Multiple bridge plugs are typically set at intervals to isolate stages in multistage hydraulic fracturing jobs in horizontal wells. Wellbore cleanout is usually performed to remove milled plug debris. In the last few decades, CT technology gained popularity, over conventional jointed-pipe snubbing with rig-assist snubbing units, for post-stimulation milling of plugs and wellbore cleanout, especially in the shale market. Well cleanouts operations are by far the most common CT service, accounting for over half of all coiled tubing operations worldwide. This technology has proven to be more reliable and cost-effective compared to the other methods. However, with the increasing demand for lateral extensive unconventional wells to expose more shale reservoir intervals and due to CT’s inability to rotate like a workover rig’s drill string, the Coil Tubing industry continues to work persistently to remain efficiently and economically competitive.

The cleaning techniques initially utilized in the composite plug drillouts operations are adapted mainly from the drilling and workover operations in vertical wells. These conventional techniques are continuously being adjusted to address and solve the challenges encountered in horizontal shale wells. Despite that, the focus of this discussion is geared more towards the shale market. The discussed cleaning scenarios are still applicable in both workover and post-fracturing operations. As such, the overall objective of this research is to optimize the well cleaning process by better understanding the current practices and recommend the best techniques and suggesting utilizing real-time monitoring tools and machine learning as the best ways forward.

Several studies have been conducted in the Oil and Gas industry, addressing different coil tubing cleaning challenges. This operation is complicated as its efficiency relies on many variables, including formation and completion properties, components of bottom hole assembly, motor and mill selection, type of circulating fluid, and choice of equipment.

It was indicated that most of the fracture-stimulated completions use plug n’perf. The technique requires either a connected pipe or a CT with a positive displacement motor (PDM) to mill out the composite plugs (CPs) after the fracturing maneuver is completed. In the US alone, between 100,000 to 140,000 composite plugs are set in horizontal fractured wells and milled out every year. Efficient milling is critical in order to achieve operational goals. Most of the milling operations are currently decided at the well-site, and risk assessment is only done after operational complications arise. The authors highlighted best practices after reviewing many operations across the US. The newly developed scheme used information about pre-job planning, mill design, motor selection, fluid selection, weight on bit, wiper trip frequency and speed, whether to produce the well during milling, surface flowback equipment, and operational contingencies [1].

CPs had been used for isolation in multistage stimulation for over 15 years in the US. The technique is cost-effective as hundreds of thousands of plugs are available to the market at competitive prices. These plugs must be removed effectively through wellbore cleanout techniques that involve milling with coiledtubing. The authors emphasized that full-scale milling tests must be performed on different plugs to optimize milling performance, particularly in tight-economic plays. The authors indicated that a millability test should involve many plugs since plugs are milled out sequentially. They added that during the experiments, real bottomhole conditions should be simulated, and millout time, penetration rate, debris size, and stump size should be measured. The authors demonstrated howmilling tests could improve the design of a plug, enhance mill selection, and improve the economics of the completion operation [2].

The Duvernay Formation in North Western Alberta contains 443 × 10¹² ft³ of gas and 61.7 × 10⁹ barrels of oil. Many laterals have been fractured to maximize the reservoir surface area to flow. Plug n’ perf is the most common method used to hydraulically fracture the horizontal wells. Plugs have been milled using coiled tubing in order to initiate production. The authors explained that to optimize the milling process and maximize production, mill bits, motors, and plug designs must be investigated. They added that research and development had improved mill bit design and motor life. The disintegration of new plugs has become easy and quick [3].

Nonetheless, little research has been done on solids transport in the wellbore. Service companies have been relying on rules of thumb and experience to identify the type of gels or determine the needed number of wiper trips during milling and wellbore cleaning operations. Most milling techniques leave a substantial amount of debris that leads to stuck pipes and incur additional completion costs. The authors explained that a new method had been devised to improve plug milling efficiency. The goal was to study fluid rheology to optimize the complete milling procedure. In the paper, case histories comparing the results before and after the implementation of the state-of-the-art concept as well as project economics have been highlighted. The authors conclude that conventional practices including relying heavily on gel sweeps to clean the well are inapplicable where there is no pipe rotation and highly recommend utilizing a fluid monitoring and controlling system (FMCS) to closely examine the amount of chemicals added to the milling fluid to reduce operational costs.

Operators needed to reduce the completion cost of gas wells due to low gas market prices. For that matter, the mill out cost of composite frac plugs and frac sleeves have been lowered. Shell’s Eagle Ford group has abandoned traditional milling procedures that used frequent short trips, best known as wiper trips. They incorporated an optimized fluids’ system technology and an advanced bottomhole assembly (BHA) that eradicated short trips and produced cleaner wellbores. The new techniques led to a significant reduction in operational time and cost. In addition, the authors shed some light on the past CT performances and the necessity for change. They reviewed improvements made on fluid quality assurance and controls (QA/QC), changes made on CT selection, and BHA design. The authors indicated that these changes resulted in a substantial reduction in time and cost per well and cycle time. The changes are now the new standard for Shell’s Eagle Ford CT operations [4].

It was indicated that composite plug coil-tubing drillout (CTDO) must be efficient and remain economically competitive. The author suggested that by reducing and perhaps eradicating wiper trips, CTDO could be a viable technique since less fluids and chemicals are pumped, and henceforth, pipe fatigue is reduced. This paper highlights how CT string design, bottomhole assembly (BHA), flowback arrangement, drilling techniques, fluid system, and monitoring and control can make CTDO an effective process. The author presented two case histories, where wiper trips were successfully curtailed/disregarded. The first is in the WillistonBasin. The 10,135-ft vertical and 9868-ft lateral well with 24 composite plugs was drilled out by a 22,230-ft, 2 3/8-in.-CT. The second is in the Denver-Julesburg basin. The 6780-ft vertical and 7815-ft lateral well with 39 composite plugs was drilled out with a 17,990-ft 2-in.-CT.The study recommended a more standardized drillout approach that takes into consideration flowback arrangements, drilling techniques, Bottomhole Assembly, CT string design, and continuously monitoring and controlling the operation parameters. The process has proven to be viable since the operator saved on the well intervention time [5].

Challenges to mill out composite bridge plugs (CBPs) across an over-pressured zone in a NewZealand field were discussed. While milling the plugs that isolate the over-pressured zone, the differential pressure applies substantial forces on the milling assembly that lead to buckling and possibly failure of the CT unit. A bridge plug with anequalizing path was utilized to overcome the differential pressure. During the milling process, the system indicated that pressure equalization took place. The process was monitored using a CT, real-time downhole communication unit that provided a better understanding of the downhole conditions and motor performance during the milling of plugs, thereby optimizing the operation [6].

It was also indicated that it is imperative to optimize plug milling and well cleanout techniques. They added that well cleaning continued to be a challenge due to the large number of operational variables involved. Considerable amounts of solids remain in the horizontal section of the wellbore after the cleanout is completed due to the fact that most operators use cleaning techniques that are best suited for vertical wells. This paper highlights cutting-edge cleanout procedures that utilize non-viscous fluids. Hole cleaning performance drastically improved when viscous pills were eliminated, and chemical dosage was controlled. The use of non-viscous fluids improved overall efficiency during the cleaning and milling operations from the economic and environmental viewpoints [7].

Vertical wells cleanout best practices had been applied for horizontal wells. Operators have been pumping a high volume of viscous gel sweeps and have been performing several wiper trips to remove debris from the wellbore. Recently, to perform milling operations, CT with a larger diameter has been used due to its enhanced set-downforce and amplified annular velocities. The amount of wiper trips with larger diameter coil have been reduced, and time and money have been saved. The authors added that proper fluid rheology needs to be maintained throughout the operation. They noted that larger diameter 2-inch CT and proper cleaning fluid rheology had reduced mill out costs by 40% [8].

Raageshwari gas field was studied. The formation tops ranged from 9186 to 11,483 ft, and the reservoir is unconventional volcanic with pore pressure and temperature of 5200 psi and 284 ˚F, respectively. Wells were completed with a 3-1/2” tubing and a plug n’ perf (PnP) technique was utilized in order to frac the formation. 3-1/2” 10,000-psi bridge plugs were placed between fractures and then milled with coiled tubing (CT) unit. The authors indicated that substantial amounts of gas and condensate were produced during milling. This led to increased wellhead pressures. Circulation pressure and weight on bit (WOB) were then hiked to speed up milling. Motors, unfortunately, stalled frequently and led to ineffective milling. The authors also described how optimized WOBs, penetration rates, and a newly devised milling fluid system led to increased frac fluid recovery and more considerable differential pressures across the motor. The study presented values and optimum ranges for critical parameters for efficient plug milling operations in deep retrograde gas condensate type of reservoirs. The process reduced the bridge plug milling time and saved about 5% of the well cost [9].

The experience gained from plug drillout techniques in the VacaMuerta formation in Argentina was presented. The authors analyzed drillout times, debris size and shapes, hole cleaning, and tool dull grades for different mills, roller-cone, and polycrystalline diamond compact (PDC) bits. They investigated drilling parameters, several runs to complete a well, drilled plugs per run, and tolerances regarding casing drift. They examined more than 500 plugs that were drilled out in more than 60 wells. The authors explained that to reduce drilling time, improve economics, minimize risks, and reduce CT system fatigue, the use of five to six-blade PDC bit outperformed the other tested bits. The bit minimized debris size, reduced stuck pipe risks, and improved well production [10].

A factory model that has been utilized for 3 years and has become an industry standard. The approach applies to any area irrespective of lateral length, several plugs, and reservoir measured depth. The method enhanced CT drillout routine and has been successful, repeatable, and operationally safe [11].

The results from over 200 frac-plug drillout (FPDO) and flowback operations from the United States and Argentina were presented. Positive FDPO implies that the coiled tubing injection rate is higher than annular formation flowback rates. The authors showed that monitoring flowback with a high-frequency flowmeter was a useful technique for optimizing well productivity during well clean-up. The paper also indicated that the use of a dual-energy gamma-ray multiphase flowmeter equipped with a dual-energy gamma-ray helped quantify the amount of debris produced during FPDO and flowback [12].

A composite frac plug was set off-depthin an unconventional shale well in the northern panhandle of West Virginia. The operator used a wireline-conveyed tractor and tractor-based milling tool to mill the plug. The authors indicated that other tubing-based solutions required more mobilization time and intricate logistics to rig down and mobilize the equipment. The tractor took only 50 minutes to drive down to plug in a 1718-ft lateral section. The authors explained that the capacity to release plugs via wireline is more economical than the tubing-based practices. These take days, sometimes weeks, to mobilize at considerable cost to operators [13].

The use of a generation four CT unit and the application of fluid rheology resulted in an efficient milling operation along a West Central Alberta horizontal well that extended for 27,379 ft. The use of heavy-walled tubing provided the required WOB for the milling operations. The use of fluid rheology applications and real-time monitoring of Reynold’s number ensured downhole turbulent flow conditions. It helped optimize solid-transport and chemical consumption as well [14].

It was also stated that PnP is a commonly used technique for multistage fracturing procedures. The authors added that CT-assisted PnP operations in high-pressure/high-temperature (HP/HT) wells are challenging. The “A” field in India is HP/HT with a bottomhole temperature (BHT) of around 310 ˚F and a reservoir pressure of about 9000 psi. Best practices presented in the PnP cycle included operational parameters such as CT speed, pumping rates, and bottomhole assemblies with smaller outer diameters. The parameters led to doubled operational efficiency during the maneuvers [15].

Table 1 is a summary of the current field practices’ developments and findings. The table also summarizes the recommended ranges of key performance indicators (KPIs). These indicators include annular velocity (AV), Reynolds Number (Re), cleaning gel viscosity, return rate, wiper trip times, cleaning time, and circulation rate:

In the reviewed field cases, formal training and effective communication between all stakeholders were of prime importance. It was also proposed to use a dynamic well construction workflow to predict early well performance issues and perform wellbore conditioning before setting up bridge plugs. The importance of millability tests has been emphasized. It was suggested to test at least two plugs as a pre-job routine. It was also advised to use generation four CT with larger diameter and smaller outer diameter bottomhole assemblies to improve milling performance and reduce wiper trips. PDCs with five to six-blades minimized debris size and enhanced well production. It was also recommended to utilize an early-stage fluid and flowback real-time monitoring and controlling system to reduce milling time, coil tubing, fatigue stresses, and cost. Fluid quality

assurance and controls (QA/QC) assessments and coil tubing selection should be revised. It was also suggested to use non-viscous fluids for better cleaning performance. Some other practices recommended the use of a wireline-conveyed tractor over tubing-based operations.

Furthermore, data analytics’ studies conducted to improve and better understand hole cleaning in horizontal wellbores are presented in the next section.

Many authors ( [11] [16] - [23] ) used data analysis to better understand and optimize hole cleaning operations.

Based on the Buckingham-π theorem, several dimensionless groups were defined and a semi-empirical correlation to predict the critical deposition velocity was developed. As shown in Equation (1), velocity in the open area above the bed was assumed to be the same as the critical velocity [16].

V s 0 = v o p e n (1)

where, V s 0 is initial cuttings volume remaining in the wellbore annulus and is the in-situ liquid velocity in the open area above which is the used circulating cleaning velocity.

The authors also predicted hole cleaning time (t), when circulating, as:

t = V s 0 C 0 q l ln ( V s 0 V s ) (2)

where, C 0 is the initial cleaning concentration, q l is the circulated mud flow rate, and V s is the total cuttings volume remaining in the wellbore annulus.

The optimum cleaning speed ( v o p t i m u m ) was defined as:

v o p t i m u m = C q l ( 1 − C ) C b A b (3)

where, C is the cuttings pick-up capacity, C b is the cuttings volume concentration in the bed, and A b is the cuttings bed cross-sectional area.

A hole cleaning model for CT operations was also developed [17]. The model was used to determine the maximum flow rate the CT can deliver and the specific annular liquid velocities using flow friction, bottomhole pressure, and configuration of the work string.

Additionally, a mathematical model for solid-liquid flow at low velocities for different configurations of horizontal and eccentric annuli was published [20]. The authors defined the parameters that affect the critical velocity given by:

v c = f [ d p , d 1 , d 2 , h b , C v , ( ρ s − ρ l ) g , ρ l , μ l , β , H L , ε ] (4)

where, d 1 is drillpipe diameter, d 2 is openhole diameter, C v is cuttings volume concentration, μ l is mixture viscosity, β is wellbore inclination angle from vertical, H L is liquid holdup, and ε is eccentricity of drillpipe.

Furthermore, a one-dimensional unsteady-state two-phase mechanistic model of cuttings transport with foam was developed [22]. The authors introduced a new critical deposition velocity correlation for foam-cuttings flow. The model is solved numerically to predict cuttings bed height as a function of drilling rate, gas and liquid injection rates, rate of gas and liquid influx from the reservoir, and borehole geometry. Other foam modeling efforts in horizontal wells have also been highlighted ( [18] [19] [21] ).

To study variation in downhole tool data and optimize the procedure, real-time data analysis was utilized [11]. Technology implementation helped remove unnecessary short trips and eliminated stuck coiled tubing unit operations. As a result, more than 320 horizontal wells with a depth ranging from 5000 to 10,000 ft have been successfully achieved. Sixty percent of the completion time and sixty-one percent of the cost were saved.

Besides, post drillout data in the Permian Basin and the Eagle Ford was analyzed [23]. The authors used 63 wellbores completed with 2664 composite plugs, 60 wellbores used a combination of 382 dissolvable and 1792 composite plugs, and 16 wellbores completed with 543 dissolvable plugs. The intent was to correlate key performance indicators (KPIs) such as time in well, average milling time, wash time, plug spacing, and the total number of plugs with well geometry, lateral length, and plug type. Statistical analysis indicated that the overall procedure time when plugs were combined was reduced by 46% compared to that spent on dissolvable plugs only. Data analysis also showed that at least 45% of dissolvable plugs did not dissolve, and CT intervention was required. Table 2 summarizes study findings in the area of hole cleaning model development.

Best practices indicated that KPIs like circulation rate, fluid viscosity, annular velocity, and Reynolds’ number impact wellbore cleaning efficiency. Optimum circulation rate ranged from 2.1 to 4 bbl/min depending on the size of the CT-casing pair. Viscosity in the range of 5 to 15 cp provided maximum solids suspension. Typically, annular velocities in the range of 143 to as high as 300 ft/min,

depending on the size of the CT-casing pair, and Reynolds’ number above 20,000 ensured optimum transport efficiency.

The paper review was also used to identify research gaps in the areas of real-time monitoring and data analysis, data analytics, and machine learning.

Important advances had been made in the area of bridge plug cleaning techniques. In the discussed case histories, cleaning techniques for bridge plug drillouts came a long way. Operators are drifting away from vertical well cleaning techniques to adapted techniques for horizontal wells. Case histories indicated that the use of a fourth generation CT unit is a viable solution compared to jointed pipe. The method provided more WOB and reduced buckling problems. More advanced bridge plugs have also been used. Research showed that milling bridge plugs is more efficient than soluble plugs. The use of rheology and real-time monitoring has also led to better cleanout performance.

A significant research gap exists in the area of bridge plug drillouts’ cleaning model development. Despite the fact that an immense data base for hole cleaning exists, attempts made in the development of a hole cleaning coefficient remains insignificant because of the complexity and lack of understanding of the physics of the cuttings transport problem.

The way forward is to use data analytics and machine learning using python as a coding language to develop empirical models that can be used to predict hole cleaning efficiency.

The data supporting the conclusions of this paper are included within the paper. Any queries regarding these data may be directed to the corresponding author.

The authors have agreed to submit it in its current form for publication in the journal.

The authors declare that they have no competing interests.

Trabelsi, H., Seibi, A., Liu, N., Boukadi, F. and Trabelsi, R. (2021) Bridge Plug Drillouts Cleaning Practices—An Overview. Natural Resources, 12, 19-33. https://doi.org/10.4236/nr.2021.122003

A a n n u l a r = annular area, m²

A b = cuttings bed cross-sectional area, m²

C = cuttings pick-up capacity, in decimal

C 0 = initial cuttings concentration, in decimal

C b = cuttings volume concentration in the bed, assumed as 0.52

C b = A b A a n n u l a r × 1 00 % , the bulk volume concentration in the annular test section,

percentage

C v = cuttings volume concentration, cuttings true volume/(cuttings true volume + mud volume), %

d P = particle diameter, m

d 1 = drill pipe diameter, m

d 2 = openhole diameter, m

FTFPs = flowthrough frac plugs

g = acceleration caused by gravity, m/s²

h b = bed height, m

H L = liquid holdup, in decimal

q l = circulated mud flow rate, m³/min

t = time, seconds

v o p e n = in-situ liquid velocity in the open area above, m/s

v o p t i m u m = optimum cleaning speed, m/min

V s = total cuttings volume remained in the wellbore annulus, m³

V s 0 = initial cuttings volume remained in the wellbore annulus, m³

μ l = viscosity of mud or mixture viscosity of gas and liquid, Pa.s.

ρ l = liquid density, lbm/gal

ρ s = solid density, lbm/gal

β = wellbore inclination angle from vertical; 90° refers to the horizontal position

ε = eccentricity of drill pipe, 2 χ d 2 − d 1 , in decimal