SPE Copyright 2012, Society of Petroleum Engineers

SPE 155037 Formation-Tester Pulse Testing in Tight Formations (Shales and Heavy Oil): Where Wellbore Storage Effects Favor the Determination of Reservoir Pressure Hamid Hadibeik, The University of Texas at Austin; Mark Proett, Dingding Chen, and Sami Eyuboglu, Halliburton Energy Services; Carlos Torres-Verdín and Rooholah A. Pour, The University of Texas at Austin Copyright 01, Society of Petroleum Engineers This paper was prepared for presentation at the SPE Americas Unconventional Resources Conference, Pittsburgh, Pennsylvania, USA, 5 7 June 01. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright. Abstract Tight formation testing when mobilities are lower than 0.01 md/cp poses significant challenges because the conventional pressure transient buildup testing becomes impractical as a result of the large buildup stabilization time. This paper introduces a new automated pulse test method for testing in tight formations that significantly reduces testing time and makes the determination of formation pressure and permeability possible. A pulse test is defined as a drawdown followed by an injection test, and the source is shut in to record the pressure transient. Based on pressure data during the shut-in period, the next drawdown or injection test is designed, such that the flow rate is a fraction of the initial pulse rate, followed by another shut-in test. This procedure continues until the difference in pressure at the beginning and at the end of the shut-in period is reduced to within a specified limit of pressure change; then, an extended transient is recorded to a stabilized shut-in pressure. The overall advantage is to reduce the pressure stabilization time by implementing an adaptive pressure feedback loop in the system. The method can be applied to a straddle packer test using conventional drillstem testing tools or formation testers, using either straddle packers or probes. The effects of wellbore storage and fluid compressibility are found to reduce the pressure drop and positive pressure pulse in the drawdown and injection tests, respectively; they also affect the decay rate to the asymptote of the shut-in pressure response. Consequently, the combined pulse test method with the pressure feedback system and wellbore storage effect reduces the reservoir pressure testing time in tight formations. The automated pulse-test method has been successfully validated with consideration of the effects of wellbore storage and overbalance pressure in tight gas and heavy oil formations. In addition, the effects of invasion with water- and oil-based mud filtrate were considered in the modeling. The method uses successive pressure feedbacks and automated pulses to yield a pressure to within 0.5% range of the initial reservoir pressure while decreasing the wait time by a factor of 10 for a packer type formation tester. To account for various tool options and storage effects, the packer-type, oval probe, and standard probe-type formation testers have been simulated in various tight formation conditions. The method enables a rapid appraisal of pressure measurements in comparison to conventional testing. Simulations also indicate that the analytical spherical model can be used to analyze a pulse test, even when encountering multi-phase compositional fluid effects. Introduction Important reservoir properties, such as formation pressure and permeability, can be measured with formation-testers (Angeles et al. 010; Proett et al. 004; Zazovsky et al. 005; Elshahawi et al. 1999). Hadibeik et al. (009 and 010) tested fluid sampling by means of various probe-type formation testers in laminated reservoirs under the influence of dynamic mudfiltrate invasion. The next advancement in the transient analysis of formation-tester measurements is the consideration of the effect of flowline storage or wellbore storage, which significantly slows down the pressure transient in low mobility formations (Proett et al. 1998; Goode et al. 1987; Yildiz et al. 1991; Chin et al. 007). Consequently, tight formation testing poses significant challenges when using the conventional drawdown buildup method. Another complication for testing in tight formations is that the measured pressure is supercharged and is greater than the reservoir pressure. The measured shut-in pressure is usually assumed to be the formation pressure. In a permeable formation, mudcake can form quickly and is normally very effective in slowing down invasion and maintaining the wellbore sandface

SPE 155037 pressure to near that of the formation pressure. However, this assumption is unrealistic, especially in low mobility formations in which there could be no sealing mudcake to isolate the reservoir from hydrostatic pressure. In tight formations, the invasion rate is slowed by the formation, and mudcake may form slowly or it may not exist (Proett et al. 000; Schrooten et al. 007; Britt et al. 004). Therefore, the measured pressure in these cases is substantially greater than the formation pressure as a result of the lack of sealing mudcake. There have been some attempts to determine the reservoir pressure in tight sand formations (Yang et al. 007; Proett et al. 1994). To solve this challenge, a pulse test technique is used in combination with the spherical flow model to obtain the true formation pressure. Early transient and spherical flow period can be distinguished from late transient flow period by calculating the streamline evolution of flow near the source (Hadibeik et al. 011). To measure the reservoir pressure in tight-formation testing, current technologies require several hours of wait time to reach the stabilized pressure, and it is often difficult to reach this stabilization or to determine when it occurs (Proett et al. 1996). The automated harmonic testing method introduced in this paper substantially reduces the testing time when using a packer or probe-type formation tester and automatically converges to the shut-in pressure. Problem Statement There are two main practical difficulties in controlling a low mobility test. First, although drawdowns are very fast, the pressure buildup is very slow, and might require several hours or days for a buildup to stabilize. A pressure transient near the stable shut-in pressure is required to identify infinitely-acting Darcy flow and to interpret formation pressure and permeability (Fig. 1). Before infinitely-acting Darcy flow is identified, the pressure transient is dominated by flowline storage and skin. This causes the drawdown to decline rapidly but buildups to increase slowly that delays the stable infinitely-acting Darcy flow period. Drawdown pressure control is difficult because of its rapid declining rate. It is not unusual to observe fluid property changes, or even phase changes, which make the buildup stabilization pressure even more difficult or, in many cases, impossible to achieve. Second, the mud-filtrate invasion in these formations causes the supercharged effect to occur, which means that the stabilized pressure is different from the true initial reservoir pressure. In wells that are overbalanced, the stabilized wellbore pressure is supercharged as a result of the greater hydrostatic pressure. In wells that are underbalanced, the wellbore pressure will be near the hydrostatic pressure, but the formation pressure is actually greater. Therefore, depending on the drilling situation (overbalanced or underbalanced drilling), this pressure will be greater or less than the initial reservoir pressure. In this paper, the automated pulse-test technique and a pressure feedback address these testing issues. Flowline volume - storage Probe isolation valve Reduced test time Slow buildup Fig. 1 Formation testing in low mobility reservoirs results in a slow buildup. The use of a shut-in valve can decrease the stabilization wait time; however, field data have shown that even with a shut-in valve, the wait time can be significant in micro- and nano-darcy formations. Numerical Model for Pulse-Test Method The numerical model used in this study is a compositional multi-phase near wellbore finite-difference model (Pour 011). To account for the flowline storage effect, the first gridblock adjacent to the source is assigned a large permeability and porosity value to perform as a hollow chamber rather than a porous medium. The thickness of the gridblock is adjusted to account for the volume of fluid present in the flowline. The model was developed for the straddle packer, an oval probe, and a single probe. An analytical single-phase spherical model (Proett et al. 1996) is then used to validate these models. Proett et al. (1998) presented this spherical flow model with the wellbore storage effect. The first step is to validate the numerical model with the analytical model for the single-phase flow in a tight sand oil reservoir invaded with an oil-based mud filtrate. The next step is to implement the wellbore storage effect in the model. Although the numerical method was validated using a single-phase fluid, it can be used to model the multi-phase flow, especially tight gas sands invaded with water-based mud (WBM) or oil-based mud (OBM).

SPE 155037 3 Without Wellbore Storage Effect. A homogenous isotropic tight formation is used as a base case model to validate the numerical method without wellbore storage effect. The properties of this formation are shown in Table 1. We assume that there is no mud-filtrate invasion; consequently, there is no supercharge effect, and the reservoir is a single-phase oil reservoir. Tables through 4 summarize the tester properties, and Figs., 3, and 4 summarize the results of the validation of the numerical method. In these examples, the pulse test consists of a drawdown followed by an injection of equal flowrate and pulse-time period. As shown in Figs., 3, and 4, the analytical model indicated by the blue curve matches the numerical model indicated by the red curve very closely. The differences between the two models can be accounted for in that the numerical model contains all geometry of the wellbore and probes, whereas the spherical model assumes an equivalent probe diameter in a purely mathematical spherical space. TABLE 1 SYNTHETIC TIGHT SAND RESERVOIR MODEL. Formation properties Permeability[μD] 1 Anisotropy (k v /k h ) 1.0 Porosity [%] 10 Fluid viscosity [cp] 1.0 Fluid compressibility [psi -1 ] 3.0x10-6 Formation pressure [psi] 0,000. x 10 4 Numerical Model Analytical Model TABLE STRADDLE PACKER, TESTER PROPERTIES. Properties Flow line fluid compressibility [psi -1 ] 3.0x10-6 Flow line volume [bbl] 0.0 1.6 Wellbore diameter [ft] - ([in.]) 0.71 (8.5) Packer height [ft] 4. Packer equivalent probe radius [in.] 5 Pulse rate [bbl/d] -([cc/sec]) 1.0 (4) 1.4 0 4 6 8 10 Fig. Comparison of the numerical method to the analytical model for a straddle packer. This result is used to validate the numerical parameters used in the simulations..4 x 104. 1.6 Numerical Model Analytical Model 1.4 0 4 6 8 10 Fig. 3 Comparison of the numerical method to analytical model for an oval probe formation tester. There is a close match between the two models when the wellbore storage is zero. Pulse time [sec] 30 The flow line volume is zero to account for zero wellbore storage. TABLE 3 OVAL PROBE, TESTER PROPERTIES. Properties Flow line fluid compressibility [psi -1 ] 3.0x10-6 Flow line volume [bbl] 0.0 Wellbore diameter [ft] 0.71 Oval probe height [in.] 9 Oval probe width [in.] 1.75 Oval probe radius [in.].04 Probe geometric factor 1. Oval probe equivalent radius [in.] 1.49 Pulse rate [bbl/d]-([cc/sec]) 0.009 (0.0184) Pulse time [sec] 30 The flow line volume is zero. The numerical method uses the dimensions of the probe to simulate the measurements; however, the analytical method uses the oval probe equivalent radius with the probe geometric factor.

4 SPE 155037.3 x 104..1 Numerical Model Analytical Model TABLE 4 STANDARD PROBE, TESTER PROPERTIES. Properties Flow line fluid compressibility [psi -1 ] 3.0x10-6 Flow line volume [bbl] 0.0 Flow line shut-in volume 0.0 Well bore diameter [ft] 0.71 (8.5) Probe radius [in.] 0.38 1.7 0 4 6 8 10 Fig. 4 Comparison of the numerical method to analytical model for the standard-probe formation tester. Because the effect of wellbore storage is subsided, the amplitude of the positive pulse is the same as the negative pulse. Probe geometric factor 0.6 Probe equivalent radius [in.] 0.39 Pulse rate [bbl/d] - ([cc/sec]) 0.0011 (0.00) Pulse time [sec] 30 The analytical model uses the probe equivalent radius and geometric factor to simulate the measurements. Although the pulse rate changes with respect to the probe size and shape, the pulse duration remains constant to evaluate the source type effect. With Wellbore Storage. The effect of wellbore storage must be included in testing models, particularly in tight formations in which the storage effect can dominate most of the pressure transient data. The storage boundary effect consists of the fluid volume and compressibility connected to the formation. The analytical method described in Proett et al. (1998) includes the storage effect in the boundary value problem definition. In the numerical model, the storage effect is simulated by using the first gridblock adjacent to the wellbore and assigning it a large permeability and porosity values to act as the storage fluid volume, which is inherently connected to the porous medium grid. The reservoir properties and tester properties remain the same as before, but now includes wellbore storage. Table 5 summarizes the amount of storage for each tester. Figs. 5, 6, 7, and 8 display the validation models for the straddle packer, oval probe, standard probe, and DPS (dual probe section) of the formation-testing tool. TABLE 5 FLOWLINE VOLUME (STORAGE EFFECT) OF FORMATION TESTERS. Type of Formation Tester Volume [cc] Straddle packer 37,000 Oval probe 50 Standard probe 160 In these examples, the storage effect clearly changes the pressure transient behavior. First, both the drawdown and injection pressure pulse magnitudes are reduced. Most notably, the positive pulse overshot pressure is now reduced from several thousand psi to less than a few hundred psi. Because of this, the shut-in pressure appears to be reached in less time. In addition, the observed reduction in the positive pulse is much smaller in the straddle packer than in the probe cases and for the lower permeability cases shown in Fig. 8. Therefore, for the given cases, storage appears to actually help to reduce the testing time with this simple pulse test method. However, it may become more complicated when wellbore effects, such as an over- or underbalanced wellbore hydrostatic mud are considered, and further refinements are needed to the pulse test method.

SPE 155037 5 x 10 4.05 x 104 Numerical Model Analytical Model 9 8 5 7 Numerical Model Analytical Model 6 0 5 10 15 Fig. 5 Validation of the numerical method with the analytical model for a straddle packer with storage effect. The effect of storage tends to reduce the positive pulse overshot in pressure resulting from the injection segment of the pulse test.. x 104.1 1.7 1.6 Numerical Model Analytical Model 1.5 0 5 10 15 Fig. 7 Validation of the numerical method with the analytical model for the standard probe with storage effect. Pressure overshot and drawdown amplitude increase when the size of the probe becomes smaller with less storage. 5 0 5 10 15 Fig. 6 Validation of the numerical method with the analytical model for an oval probe with storage effect. The pressure response of a test with storage effect stabilizes the reservoir pressure more quickly than the case without storage..05 x 104 5 K = 1 μ D, Numerical 5 K = 0.1 μ D, Numerical K = 1 μ D, Analytical K = 0.1 μ D, Analytical 0 4 6 8 Fig. 8 Validation of the numerical method with the analytical for standard probe formation testing tool for two different tight sand reservoirs. New Optimized Pulse Testing Method Method and Algorithm. Previous studies of a pulse test have shown some advantages over traditional drawdown buildup testing methods (Proett et al. 000); however, the previous pulse testing methods used a continuous harmonic pulse to determine permeability or anisotropy and did not necessarily improve testing in tight reservoirs. After the pulsing stopped, the stabilization time would not be any faster for tight formations (k < 0.001 md). However, if the pulse is varied, then it may be possible to further reduce the stabilization time. Some comparisons of pulse types are shown in Fig. 9 for a standard probe testing in a tight sand formation, and Table 6 summarizes the fluid and reservoir properties for an oil reservoir invaded with one day of OBM filtrate invasion. In the example shown, the injection pulse with half the volume of the drawdown improves the stabilization time, as compared to a standard drawdown buildup and a drawdown and injection pulse of the same volume and flow rate. The examples shown in Fig. 9 suggest that a pulse method could be further improved by using optimization techniques. This possibility led to the discovery of using an ongoing pressure transient feedback loop to monitor the pressure changes and determine the next action which could be a drawdown or an injection at a volume and rate designed to minimize testing time. This method can begin from a drawdown or injection test, depending on the drilling situation (overbalanced or underbalanced drilling). Consequently, a shut-in test (buildup or builddown test) is performed, and a pressure gauge records the pressure during the shut-in period. If the pressure decreases, the next move should be a drawdown test; otherwise, if the pressure

6 SPE 155037 increases, the next choice should be an injection test. The rate of pressure increase or decrease (i.e., slope or derivative) can also be used to design the new pulse. The simplest way to monitor these short shut-in periods is to observe the start pressure and ending pressures of the shut-in, which is used here for simplicity. However, real data can be noisy, and it is desirable to use all of the data to determine the slope or derivative. A logic flow diagram of the algorithm of this automated pulse test technique is shown in Fig. 10. Fig. 11 depicts a pressure response of the automated pulse test method obtained from a straddle packer, and Table 7 presents its tester properties. TABLE 6 RESERVOIR AND STANDARD PROBE TESTER PROPERTIES FOR AN OIL RESERVOIR INVADED WITH 1-DAY OF OIL-BASED MUD FILTRATE INVASION. Properties k [μd] 1.0 φ [%] 10 c t [psi -1 ] 3x10-6 μ [cp] 1.0 V fl [cc] 160 Q [cc/min] 0.55 Flow line volume (V fl ) and pumpout flow (Q) rate was chosen based on tool design and the amount of pressure drop created in the test..15 x 104.1.05 5 5 1.75 1.7 = ½ Q - Simple DD 00 400 600 800 1000 BD T = 65.8 min BD T = 31.1 min BD T = 153.5 min Time [sec] Fig. 9 Comparison of a conventional drawdown test (DD) with a single pulse test ( =Q - ) and a test that combines a drawdown and a half-injection test ( =1/Q - ). The reservoir pressure is 0,000 psi, and the overbalance pressure is 1000 psi. Mud filtrate invasion stops during the test. A single pulse test is not sufficient to create a fast builddown stabilization time. Start of Test Perform a Pulse Test.15 x 104.1 Pumpout rate decreased relative to first DD.05 DD: Drawdown INJ: Injection SH: Shut-in N : Number of pulses performed Perform a Buildup Record Start Pressure (P 1 ) Record Final Pressure (P ) Yes Perform a Drawdown is P <P 1? No T : Default Setting of Number of Loops Perform an Injection 5 5 1.75...... 1.7 100 00 300 400 500 600 700 800 900 1000 DD DD Time [sec] INJ SH SH INJ SH INJ Final SH First Pulse End the task if N > T Perform Final Buildup Fig. 10 Automated pulse test algorithm flow diagram with pressure feedback. The number of repetitions of the feedback pressure loop can be set to a default number or determined when the pressure change is within a bound criterion, which stops the iterations.. Fig. 11 Automated pulse test example. The initial reservoir pressure is 0,000 psi, the overbalance pressure is 1,000 psi, and during the test, the invasion is stopped. A complete pulse test was performed before starting the automation, which can be removed in the actual test. Each section of the test (injection or pumpout) is followed by a shut-in period. The pulse durations and pumpout rate direction (drawdown or injection) can be determined automatically. In this example, the flow rate of each new action is half of its previous action, and the injection or drawdown test s pulse duration is 30 seconds, followed by a 60- second shut-in period. It is also possible to optimize the flowrate selection based on the optimization technique.

SPE 155037 7 TABLE 7 RESERVOIR AND TESTER PROPERTIES FOR A STRADDLE PACKER TO PERFORM THE AUTOMATED PULSE TEST. Properties k [μd] 1.0 φ [%] 10 c t [psi -1 ] 3x10-6 μ [cp] 1.0 V fl [cc] 160 Q [cc/min] 0.55 In the previous example, the assumption is made that invasion stops during the test. However, when mud-filtrate invasion continues in this overbalanced situation, then the wellbore supercharge becomes a factor to be considered. Fig. 1 illustrates the numerical simulation results, taking into account the effect of invasion during the test for the straddle packer with properties listed in Table 7. Mud-filtrate invasion is simulated before sealing the packer to the borehole, and then the invasion is stopped in this case. Two pulse tests are simulated in Fig. 1. The first test is a simple pulse test with identical drawdown and injection volumes and rates. The second test is based on the automated pulse test algorithm. The automated pulse test reaches a stable pressure much more quickly than the simple pulse test, but both converge to a supercharge pressure that is greater than the formation pressure. This is also the case with a standard drawdown buildup, but the stable pressure is obtained much more quickly, particularly for the automated pulse test..15 x 104.1.05 5 5 Difference 1.75 0 0. 0.4 0.6 0.8 1 Fig. 1 Automated pulse testing method was performed in a tight sand reservoir with the supercharge effect. The invasion causes the stabilized pressure to be different from the true reservoir pressure. This difference depends on the formation pressure, overbalance pressure, and formation properties. The green line indicates a pressure gauge response of a probe set on the borehole above the packer to monitor bottomhole pressure changes. Because there is no mudcake, the borehole and probe pressure are nearly identical. The red curve indicates a single pulse test. A comparison of the automated pulse test (blue curve) and a single pulse test (red curve) demonstrates that the stabilized pressure is achieved more quickly in the automated pulse test. However, the stabilized pressure is still supercharged more than 500 psi above the formation pressure. Supercharge Effects on Pulse Testing Overbalance pressure makes the pressure stabilize at a greater pressure than the formation pressure as a result of the leaks around the straddle packer. This is also the case for a probe-type formation tester, but the probe provides even less protection from supercharging. This supercharging effect also causes the pulses to be effected in a very different way than if there was no overbalance. The initial drawdown pulse can be considered in two parts. In the initial stage of the drawdown, little if any fluid is withdrawn from the formation and is simply decompressed by the pretest piston movement. The process can be described as if there is an overbalance; then part of the drawdown is above a stable pressure and, until the pressure falls below a threshold, no fluid is withdrawn from the formation. Consider the example of a 1000 psi overbalance with a flow rate of 0.1 cc/sec. After the straddle packers are set, the pressure will tend to slowly decline toward a stable pressure. If we have flowline storage of 00 cc and a compressibility of 3e-6 psi -1, then it will require approximately 6 seconds to reduce the flowline pressure to the supercharge pressure of 500 psi above the formation pressure. Therefore, the decompression volume is: 6 1 V = 00( cc) 3 e ( psi ) 1000( psi) = 0.6cc,... (1)

8 SPE 155037 and the decompression time can be calculated: V 0.6( cc) T = = = 6sec Q cc 0.1 sec... () This decompression time must be subtracted from the drawdown production time. Therefore, if the pulse is specified as 30 seconds, the 6-second section is non-productive, leaving a remaining 4 seconds for a drawdown below the supercharge pressure. Then, on the positive pulse side, a full 30 seconds is applied. The implication of this is that the pressure rise when injecting will be greater than predicted when there is no overbalance. In larger permeabilities, the pressure increase is lower which is due in part to the fact that higher permeable formations have a sealing mudcake. If there is no permeability, then the pressure will increase back to the hydrostatic pressure because the fluid is only being compressed and decompressed. Figs. 13 and 14 depict the pressure response of a dual packer and an oval probe in a single pulse test. Tables 1,, 3, and 5 summarize the formation and these tester properties. These figures indicate that if the injection portion of the pulse test can be optimized, it is possible to reach a fast stabilization pressure with a single pulse..15 x 104.1.05 5, BD Stab. T > 6.5 hrs = 1/ Q -, BD Stab. T = 0.47 hrs = 1/4 Q -, BD Stab. T = 1.7 hrs 5 10 15 Fig. 13 Effect of overbalance pressure and mud filtrate invasion on the pulse test method for the straddle packertype formation tester. The test begins with a drawdown followed by an injection. The stabilization time recorded for this case indicates that the half-pulse rate injection reaches the stabilized pressure more quickly. x 10 4..1 1.7, BD Stab. T > 1.67 hrs 1.6 = 3/4 Q -, BD Stab. T = 0.41 hrs 1.5 = 1/ Q -, BD Stab. T = 1.01 hrs = 1/4 Q -, BD Stab. T > 1.67 hrs 1.4 5 10 15 Fig. 14 Effect of overbalance pressure and variation of pulses on an oval probe-type formation tester. In this case, when a ¾ factor is used to reduce the pulse flow rate, the stabilization time is the lowest of all cases; however, this may apply only to this particular set of conditions. Automated Pulse Test Method The automated pulse test technique is based on implementing a pressure monitoring feedback and control system in which the pulse flow rates and durations can be software-controlled by the testing tool. Figs. 10 and 11 illustrate the algorithm and feature an automated pulse test; the following simulations show a comparison of the automated pulse testing method to other methods (i.e., standard drawdown buildup, drawdown with identical injection volume, and drawdown with half-injection volume). Table 8 shows the properties of three different tight sand formations (i.e., K 1, K and K 3 ) where the straddle packer, oval probe, and standard probe were tested. Table through Table 5 also disclose the operating parameters of these testers. Figs. 15, 16, and 17 present the results of a straddle packer in tight formations with four different tests. In Figs. 18, 19, and 0, show a comparison of the four sets of pressure tests for the oval probe-type formation tester. Figs. 1,, and 3 illustrate the comparison results for the standard probe. In all but one case, the automated pulse test method obtains a stable shut-in pressure more quickly that the other test types. In Fig. 15, the half-injection pulse test reaches the formation pressure faster than the automated pulse test. However, this observation is not general and is only effective in this case because the half-injection pulse is arbitrary. In all of the results, the time in which the pressure reaches the 0.35% of the initial formation pressure is recorded as the stabilized pressure. In the automated pulse test, the flow rate of each test is half of its preceding test flow rate. It was mentioned that this ratio can be optimized based on the pressure feedback, however. The number of iterations between pumpout and injection tests is defined as six iterations, followed by an extended shut-in test to determine the stabilized pressure.

SPE 155037 9 TABLE 8 PROPERTIES OF THREE SYNTHETIC HOMOGENOUS RESERVOIRS WITH DIFFERENT PERMEABILITIES. Properties φ [%] 10 μ [cp] 1.0 c t [psi -1 ] 3x10-6 K 1 [μd] 1.0 K [μd] 0.1 K 3 [μd] 0.01.1 x 104.1.08.06.04.0 Drawdown, BD Stab. T > 1 hrs, BD Stab. T > 1 hrs = 1/ Q -, BD Stab. T = 31.77 min Automated, BD Stab. T = 39.8 min.15 x 104.1.05 Drawdown, BD Stab. T > 1 hrs, BD Stab. T > 1 hrs = 1/ Q -, BD Stab. T = 10.1 hrs Automated, BD Stab. T = 6.48 min 8 5 6 4 4 6 8 10 1 14 16 Fig. 15 Comparison of four different tests to determine the initial reservoir pressure for the first synthetic case (k 1 = 1.0 μd). The time is recorded that the pressure falls within a bound, which is 0.35% of the initial pressure. The reservoir pressure is 0,000 psi, and the overbalance is 1,000 psi. A simple drawdown or a single pulse test ( =Q - ) does not reach the formation pressure in the previously mentioned bound range within 1 hours of shut-in time. For the pulse, when its injection rate ( ) is half of its pumpout rate (1/Q - ), the time is 31.77 minutes; for an automated pulse test, it is 39.8 minutes. 4 6 8 10 1 14 16 Fig. 16 Comparison of four test procedures to determine the reservoir pressure for the straddle packer in the second synthetic reservoir (k = 0.1 μd). Although a drawdown test and a pulse test ( =Q - ) do not reach to the 0.35% limit of the reservoir pressure, it requires 10.1 hours for a half-pulse injection test ( =1/Q - ) to reach this point. However, it requires only 6.48 minutes for the automated method to reach this level..15 x 104.1 Drawdown, BD Stab. T > 1 hrs, BD Stab. T > 1 hrs = 1/ Q -, BD Stab. T > 1 hrs Automated, BD Stab. T = 0.30 min.15 x 104.1 Drawdown, BD Stab. T = 130.9 min, BD Stab. T = 53.7 min = 1/ Q -, BD Stab. T = 16.67 min Automated, BD Stab. T = 13.17 min.05.05 5 5 4 6 8 10 1 14 16 Fig. 17 Comparison of four test procedures to reach the stabilization pressure for the third synthetic reservoir (K 3 = 0.01 μd). Of these tests, only the automated pulse test reaches the 0.35% boundary of the initial reservoir pressure. 10 0 30 40 50 Fig. 18 Comparison of four test procedures for an oval probe, testing the first synthetic reservoir (k 1 = 1.0 μd). In this case, the automated pulse test yields the formation pressure more quickly than other procedures, and approximately 10 times faster than that of a drawdown test.

10 SPE 155037 x 10 4.1.05 Drawdown, BD Stab. T = 187.0 min, BD Stab. T = 74.7 min = 1/ Q -, BD Stab. T = 69 min Automated, BD Stab. T = 9.30 min.15 x 104.1.05 Drawdown, BD Stab. T > 5 hrs, BD Stab. T > 5 hrs = 1/ Q -, BD Stab. T > 5 hrs Automated, BD Stab. T = 1.5 min 5 5 10 0 30 40 50 Fig.19 Comparison of four test procedures for the second synthetic reservoir (k = 0.1 μd) for an oval probe tester. The automated pulse test technique successfully yields the stabilized pressure. Its stabilization time is approximately 0 times, 8 times, and 7 times less than that of a drawdown test, pulse test, and half-injection test..15 x 104.1.05 5 10 0 30 40 50 Fig. 0 Comparison of four different test procedures for an oval probe-type formation tester for the third synthetic reservoir (k 3 = 0.01 μd). In this reservoir, only the automated pulse test is able to locate the initial reservoir pressure in its 0.35% limit within a reasonable testing time..15 x 104.1.05 Drawdown, BD Stab. T > 5 hrs, BD Stab. T > 5 hrs = 1/ Q -, BD Stab. T > 5 hrs Automated, BD Stab. T = 15.8 min 5 5 Drawdown, BD Stab. T = 153.5 min, BD Stab. T = 65.8 min 1.75 = 1/ Q -, BD Stab. T = 31.1 min Automated, BD Stab. T = 11.74 min 1.7 10 0 30 40 50 Fig. 1 Results of four types of formation testing for the first synthetic tight sand reservoir (k 1 = 1.0 μd) to obtain the reservoir pressure with a standard probe. The automated pulse test technique reaches the stabilization pressure in 11.74 minutes..15 x 104 5 5 1.75 1.7 5 10 15 0 5 30 35 40 45 50 Fig. Results of four types of formation testing for the second synthetic tight sand reservoir (k = 0.1 μd) to obtain the reservoir pressure with a standard probe. In this case, only the automated method is able to reach the stabilization pressure..1.05 5 5 Drawdown, BD Stab. T > 5 hrs, BD Stab. T > 5 hrs 1.75 = 1/ Q -, BD Stab. T > 5 hrs Automated, BD Stab. T = 10.89 min 1.7 10 0 30 40 50 Fig. 3 Results of four types of formation testing for the third synthetic tight sand reservoir (k 3 = 0.01 μd) to obtain the reservoir pressure with a standard probe. The automated pulse technique is a non-linear search optimization; this is the reason that the stabilization time in this case is less than the previous synthetic case. Q -Q ½ Q -¼ Q 1 / 8 Q - 1 / 16 Q 1 / 3 Q Fig. 4 Schematic showing a typical harmonic pulse test. The harmonic nature of the positive and negative pulses is designed to quickly approach a stabilized pressure by systematically reducing the flow rate of eachpulse.

SPE 155037 11.15 x 104.1 Harmonic method Automated method.05 5 4 6 8 Fig. 5 Comparison of the automated pulse test and a harmonic test for the straddle packer in the first synthetic case reservoir. The stabilization time of the automated pulse test is faster than that of the harmonic pulse test. The harmonic pulse test biases the stabilized pressure and perturbs the pressure without any information (feedback) from the reservoir. Automated Pulse-Test vs. Harmonic Pulse-Test Method. The harmonic pulse test consists of a series of drawdown and injection tests. After each step, the flow rate is decreased by a factor. However, there is no feedback control in the system to monitor the pressure. Fig. 4 shows a typical harmonic pulse test, and Fig. 5 depicts the comparison of a harmonic pulse test with an automated one. The results show that the stabilization time of the automated pulse test is faster than that of the harmonic pulse test. Compositional Oil and Gas Modeling. The analytical method was used to validate the numerical models. However, to understand the change of fluid properties with temperature and pressure, a numerical simulation is needed, particularly for low mobility reservoir where changes in pressure and temperature are more pronounced in the pumpout process. In practice, the fluid property change during a test as a result of invasion and pressure / temperature variations can be evaluated using compositional equation of state (EOS) modeling. For the rest of reservoir testing simulations straddle packer was used, and the properties of the tester are the same as those shown in Table and Table 5. Overbalanced Drilling and Invasion. With numerical modeling the effects of overbalanced drilling are also considered. In the cases that follow, the reservoir pressure is 0,000 psi, and the overbalanced pressure is 1,000 psi greater than the formation pressure. The type of mud is an OBM or a WBM, and the filtration viscosity is constant (μ = 1 cp). The invasion time is 1-day, and the pumpout and injection duration of the pulse is 30 seconds. Because of the overbalanced drilling condition, the stabilized pressure is greater than the formation pressure. The stabilization time is reported when the relative pressure change of a shut-in test is within 0.35% of the final pressure. Tight Sand Formation. In this case, an OBM filtrate invades an oil reservoir in which the fluid properties of mud-filtrate and in-situ oil are the same. Table 9 describes the oil properties, and Table 10 presents the tight sand formation properties. Fig. 6 compares four different testing methods in this tight sand reservoir demonstrating that the automated method reaches a stable pressure much more quickly than the other testing methods. TABLE 9 PROPERTIES OF IN-SITU OIL AND OIL-BASED MUD FILTRATE FOR THE TIGHT SAND FORMATION. Properties FC 10 Molecular weight 100 Critical pressure 5.01 Critical temperature 6.1 Acentric factor 0.044 Critical volume 0.51 TABLE 10 SYNTHETIC TIGHT SAND RESERVOIR PROPERTIES. Formation Properties Permeability [μd] 0.01 Anisotropy (k v /k h ) 1.0 Porosity[%] 10 Fluid viscosity [cp] 1.0 Fluid compressibility [psi -1 ] 3.0x10-6 Formation pressure [psi] 0,000

1 SPE 155037.16 x 104.14.1.1.08.06.04.0 8 Drawdown, BD Stab. T > 1hrs, BD Stab. T > 1 hrs = 1/ Q -, BD Stab. T = 480.8 min Automated, BD Stab. T = 183.1 min 6 0 0 40 60 80 100 Fig. 6 Tight sand formation testing with a straddle packer-type formation tester in an overbalanced drilling situation. A comparison between the various tests shows that an automated pulse test technique is able to reach a stabilized pressure more quickly than other tests. Shale Gas Reservoir. In this case, an OBM invades a shale gas reservoir. The straddle packer was used to perform the synthetic testing in this reservoir (Table and 5). Table 11 describes the OBM and in-situ gas properties. The reservoir properties are summarized in Table 1. Fig. 7 shows the results of testing in this synthetic shale gas reservoir. TABLE 11 PROPERTIES OF IN-SITU GAS AND OBM INVADING THE SHALE GAS RESERVOIR. Properties CH 4 C H 46 Molecular weight 16.04 310.0 Critical pressure 45.35 13.6 Critical temperature 190.4 804.4 Acentric factor 0.008 0.87 Critical volume 0.098 1.16 TABLE 1 SYNTHETIC SHALE GAS RESERVOIR AND OBM PROPERTIES. Formation Properties Permeability [μd] 0.01 Anisotropy (k v /k h ) 1.0 Porosity [%] 10 In-situ gas viscosity [cp] 0.01 Oil-based mud filtrate viscosity [cp] 1.0 Fluid compressibility [psi -1 ] 3.0x10-6 Formation pressure [psi] 0,000.1 x 104.1.08.06.04.0 Drawdown, BD Stab. T = 465.4 min, BD Stab. T = 389.0 min = 1/ Q -, BD Stab. T = 34.9 min Automated, BD Stab. T = 169.1 min 8 0 0 40 60 80 100 Fig. 7 Shale gas reservoir testing in an overbalanced drilling situation using OBM. A comparison shows that the automated method is approximately three times faster than other methods to reach its stabilized pressure. The method loses its competency because of an increase in the mobility of this reservoir. Because the gas viscosity is 0.01 cp, the reservoir mobility is 100 times greater than in the previous case. This demonstrates once again how effective the automated pulse test technique is in low mobility formations. For formations with higher mobilities, the method becomes less effective but still can offer an improvement over conventional methods.

SPE 155037 13 Heavy Oil Formation. In the heavy oil synthetic reservoir, the WBM was used in the overbalanced drilling situation. The tests were performed with a straddle packer (see Table and Table 5); Table 13 summarizes the reservoir rock and fluid properties. Fig. 8 compares various test results for the heavy oil reservoir. Table 14 summarizes the overbalanced drilling situation in these three formations (tight sand, shale gas, and heavy oil). TABLE 13 SYNTHETIC HEAVY-OIL RESERVOIR ROCK AND FLUID PROPERTIES..1 x 104 Formation Properties.1 Permeability [md] 1.0 Anisotropy (k v /k h ) 1.0 Porosity [%] 10 In-situ oil viscosity [cp] 1000.0 Water-based mud filtrate viscosity [cp] 1.0 Fluid compressibility [psi -1 ] 3.0x10-6 Formation pressure [psi] 0,000.08.06.04 Drawdown, BD Stab. T = 380. min.0, BD Stab. T = 396.4 min = 1/ Q -, BD Stab. T = 81. min Automated, BD Stab. T = min 0 0 40 60 80 100 Fig. 8 Heavy oil reservoir testing in a WBM overbalanced drilling case. TABLE 14 COMPARISON OF STABILIZATION TIME TO DIFFERENT TEST PROCEDURES FOR THREE RESERVOIR TYPES. Test Type Drawdown Single Pulse Half-Injection Automated Pulse Reservoir Pulse Type 1. Tight sand > 1 hrs > 1 hrs 480.8 min 183.1 min. Shale gas 465.4 min 389.0 min 34.9 min 169.1 min 3. Heavy oil 380. min 396.4 min 81. min min In these cases, the stabilization time is faster using the automated pulse method. The permeability of the first and second reservoirs are the same, however, the mobility of the second reservoir is 100 times bigger than that of the first one due to the gas viscosity. Similarly heavy oil causes a similar contrast in mobility when water invades. The mobility is one of the factors that causes the stabilization time to be somewhat slower in the second and third reservoirs. Underbalanced Drilling. The underbalanced case is the same as the overbalanced drilling, except that the bottomhole pressure is 500 psi less than the formation pressure (0,000 psi). The reservoir and fluid properties are the same as the overbalanced case, and the stabilization time is measured when each test reaches within the 0.35% limit of stabilized pressure. In this section, the test begins with an injection, rather than a pumpout test, because the formation pressure near the wellbore is less than the far-field pressure; therefore, an injection is necessary to increase the pressure. However, the choice of injection or drawdown is arbitrary, attempting to reduce the iterations. In many cases, it may be difficult to determine if a formation is under- or overbalanced, but the automated pulse testing method will work even if the initial pulse is not in the optimal direction. Tight Sand Formation. Properties of the tight sand synthetic formation are described in the overbalanced section. Fig. 9 shows the comparison of the tests for a straddle-packer formation tester.

14 SPE 155037.14 x 104.1.1.08.06.04 Drawdown, BD Stab. T > 1 hrs, BD Stab. T = 457. min = 1/ Q -, BD Stab. T = 511. min Automated, BD Stab. T = 17.5 min.06 x 104.05.04.03.0 Drawdown, BD Stab. T = 456.6 min, BD Stab. T = 3 min = 1/ Q -, BD Stab. T = 358.3 min Automated, BD Stab. T = 13.7 min.0.01 8 6 0 0 40 60 80 100 Fig. 9 Tight-sand formation testing in an underbalanced drilling case. The drawdown test is in fact an injection test to increase the pressure because the wellbore pressure is less than the formation pressure. All tests begin with an injection test. 9 8 0 0 40 60 80 100 Fig. 30 Shale gas reservoir testing with a straddle packer automated pulse testing comparison..08 x 104.06.04 Drawdown, BD Stab. T = 398.3 min, BD Stab. T = 87.5 min = 1/ Q -, BD Stab. T = 343.0 min Automated, BD Stab. T = 18.6 min.0 8 6 4 0 0 40 60 80 100 Fig. 31 Heavy oil reservoir testing with a straddle packer automated pulse testing comparison. Shale Gas Reservoir. In shale gas underbalanced testing the bottomhole pressure is less than the formation pressure; therefore, the formation is bleeding into the wellbore. Fig. 30 shows that the automated pulse test is more effective in reaching the stabilized pressure. Heavy Oil Formation. A comparison of the heavy oil testing indicates that the automated pulse test method is at least twice as fast as other methods for this synthetic case (Fig. 31). As previously mentioned, it is apparent that the lower the mobility of the formation, the better is the performance of an automated pulse test relative to other methods, at least for the examples shown. TABLE 15 COMPARISON OF STABILIZATION TIME FOR THE UNDERBALANCED DRILLING CASE IN THREE DIFFERENT RESERVOIRS. Test Drawdown Single Pulse Half-Injection Automated Pulse Reservoir Pulse Type Tight sand 1 > 1 hrs 457. min 511. min 17.5 min Shale gas 456.6 min 3 min 358.3 min 13.7 min Heavy oil 3 398.3 min 87.5 min 343.0 min 18.6 min 1: Mobility: 10-5 [md/cp], : Mobility: 10-3 [md/cp], 3: Mobility: 10-3 [md/cp] The stabilization times for these cases are less than that of overbalanced drilling. One of the reasons is that the bottomhole pressure is 500-psi nearer to the formation pressure in the underbalanced drilling.

SPE 155037 15 Discussion about Interpretation Method The automated pulse test method is practical and can be performed in the field. The flowrate and pulse period for each test step can be calculated with an optimized algorithm. This optimization can be done with a simulation model and optimization algorithm if formation and flow-line parameters are known. When applying automated pulse testing in the field without exact knowledge of formation pressure and other rock/fluid properties, a test design procedure is recommended (flowrate and pulse period) based on the best prior knowledge of the reservoir. Then, the pressure feedback will adjust the flowrate based on reservoir response during the test. The same forward models are used for test design, and a test can also be used to match the pressure testing results from the field to determine the formation pressure, supercharge effects, mobility/permeability and skin. For gas and heavy oil simulation examples, the fluid compressibility was chosen as 3.0 x 10-6. This was assumed because the flow line normally has mud filtrate and this fluid compressibility forms the dominant flow regime during fluid injection or pullback from the reservoir during the test. Since the pulse duration is short, and the reservoir mobility is so low, the assumption of constant fluid compressibility is reasonable for simulations with water based muds or dead oil filtrates. In a continuation of this study, further pressure transient analysis and inversion techniques can be performed on the automated pulse test measurement to determine other reservoir properties such as mobility, damage skin, and wellbore storage. Conclusions The automated pulse test method measures the reservoir pressure more quickly than any other available method in the industry; it is designed for tight formations, shale gas, heavy oil and, in general, to operate in low mobilities, where the new method presents a better answer than other methods. The automated pulse test algorithm was designed with a pressure feedback circuit, and can be implemented in formation-testing tools to perform the job without operator interference. In the automated pulse test, there is no need to have prior information about the reservoir pressure. In presented examples, the automated pulse test was able to find the reservoir pressure at least 3 to 10 times faster than that of conventional pressure-testing techniques. The recorded builddown stabilization time was the time required to reach 0.35% of the initial reservoir pressure. In the case of overbalance and underbalance drilling, the stabilized pressure is not the initial reservoir pressure and further analysis should be done to find the true reservoir pressure. However, the advantage of the automated pulse test over other methods is to reach this stabilized pressure faster. The automated pulse test is a non-linear method, similar to that found in control systems with a feedback loop. This system shows some oscillation in the objective response to find the local minimum (in this case, reservoir pressure). To remove the effect of noise on the feedback pressure loop in the automated pulse test method, it is possible to use the change in the slope of the line to make the decision about the next move, rather than observing the two pressure end-points. It is also possible to use a noise reduction device, such as a low-pass filter in the design. Nomenclature c t : Compressibility, [psi -1 ] Μ : Mobility, [μd/cp] T : Time, [s] φ : Porosity, [%] V fl : Flowline volume, [cc] μ : Viscosity [cp] Q : Pulse rate, [cc/min] k : Absolute permeability, [μd] Acronyms BD : Build-down stabilization RDT : Reservoir description tool DD : Drawdown test DPS : Dual probe section BU : Build-up test OBM : Oil-based mud INJ : Injection test WBM : Water-based mud References Angeles, R., Torres-Verdin, C., Hadibeik, A., et al. 010. Estimation of Capillary Pressure and Relative Permeability from Formation- Tester Measurements Using Design of Experiment and Data-Weighing Inversion: Synthetic and Field Examples. JPSE, 75, (1-). Britt, L., Jones, J., Heidt, J., et al. 004. Application of After-Closure Analysis Techniques to Determine Permeability in Tight Formation Gas Reservoirs. Paper SPE 90865 presented at the SPE Annual Technical Conference and Exhibition, Houston, Texas, USA, 6 9 September. Chin, W. and Zhuang, X. 007. Formation Tester Inverse Permeability Interpretation for Liquids in Anisotropic Media with Flowline Storage and Skin at Arbitrary Dip. Paper SPWLA presented at the 48th Annual Logging Symposium, June 3 6, 007, Austin, Texas. Elshahawi, H., Fathy, K., and Hiekal, S. 1999. Capillary Pressure and Rock Wettability Effects on Wireline Formation Tester Measurements. Paper SPE 5671 presented at the SPE Annual Technical Conference and Exhibition, Houston, Texas, USA, 3 6 October. Goode, P.A. and Thambynayagam, R.K.M. 1987. Pressure Drawdown and Buildup Analysis of Horizontal Wells in Anisotropic Media. SPE Formation Evaluation,, 4, December.

16 SPE 155037 Hadibeik, A., Proett, M., Torres-Verdín, C. et al. 009. Wireline and While-Drilling Formation-Tester Sampling with Oval, Focused, and Conventional Probe Types in the Presence of Water- and Oil-Base Mud-Filtrate Invasion in Deviated Wells. Paper SPWLA presented at the SPWLA 50th Annual Logging Symposium, The Woodlands, Texas, 1 4 June. Hadibeik, H., Proett, M., Torres-Verdin, C., et al. 010. Effects of Highly Laminated Reservoirs on the Performance of Wireline and While-Drilling Formation-Tester Sampling with Oval, Focused, and Conventional Probe Types. Paper SPWLA presented at the SPWLA 51 st Annual Logging Symposium, Perth, Australia, 19 3 June. Hadibeik, H., Pour, R., Torres-Verdín, C., et al. 011. 3D Multiphase Streamline-Based Method for Interpretation of Formation-Tester Measurements Acquired in Vertical and Deviated Wells. Paper SPE 146450 presented at the SPE Annual Technical Conference and Exhibition, Denver, Colorado, USA, 30 October November. Proett, M., Chin, W., Lysen, S., et al. 004. Formation Testing in the Dynamic Drilling Environment. Paper SPWLA presented at the SPWLA 45th Annual Logging Symposium, Noordwijk, The Netherlands, June 6 9. Proett, M., Waid, M., Heinze, J., et al. 1994. Low Permeability Interpretation Using a New Wireline Formation Tester "Tight Zone" Pressure Transient Analysis. Paper SPWLA presented at the 35th Annual Logging Symposium, Tulsa, Oklahoma, June 19. Proett, M. and Chin, W. 1996. Supercharge Pressure Compensation Using a New Wireline Testing Method and Newly Developed Early Time Spherical Flow Model. Paper SPE 3654 presented at the SPE Annual Technical Conference and Exhibition, Denver, Colorado, USA, 6 9 October. Proett, M. 1996. Real Time Pressure Transient Analysis Methods Applied to Wireline Formation Test Data. Paper SPE 8449 presented at the SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, 5 8 September,. Proett, M. and Chin, W. 1998. New Exact Spherical Flow Solution with Storage and Skin for Early-Time Interpretation with Applications to Wireline Formation and Early-Evaluation Drillstem Testing. Paper SPE 49140 presented at the SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, 7 30 September. Proett, M., Chin, W., and Mandal, B. 000. Advanced Dual Probe Formation Tester with Transient, Harmonic, and Pulsed Time-Delay Testing Methods Determines Permeability, Skin, and Anisotropy. Paper SPE 64650 presented at the International Oil and Gas Conference and Exhibition, Beijing, China, 7 10 November. Pour, R. 011. Development and Application of a 3D Equation-Of-State Compositional Fluid-Flow Simulator in Cylindrical Coordinates for Near-Wellbore Phenomena. Dissertation for the Degree of Doctor of Philosophy at Graduate School of the University of Texas at Austin, December, Austin, TX. Schrooten, R.A., Boratko, E.C., Singh, H., et al. 007. Measurements to Improve Reservoir Characterization in Tight Formation Gas- Wamsutter Field, Wyoming. Paper SPE 109565 presented at the SPE Annual Technical Conference and Exhibition, Anaheim, California, USA, 11 14 November. Yang, Y., O'Connell, W., and Birmingham, T. 007. A Novel Method to Calculate the Reservoir Pressure of Tight Sand Using the Hydraulic Fracturing Information in the Denver-Julesburg Basin. Paper SPE 111089 presented at the Eastern Regional Meeting, Lexington, Kentucky, USA, 17 19 October. Yildiz, T., Desbrandes, R., and Bassiouni, Z.A. 1991. Flowline Storage Effect on Wireline Formation Tester. Paper SPE 753 presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA, 6 9 October. Zazovsky, A., Haddad, S., and Tertychnyi, V. 005. A Method for Formation Temperature Estimation Using Wireline Formation Tester Measurements, paper SPE 96 presented at the SPE Western Regional Meeting, Irvine, California, USA, March 30 April 1.