decades. The foundation of the traditional numerical reservoir simulation technology is our current understanding of the physics of the fluid flow in the porous media. When this technology is applied to shale plays, given the state of our current understanding of the complex physics of completion and production in shale wells, instead of being a scientifically accurate modeling technology, the traditional numerical reservoir simulation turns into a tool that is used to accommodate the interpretations, biases, and preconceived notions of the practitioners.

The degree of simplifications and the vast number of assumptions that are included in the traditional numerical reservoir simulation technology when it is applied to shale plays turns this originally scientific tool into a justification tool for the decisions that has already been made (prior to the modeling) by the practitioner. The major problem with the use of traditional numerical reservoir simulation in shale plays is that it can be used to justify completely contradictive decisions. In other words, no matter what is the nature of the decisions that are made and how opposite of one another they are (Increasing versus decreasing the stage length, small vs. large number of clusters per stage, smaller vs. larger amounts of proppant, fine vs. coarse proppant, high vs. low injection rates and pressure, etc.), the traditional numerical reservoir simulator can be designed such that to justify them all. This is due to the massive amount of simplifying assumptions that are involved in this modeling process that have nothing to do with the realities faced by the operators in the field.

The ideas presented here have been first mentioned (using a different approach) in 2013 (“Reservoir Modeling of Shale Formations”. Journal of Natural Gas Science and Engineering. Vol. 12, [2013], pages 22-33, and “A Critical View of Current State of Reservoir Modeling of Shale Assets”. Mohaghegh, S. D., SPE 165713 - SPE Eastern Regional Conference and Exhibition. Pittsburgh, PA , 20-22 August 2013). Besides all the well-known assumptions and simplifications that are included in the traditional numerical reservoir simulation technology when this technology is applied to conventional plays, its application to shale plays brings about a new set of categories of gigantic simplifications and assumptions that completely compromise any and all applications of this technology to shale plays. Some of the assumptions and simplifications associated with the traditional numerical reservoir simulation application in shale plays can be divided into the following three categories:

  1. Reservoir
    • Naturally fractured reservoir
    • Reservoir characteristics
      • Petro-physical characteristics
      • Rock-Fluid Characteristics
      • Geo-mechanical characteristics
      • Geo-physical characteristics
  2. Completion
    • Hydraulic Fracturing characteristics (Fracture Half length, Fracture Height, Fracture Width, Fracture Conductivity)
    • Geometric distribution of the induced fractures
    • Impact of reservoir characteristics on the Geometric distribution of the induced fractures
    • Impact of number of clusters, stage length, and pumping characteristics on the induced fractures
  3. Production
    • Flow Back
    • Impact of the Operational constraints (Choke setting) on production behavior
    • Which clusters and/or stages are contributing to the hydrocarbon production.

The above list includes some of the assumptions and simplifications that are made in the application of traditional numerical reservoir simulation technology in shale plays. Details of this assumptions and simplifications will be covered in a separate article. Anyone that has used traditional numerical reservoir simulation in order to model fluid flow, completion, and production in shale plays can provide much details about these and many other assumptions that are involved in this modeling technology.

Is there an alternative?

Yes, There is.

It is called “Dynamic Shale Analytics”.

Dynamic Shale Analytics (DSA) introduces a new technology for building a comprehensive, full field, dynamic and coupled reservoir-wellbore simulation model that avoids any and all assumptions and simplifications. Dynamic Shale Analytics is a reservoir+wellbore simulation model based only on field measurements and data that is gathered from drilling, completion, operation, and production of all the shale wells in a given asset. Dynamic Shale Analytics has the capability of simultaneously history matching hundreds or thousands of shale wells that have been drilled through hundreds of pads.

Dynamic Shale Analytics discovers the patterns of induced fractures at every cluster as a function of well placements, reservoir characteristics, and completion practices, as well as operational constraints in order to predict and mitigate interactions between multiple Parent and Child wells. Completion optimization, well spacing and stacking, as well as management of operational constraints (choke setting and wellhead pressure) for the production optimization are part of the Dynamic Shale Analytics deliverable.

When it comes to full filed simulation and modeling of hydrocarbon production from shale wells, another distinguishing factor of the Dynamic Shale Analytics is the use of hydrocarbon production data as the model output instead of using it as input to the model. In traditional numerical modeling, Flowing Bottom-Hole Pressure (FBHP) is used as the model output and hydrocarbon production is used as input to the model. The fact is that Flowing Bottom-Hole Pressure (FBHP) is hardly ever measured in all the wells in a Shale play and using it to check the model validity leaves much room for interpretation and uncertainty. In Dynamic Shale Analytics, wellhead pressure (or choke setting depending on the availability of field measurements) is used as model input and measured hydrocarbon production (details production profiles [oil, gas, and water]) is used as model output. Confirming the validity of the simulation model when the measured hydrocarbon production (details production profile) are matched can be much stronger and far more acceptable.

Other operations such as re-frac candidate selection and design as well as recovery optimization and production enhancement by gas injection into the shale wells can also be modeled realistically through Dynamic Shale Analytics using the pattern recognition capabilities of Artificial Intelligence and Machine Learning. Current limitations associated with the application of Artificial Intelligence and Machine Learning regarding the completion and production optimization is mainly due to the static nature of the modeling and data analytics techniques that are used for this purpose. In order to model Frac-Hit, dynamic interaction between multiple parent and child wells must be modeled as a function of time throughout the life of each shale well. Dynamic Shale Analytics overcomes these limitations and accommodates comprehensive dynamic reservoir and field management in a unique and unmatched manner.

Dynamic Shale Analytics is part of a new modeling technology in the oil and gas industry known as “Subsurface Analytica” (www.IntelligentSolutionsInc.com). This new technology is changing the way modeling is done in our industry. Instead of starting with formulating a physics-based model of the fluid flow through porous media and then modifying the geological models to achieve a history match, “Subsurface Analytica” takes a completely different approach. Dynamic Shale Analytics builds a physics-based model of the fluid flow through porous media that is not through detail formulation, rather through field measurements, pattern recognition, and data-knowledge fusion. “Subsurface Analytica” is applicable to both conventional and unconventional resources.

In Dynamic Shale Analytics (an AI-based approach to reservoir modeling in unconventional resources) field measurements are the foundation and the building blocks of modeling the physics of fluid flow in shale. In Dynamic Shale Analytics we model the physics of fluid flow through the complex unconventional porous media that has been highly impacted by the operators through vast number of induced hydraulic fracture. Dynamic Shale Analytics generates a better understanding of the physics by discovering the highly complex relationships between all measured data. In order to match the field measurements (hydrocarbon production), Dynamic Shale Analytics discovers the relationships between all the existing field measurements (reservoir characteristics, well placement, completion, operational conditions, etc.), rather than making assumptions in order to serve the practitioners interpretations, biases, and preconceived notions.