Study of Expandable-Tubular Collapse Leads To Risk-Based Strength Development
You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.
Solid expandable technology is helping the industry reach more challenging wells. The objective of this study is to investigate comprehensively the effects of the expansion process on the pipe mechanical properties and collapse performance and to develop a reliable risk-based collapse-strength design for the expandables.
Expansion technology allows for tubulars to expand in situ, thus maintaining downhole size. It has been used successfully for zonal isolation in depleted or trouble zones as a drilling liner. It was also found to be very useful for extended reach in brownfields without losing the tubular diameter for maximum production. Another application is to repair existing casing worn by drilling. Several expansion technologies have been developed since the first concept test was performed in 1993. Currently, an expansion system called top anchor and pull (TAAP) is being developed, and multiple liners have been installed in the Gulf of Mexico.
The TAAP expansion system currently uses 50-ksi-grade expandable seamless tubulars. This material exhibits a large degree of work-hardening and, therefore, is suited for use in expandable tubulars. The collapse strength of tubulars, one of the major concerns during well design, is generally determined by the diameter-to-thickness (D/t) ratio as well as pipe material yield strength and geometric factors such as ovality, eccentricity, and residual stresses. Earlier studies have shown that the expansion process can affect the factors influencing tubular collapse performance significantly.
One of the objectives of this study is to develop a better understanding of collapse-failure mechanisms of expanded tubulars.
The other objective is to develop collapse-strength design formulas and methodologies for expanded tubulars with the same reliability level used currently, taking into account the additional aspects and uncertainties related to expandable tubulars.
Expandable Tubular Collapse Mechanisms
On the basis of a review of the literature, factors such as expanded pipe yield strength, stress/strain curve, geometry factors, and residual stresses, as well as their effects on pipe-collapse performance, have been investigated in the present study by both experimental measurements and numerical finite-element analysis (FEA).
Residual Stresses. The split-ring and hole-drilling methods for pipe residual-stress determination produce approximate or average measurements that do not accurately represent residual-stress distribution through the pipe wall thickness. In this study, through-wall residual stresses in an expanded pipe were measured by X-ray diffraction (XRD).
A 9⅝-in.-outer-diameter (OD), 0.435-in.-thick, 50-ksi-grade expandable pipe was expanded by a 10.1-in. cone under fixed/fixed conditions using the TAAP system. A pipe ring with a length of 12 in. was cut from the expanded pipe and marked on its OD at the thinnest wall area (the area most likely to experience the highest deformation during expansion) as the first location for XRD residual-stress measurement. A second location was marked on the pipe OD for XRD measurement 90° away from the first location in the same circumference plane.
Meanwhile, a 24-in.-long pipe ring was cut next to the 12-in.-long pipe ring from the expanded pipe for standard split-ring-hoop residual-stress measurement.
XRD residual-stress measurements were conducted in the axial and hoop directions at the two marked locations on the OD surface and at two locations on the inner diameter (ID) surface. Sectioning was required to access the ID locations. Before sectioning, a strain gauge was placed on the two ID locations to monitor the stress relaxation from sectioning. The ID residual-stress measurements were corrected for any stress relaxation from sectioning.
Regarding standard split-ring residual-stress measurements, the results revealed relatively low residual stress values. The main reason could be that the actual hoop residual stress is not distributed linearly through the pipe wall thickness. The pipe ring is also seen closing because of tensile hoop residual stress at the pipe ID, as shown in Fig. 1 above.
Yield Strength and Stress/Strain Curves. Regular round-bar tension tests were conducted on expanded and unexpanded 50-ksi-grade expandable pipes. The typical unexpanded 50-ksi-grade expandable pipe tensile stress/strain curves were recorded in longitudinal and transverse directions. No significant difference of stress/strain curve exists between the longitudinal and transverse directions for unexpanded pipes.
A laser-speckle strain-measurement system was used to measure pipe stress/strain curve under tension and followed by a compression load to study the Bauschinger effect. Previous experiences indicated a significant buckling on a regular-/standard-gauge-length tensile bar when conducting the compression test. In this study, a reduced-gauge-length tensile bar was used that overcame the buckling problems encountered with normal-gauge-length specimens.
FEA of Expanded-Pipe Collapse Strength. FEA was conducted to study the effects of residual stresses and post-expansion material-property changes on pipe-collapse performance. To simplify the modeling process, a two-dimensional (2D) finite-element model was used.
The pipe-expansion process was not simulated. Instead, a 2D post-expanded-pipe model was considered with a post-expansion compressive stress/strain curve for the material properties and linearly distributed through-wall residual hoop stress.
In the FEA, several scenarios were considered, such as pipe with different D/t ratios, pipe with and without residual stress, and pipe with material properties such as post-expansion in the hoop-compression direction and pre-expansion in the axial-tension direction. Simulated collapse pressures were compared with actual tested collapse pressures of expanded pipes and regular pipes.
The FEA study revealed that, in general, pipe experiencing the Bauschinger effect (i.e., expanded pipe) has a significantly lower collapse pressure than pipe without the Bauschinger effect (i.e., regular pipe) under the same D/t ratio. The large amount of hoop residual stress in the expanded pipe apparently does not affect collapse resistance significantly, which is different than what is seen in normal pipes.
Similarly, results showed that the hoop residual stress does not strongly affect the collapse strengths of expanded pipe at other D/t ratios (when using the post-hoop-compression stress/strain curves) either.
Pipe-Geometry Dimensional Mapping. A pipe-geometry study was conducted by use of ultrasonic sensors for continuous measurement on pipe-wall thickness and a laser flash for continuous measurement on pipe OD in a helical pattern while rotating the pipe.
The pipe specimen was a 9⅝-in.-OD×0.435-in.-wall-thickness 50-ksi-grade expandable pipe. The OD and wall thickness of the pipes were measured before and after expansion. The expansion was under fixed/fixed mechanical expansion mode using a 10.5-in.-diameter cone.
OD measurements for the 9⅝-in. pipe before and after expansion reveal that the TAAP mechanical-expansion process does not change the pipe OD variations significantly. Plotting a histogram on those measurements revealed that the OD variations are described reasonably well by normal distribution.
The pipe-wall-thickness measurements show similar results to the OD measurements. The wall-thickness variations also reasonably follow normal distribution.
The ovality and eccentricity of the pipe, important factors affecting pipe collapse in addition to OD and thickness, also have been studied. Neither changed significantly after 20% expansion.
Risk-Based Statistical Analysis
Historical Data. Collapse tests on 50-ksi-grade expandable pipe have been conducted during the past 25 years with various cone designs, various expansion ratios, and various expansion systems and expansion modes. The expanded 50-ksi-grade expandable pipes have much lower collapse pressures than average 50-ksi-grade pipes. In other words, expansion process deteriorates the pipe-collapse performance.
Collapse-Testing Data. Because strain-aged pipe is not regular expanded pipe, it was not included in the statistical analysis for collapse-strength-formula development. Rotationally expanded pipe is not used currently, so those data are not included in the analysis, either. Additional collapse tests were conducted. Some 50-ksi-grade expandable pipes were expanded with different expansion ratios and various mechanical-expansion modes.
Experimental testing, statistical analysis, and fundamental modeling of the collapse performance of 50-ksi-grade expandable pipe led to the following conclusions:
- The Bauschinger effect after expansion, which significantly reduces the yield strength and Young’s modulus in the pipe hoop direction, is the main reason expanded pipes have lower collapse performance than regular unexpanded pipes at the same D/t.
- The geometry of the pipe (e.g., OD, thickness, ovality, and eccentricity) does not have significant variations after expansion by the TAAP expansion system, as indicated by pipe dimensional-scanning techniques.
- The very high residual stresses presented in the pipe after expansion do not appear to affect the collapse performance of the expanded pipes significantly when material has a significant Bauschinger effect.
- Risk-based design-collapse equations have been developed through statistical analysis.
Study of Expandable-Tubular Collapse Leads To Risk-Based Strength Development
01 June 2018
Reliability-Based Casing Design Unlocks Reserves in a High-Pressure Gas Field
This paper presents work performed to characterize the probability of collapse as a function of reservoir-abandonment pressure using reliability-based design (RBD).
Risk-Based Statistical Approach To Predict Casing Leaks
A probabilistic approach was introduced to improve interpretation of data from electromagnetic corrosion logs. This paper presents a statistical risk-based approach to predicting casing leaks using electromagnetic corrosion logs.
Risk-based design is very important, especially for complex systems, because a deterministic approach may not provide a strong underpinning to the reliability estimation of the downhole tools. We have to reduce the probability of failure.
Don't miss out on the latest technology delivered to your email weekly. Sign up for the JPT newsletter. If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.
12 June 2018