Modeling and simulation play a key role in well construction planning, which can help to evaluate and optimize the engineering designs for a well. Today, many simulations use finite element analysis (FEA) and computational fluid dynamics (CFD) to model complex dynamic downhole conditions and behaviors of drilling tools. However, one challenge is that the complex simulation may take a few hours to run, which limits the usage to only a few well planning jobs. This limitation also poses as barrier in real-time monitoring applications, where under one second computation speed is required. In this paper, two approaches are presented for improving the performance of drilling simulations: smart depth selection logic for BHA tendency calculation, and reduced order model using machine learning for motor optimization modeling.
The ability to drill wells in high temperature formations is limited by the temperature specification of the available drilling tools. Most drilling tools currently have a temperature rating of 150°C, and there is an ongoing effort to develop tools with a higher temperature rating. A parallel effort is to develop the modeling capability to simulate the complex downhole temperature environment, to allow engineer to understand the temperature effect on drilling operation and better manage the temperature-related risks.
Many high temperature wells are planned in an extremely conservative manner. The engineer will rely on the formation temperature measured in offset wells to determine temperature gradient of the planned well. This temperature gradient will be used as a reference for all aspects of the well design, including drilling tools selection, cementing design, etc. In reality, there are many factors which affect the actual downhole temperature experienced by the tools. There is a complex interaction between heating from the formation, drilling fluid circulation, and the mechanical action of drilling tools. There are many forms of energy loss contributing to the downhole temperature, such as mechanical friction, rock cutting, and fluid friction.
A new state-of-the-art dynamic temperature model is developed to simulate downhole conditions in order to precisely predict downhole temperatures. This paper will explain the development of dynamic temperature modeling and how the model being used to plan high temperature well. The paper will also present several case studies where the modeling was used on planning high temperature well and comparison between model results and actual downhole temperature measurements.
The search for new reserves is pushing drilling into deeper reservoirs where formation temperature becomes a challenge. This becomes significantly challenging with current available drilling tools where the operating temperature can closely reach and at times surpass the downhole tools’ temperature specifications. Service companies are continuously improving technology to increase the temperature limit of the downhole tools to contend with the increasing temperatures; however, the ability to simulate downhole conditions and predict downhole circulating temperatures that the tools will be exposed to continues to be a key factor to successful drilling operations in high-temperature wells.
The current well-planning practice of designing high-temperature wells is based on static formation temperature measurement from the offset well and linear interpolation of the formation thermal gradient, which, in many cases, is too conservative, not taking into account various parameters that affect the total energy within the system that leads to the actual temperatures that the downhole tools are physically exposed to. Heat transfer occurs from the formation across the mud and downhole tools. Mud circulation carries mud with heated temperature from the bottom of the hole up to relatively cooler temperature as it approaches the surface, and back down the hole again; this will affect the actual downhole temperature that the tools are exposed to. Energy loss of torque and drag due to contact friction of drillstring with different formation and hydraulic pressure loss can also increase the borehole temperature and downhole tool temperature. A new state-of-the-art dynamic temperature model is required to more precisely predict downhole temperatures and which can be used to guide the downhole tool and services planning, along with operating parameters to be applied. This information is needed to be able to run as many downhole tools and measurements in real time to the tool temperature limit and reduce nonproductive time (NPT) due to pulling out of hole for temperature-related downhole equipment failures.
Dynamic temperature modeling takes heat transfer from virgin formation to mud and tool collars into account, while also calculating the effect of circulation, mechanical, and hydraulic friction. The dynamic temperature model can help engineers evaluate expected temperatures for different operations, guiding them to select the appropriate downhole tools for the job. The modeling was used in planning several case study high-temperature wells, and a comparison was done between model results and actual downhole temperature measurements.
This new modeling can change the high-temperature well planning perspective on the use of downhole tools in that a higher temperature rating (e.g., above 150°C) is not always required because downhole borehole temperature can be managed by applying the right drilling parameters and correct timing.
In cased completions, perforations provide the essential link between the wellbore and the reservoir. Productivity of the completion is promoted by optimizing perforation characteristics such as geometry, phasing and density, but unfortunately it is restricted by the perforation damage zone—a region of low permeability material surrounding the perforation tunnel, and created by the impact of the shaped charge jet on the rock fabric.
Perforating underbalanced has become the primary means of removing perforation damage and maximizing productivity, though the mechanism by which it does so is still not very well understood. Underbalance perforating also serves to remove some or all of the comminuted sand grains that fill the perforation tunnel immediately after penetration of the rock by the shaped charge jet. Predictions of the required underbalance to remove the damage zone or remove the comminuted fill are at best uncertain.
In this paper we describe the development of mathematical models that predict the pore pressure in the rock surrounding the perforation tunnel as the wellbore pressure drops during, for example, a dynamic underbalance operation. From this we calculate the magnitude and duration of the induced surge flow. The third stage in the analysis investigates the mechanisms through which the damage zone is removed.
As a result of these calculations we are able to predict the surge rate (and associated underbalance) required to remove the damage zone. Moreover we predict the perforation skin that results from incomplete removal of the damaged zone and for the first time we are able to determine how the skin depends on the degree and rate of underbalance.
A perforation is created by the impact into rock of the high-velocity jet from a shaped-charge explosive. This jet punches a hole at speeds sufficiently rapidly that the displacement of the rock is achieved by the creation of a large number of microfractures extending through grains of sand (Pucknell & Behrmann).
The radial displacement of the rock creates a residual elastic stress in the far-field undamaged porous medium (known as a "stress cage??). As the rock decompresses, this stress cage causes the failure of the most damaged rock adjacent to the perforation tunnel. The failed rock collapses into the tunnel, where it mixes with the remnants of the perforation jet to form a fill of loose debris.
Figure 1 shows a cartoon of a perforation immediately after it is created. The perforation tunnel is filled with a loosely packed debris of high permeability (1-10 Darcies). Immediately surrounding the perforation tunnel is a "damaged zone?? of fractured rock grains, extending a distance of order 20mm from the perforation tunnel. For liquid-saturated rocks, the porosity (and density) of the damaged zone is close to that of undamaged rock; for gas-saturated rocks, the porosity of the damaged zone may be much reduced. In the damaged zone, the pores are much smaller than in the virgin rock: many sand grains have micro-fractures, and large pore throats are filled with small fragments (Figure 2). This decrease in pore size results in the damaged zone having a permeability much decreased from that of undamaged rock Pucknell & Behrmann. Additionally, the damaged zone is partially deconsolidated: its strength is much less than undamaged rock.
The perforation tunnel creates a flow path between the reservoir (at pressure pr), and the wellbore (at pressure pw). The pressure difference pr-pw can drive a surge flow, either from the wellbore into the formation, or from the formation into the wellbore. Traditionally, wells are perforated with a "static underbalance??, with the far-field wellbore pressure less than the reservoir pressure. Usually this generates an underbalance surge flow (of timescale around 1 second), though this depends on the details of the wellbore and perforating gun design. Novel perforating techniques[2,3] exploit knowledge of the transient wellbore dynamics to engineer a surge flow that is both more rapid (building up over a timescale potentially as rapid as 10ms), and of greater magnitude than traditional perforating. (Figures 2 and 3 in ref 2 illustrate these two cases with pressure profiles measured in laboratory tests.) This dynamic underbalance can produce much cleaner perforations than traditional underbalanced perforating, leading to significantly more productive wells.[3, 4]