Sand Production

Sand production and control remain critical challenges in reservoir management and production operations. Sand results in high removal costs, equipment erosion, and significant maintenance expenditure. Sand production issues arise:

1. At the reservoir appraisal stage, where the risk of sand production must be quantified to formulate and cost the reservoir management strategy or to satisfy regulatory authorities. There are usually no historical sanding data, and the focus is on predicting the conditions under which sand production might occur.
2. In fields where sand production has been encountered and must be economically managed. The goal is to regulate and control sand production rather than the often unwarranted and widespread adoption of sand exclusion/retention measures.

Sand production prediction is not an exact science. Although theoretical analytical and numerical models exist, it is necessary to approach the problem with a good engineering-based understanding of the limitations of the rock, well and reservoir data, and an appreciation of all the other sources of information that can be targeted on the problem.

Sand management involves the development and monitoring of optimal sand control strategies that recognize the particular problems and constraints of the field development, yet maximize productivity and completion longevity. Understanding the sensitivity of productivity to different sand control methods is essential to the longer term economic success of this strategy.

Sand Production Expertise

eP offers the multi-disciplinary capability required to address the technical challenges in sanding prediction and sand control optimization. Staff within eP have extensive experience in petrophysics, geomechanical analysis, completion practices, and well performance evaluation. The work is performed either on a conventional, study basis or, interactively, in the client's offices where eP's staff act as the primary focal point for data gathering, sand production analysis, remedial planning, and coordination and monitoring of appropriate well interventions.

There are a number of advantages in choosing eP rather than a sand control service company or an individual consultant:

• eP is an independent company not associated with any particular methodology, and not linked to sand control equipment or service vendors
• Access to the accumulated expertise and knowledge of an established petroleum engineering consultancy
• eP has the capacity to address all the primary sand production issues, unlike the sand control service companies
• Expertise in the acquisition and analysis of appropriate production, petrophysical, and laboratory data
• Practical sand production prediction models
• Well performance modeling capabilities for optimum sand control and completion strategies
• Practical downhole and well site skills and experience
• Established technical links with the foremost rock mechanics and formation damage laboratories in the UK give access to state of the art rock testing technologies

Methodology of Sand Production

The methodology behind sand production assessment and control management involves 3 key stages

1. Design and implementation of a data acquisition program
2. Sand production analysis and interpretation
3. Development and monitoring of optimum sand control strategies

Mohr-Coulomb Failure Criterion

Well Data Acquisition Program

Sand production is normally caused when the shear stresses acting on the rock surrounding the perforation cavities or wellbore exceed the rock strength, and lead to rock failure. Factors which control rock failure and subsequent sand production are:

• In-situ reservoir stress magnitudes and orientation
• Well azimuth and deviation (in relation to in-situ stresses)
• Formation rock mechanical strength and petrographic properties
• Reservoir fluid properties
• Reservoir drawdowns and depletion
• Hydrocarbon and water production rates

eP can specify and manage the data acquisition program. Typical data suites will depend on available budget, but can include:

In-Situ Stress

• Well test and RFT data analysis
• Density log analysis
• Dipole sonic and temperature logs
• Calliper logs (breakout analysis)
• Analysis of drilling induced fractures on oriented core
• Differential strain analysis tests on oriented core
• Hydraulic fracturing tests
• Mini-fracs and step rate tests
• LOT tests

Rock Property Characterization

• Petrographic Analysis
  • Thin section and XRD/SEM analysis
  • Cathode luminescence
• Core Testing
  • Uniaxial compressive strength
  • Thick wall cylinder tests
  • Elastic module tests
  • Differential strain analysis
  • Formation particle size distribution
  • Triaxial tests under simulated in-situ reservoir stress regimes
  • Core flow tests at simulated reservoir conditions
• Log Analysis
  • Dipole sonic logs
  • Sonic travel time
  • Density log

Well Data

• PVT data
• PLT/TDT data
• deviatoric stress analysis (well orientation)
• Water production data
• Sand step rate tests (with in line sand detectors)
• Drawdown and rate analysis
• Well inflow performance and skin analysis

Mohr-Coulomb Compressive Strength Data

Sand Production Analysis

Predictive models are used to assess sand production potential. Typical model input includes: in-situ stresses, reservoir and drawdown pressures, and rock failure criteria which are relevant to the well geometry, rock type and field conditions. These models which have been directly derived from field observations of sanding, are constructed for a variety of completion and well geometries, and for various production and depletion scenarios. Where data exist, the models are tuned and calibrated against observed sand production.

Sand Control Strategies

If analysis indicates a significant sand production potential, or sand production has been observed, then eP can design a completion strategy to minimize the impact. The strategy adopted will depend upon a number of factors including well deviation, rock properties, well and reservoir parameters, and technical and budgetary constraints.

Options to limit, mitigate, or avoid sand production are investigated and compared, such as:

• Oriented and/or selective perforation
• Drawdown policy (establish sand-free drawdown/production rate through modeling calibrated by sand flow tests)
• Formation damage assessment
• Gravel pack/mechanical sand exclusion design and performance analysis

Formation Particle Size Analyses

The various options are evaluated using the well production simulation facility in eP's WellFlo software. In addition, the performance of the selected sand control pack or screen can be directly simulated by core tests in which the production phase is flowed at increasing drawdown pressures. This quantifies potential permeability/productivity damage caused by the physical presence of the sand control and formation fines entrainment.

Advice is also provided on sand monitoring strategies such as:

• Sand detection (e.g. downhole and surface detectors)
• Inflow performance analysis monitoring
• Water detection and shut-off

IPR Analysis for Gravel Pack/Formation Damage

Example from a new well shows log data (including compressional and shear sonic) and predicted rock strength (UCS) from correlation of log and core data. Two models (CDP1 and CDP2) predict critical drawdown pressure as shown on Track 5. CDP2 (originally adopted by the operator) proved to be too conservative and CDP1 provided a better match with observed sanding in other wells. The shaded area is the additional sand-free drawdown that can be obtained using the less pessimistic, field-calibrated model. Zones at risk of sand production under the anticipated production well drawdown conditions are highlighted.

Sand Production Training

As part of our commitment to transfer of effective technology, eP provides short course practical training on sand production, prediction, and prevention. The course has a modular format and can be adapted to meet the particular requirements of the client and participants. Besides focusing on the key issues which affect sand production, the course can be easily expanded to include more fundamental treatment of rock mechanics and formation damage, particularly with respect to laboratory test methods and analysis.

Recent Sand Production Activities

Case # 1 SE Asia, Sand Control Optimization

eP engineers were commissioned to provide the primary focal point for the client's sand control task force for a gas field in SE Asia. Working in the client's offices, in collaboration with his geoscientists, production and well service engineers, eP's staff were responsible for directing data gathering, data acquisition program design and management, sand production analysis, remedial planning and coordination, and follow up of sand control and monitoring strategies.

The field consists of natural gas and condensate accumulations found in multi-faulted reservoirs deposited in a fluvio-deltaic and coastal environment. Co-mingled production was the preferred completion method, with up to 15 perforated intervals in some wells. Many of the deviated development wells had experienced significant sanding problems but sand production could not be identified in the individual sands. The costs of sand production were significant due to loss of production and high maintenance expenditure associated with sand removal, valve and flow line replacement.

Analysis of existing log, core, well, and fluid property information in the client's databases provisionally identified the sands that were considered to be most at risk from sand production, and a log (open and cased-hole) and core data acquisition campaign was designed to improve the understanding of in-situ stresses and rock strength and failure processes. WellFlo models were constructed to determine drawdown at each sand. Excessive drawdowns, high deviatoric wellbore stresses caused by non-optimal alignment of the well paths with the principal reservoir stresses, adverse interaction between the corrosive reservoir fluids and the sands, and high water cuts were all found to contribute to sand production. Sand step rate tests with in-line sand detectors and downhole video logs were carried out to confirm the predictive sanding models. Log-based strength indicators, calibrated by core measurements, were developed and used to identify high risk sands and guide the sand control policies.

Short term intervention strategies for existing wells and longer term sand avoidance and monitoring strategies for future wells were formulated and implemented.

---

Case # 2 UK, Sanding Potential and Sand Exclusion

eP was commissioned to undertake a study of sanding potential and sand exclusion techniques for a proposed horizontal well to be drilled in the North Sea. The Eocene reservoir consists of a thin oil bearing sand sandwiched between a gas cap and an aquifer, and the DSTs had shown a high potential for gas and water coning. The prospect was initially considered to be of limited economic potential, and as a result had remained undeveloped. Recent developments in horizontal drilling technology led to a reassessment of the prospect: feasibility studies and calculations performed by the operating company in 1992/93 concluded that drilling a horizontal well would result in better performance.

A detailed analysis of the various test results led to the following conclusions and recommendations:

1. Elastic failure through a yield zone around the wellbore was predicted over a large range of drawdown conditions. The most probable dominant rupture mode of the reservoir formation was ductile/plastic.
2. The recommended sand control option was based on a series of resin coated sand prepacked screens (PPS) placed along the entire horizontal open hole section.
3. Production simulations indicated that a very high PI may be expected and that over the practical operating range, well performance will be strongly influenced by drawdown/well head pressure and tubular dimensions.
4. Due mostly to the extremely high PI expected for this well only large reductions in gravel or formation permeability will significantly affect well productivity. Such large permeability decreases are entirely possible in horizontal wells, especially if adequate clean up procedures are not implemented.

---

Case # 3 China, Sand Production Study

eP carried out an integrated sand production study for oil and gas fields, offshore China. The project also involved training operating company engineers on the methodologies of sand production evaluation and sand exclusion completion design.

Empirical and analytical models were implemented to predict the critical drawdown pressures that would result in sand production in the development wells. The analytical models indicated that vertical wells would be at risk from sand production. Horizontal wells would be less susceptible since, for the same production rates, the drawdowns would be considerably less. Vertical development wells would therefore require to be gravel packed. Horizontal wells could initially be completed using a slotted liner, which would allow pre-packed screens to be installed later in the field life should sand production occur on reservoir depletion or water influx.

Inflow performance analysis of vertical well gravel-packed completions and horizontal well pre-pack completions indicated that the productivity potential of the horizontal well would be almost 6 times greater than a vertical well in the same formation, and that 99% pre-pack permeability damage could be tolerated in a horizontal well with no significant effect on well productivity.

Given the limited data set available, the study also made recommendations for future log, core, and test data acquisition programs for the initial development wells. The objectives should be to validate, tune, and calibrate the existing sand prediction models and to develop robust field-wide rock property correlations.

---

Case # 4, SE Asia, Sand Production Study

Given that the deliverability of the production wells was not expected to be high, it was essential to establish the risk of sand production and, if sand exclusion methods were to be adopted, the impact that these might have on well productivity.

The available core, well test, log, and mineralogical data from the field was reviewed to assist in the core data acquisition program design. On the basis of the review, a rock strength test program was designed and specified by eP.

A laboratory test program was carried out which comprised:

1. Unconfined compressive strength (UCS) tests
2. Multi-stage triaxial (MST) tests and elastic module
3. Particle size distribution (PSD) tests

The UCS tests were designed to quantify the effects of saturation on unconfined compressive strength, as well as to provide quantitative estimates of rock strength. The MST tests provided quantitative data on the rock failure parameters, cohesive strength and angle of internal friction. These parameters were required in several sand prediction models implemented within the project. Characterization of particle size distributions was essential for the design of possible sand exclusion systems.

The methodology used to predict sand production was based both on empirical and theoretical models. The results from these sand prediction models have been compared and calibrated with the available field data to validate the model predictions. The results indicated that the risk of sanding was low at initial reservoir conditions, but would increase in time with reservoir depletion.

Once the degree of potential sanding had been assessed, it was necessary to design an appropriate completion. This phase involved three major objectives.

1. To perform a wellbore stability assessment for horizontal wells, which led to the determination of maximum drawdown values.
2. To analyze potential completion strategies and to evaluate the performance of different sand exclusion approaches. The analysis identified the use of screen completions as the most suitable, followed by gravel packs.
3. To investigate the degree of under balance required for perforating operations, identifying a range of pressures that would ensure clean perforations without causing sand production.

---

Case # 5, UK, Sanding Potential

The operating company intended to produce the reservoir below the dew point and were concerned that increased reservoir depletion, coupled with high drawdown pressures in the production wellbore might cause sand production. A rock test data acquisition program was also designed to acquire core mineralogical, petrographic, and rock strength data.

Theoretical sand prediction models were used to assess the sanding potential of the well, both at initial reservoir conditions and at final depletion pressures. Despite the differences between the models, the results indicate that the risk of sand production was considered to be low under typical operational conditions. Sand control would not be required at initial conditions provided the production drawdowns were kept below the critical drawdowns indicated by the prediction models, and water influx was prevented.

---

Case # 6, SE Asia , Sand Production Study

eP was commissioned to carry out a sand production study for a gas field development, offshore Bangladesh,

Log data from four exploration and appraisal wells were analyzed and a rock strength core test program was carried out to determine the rock strength characteristics. Field DST data were also provided. The log, test, and core data were integrated to provide an estimate of rock strength in the main reservoir sands. Log indicators were generally used to identify weaker horizons within the reservoir sands, so as to focus the rock testing program

Theoretical sand prediction models were used to assess the sanding potential of vertical production wells, both at initial reservoir conditions and at final reservoir depletion pressures. The theoretical model results require calibration with field data to validate their use in the field. Unfortunately, at the appraisal stage of field development, an absence of suitable field data prohibited a detailed validation of the critical drawdowns predicted by the theoretical models. These all predicted critical drawdown pressures that were significantly higher than the maximum drawdowns achieved in field tests.

On this basis, the risk of sand production under typical operational conditions in the stronger sands was considered to be low and sand control was not recommended. However, some of the reservoir sands are locally weak and the risk of sand production was considered to be higher, especially later in the field life as reservoir pressure depletes and with increased water production. For the weaker sands, selective perforating and mechanical exclusion were recommended as possible sand control options, and the former was actually implemented. The installation of a permanent sand detection system was also recommended, to optimize any future workover and/or sand control installation.

---

Case # 7 UK, Well Stability and Sanding Evaluation

Data from four vertical wells that penetrate the same formations were used to evaluate the potential for sand production in a proposed horizontal well in the Southern North Sea. A rock strength core test program, consisting of multi-state triaxial tests was carried out, to acquire data for sanding prediction modeling. The log, test and core data were integrated to provide an estimate of rock strength in the main reservoir sands.

The Mohr-Coulomb failure parameters and UCS data classified the rock as strong to very strong. However, the orientation of the rock within the in-situ stress field ultimately controls the rocks potential for failure.

Theoretical sand prediction models were used to assess the sanding potential of a horizontal well, both at initial reservoir conditions and at final depletion pressures. The proposed orientation of the borehole within the in-situ stress field results in an extreme deviatoric stress distribution at the wellbore.

The models investigated suggest that at initial reservoir conditions, the risk of shear failure is high, resulting in borehole breakout in the direction of minimum horizontal stress. Drilling overbalance and the presence of a mudcake may reduce the extent of the breakout, particularly if the well is brought onto production slowly. Following initial shear failure, plastic deformation and natural arching leads to a post failure stabilization phenomenon that makes most elastic predictions very conservative.

It is imperative that the required drawdown is obtained through a series of incremental increases over a time period that allows stabilization to occur. It was recommended to complete the well with a slotted liner and a screen, particularly if a subsea completion is used.